Artificial intelligence in cancer pathology: Challenge to meet increasing demands of precision medicine

Lai,Boan; Fu,Jianjiang; Zhang,Qingxin; Deng,Nan; Jiang,Qingping; Peng,Juan

doi:10.3892/ijo.2023.5555

Artificial intelligence in cancer pathology: Challenge to meet increasing demands of precision medicine

Authors:
- Boan Lai
- Jianjiang Fu
- Qingxin Zhang
- Nan Deng
- Qingping Jiang
- Juan Peng
View Affiliations

Affiliations: Department of Pathology, Key Laboratory of Reproduction and Genetics of Guangdong Higher Education Institutes, Key Laboratory for Major Obstetric Diseases of Guangdong Province, The Third Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong 510150, P.R. China, Department of Urology, The Third Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong 510150, P.R. China
Published online on: August 1, 2023 https://doi.org/10.3892/ijo.2023.5555
Article Number: 107

Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )

Metrics: Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )

Cited By (CrossRef): 0 citations Loading Articles...

Abstract

Clinical efforts on precision medicine are driving the need for accurate diagnostic, new prognostic and novel drug predictive assays to inform patient selection and stratification for disease treatment. Accumulating evidence suggests that a combination of cancer pathology and artificial intelligence (AI) can meet this requirement. In the present review, the past, present and emerging integrations of AI into cancer pathology were comprehensively reviewed, which were divided into four main groups to highlight the roles of AI‑integrated cancer pathology in precision medicine. Furthermore, the unsolved problems and future challenges in AI‑integrated cancer pathology were also discussed. It was found that, although AI‑integrated cancer pathology could enable the amalgamation of complex morphological phenotypes with the multi‑omics datasets that drove precision medicine, synergies of cancer pathology with other medical tools could be more promising for the clinic when making an accurate and rapid decision in personalized treatments for patients. It was hypothesized by the authors that exploring the potential advantages of the multimodal integration of cancer pathology, imaging‑omics, protein‑omics and other‑omics, as well as clinical data to decide upon appropriate management and improve patient outcomes may be the most challenging issue of cancer precision medicine in the future.

Introduction

It was estimated that there were >19 million new cancer cases and 9 million cancer-related deaths in 2020 worldwide. The top five cancer sites, based on the global cancer incidence, are as follows: Female breast cancer (BC), lung cancer (LC), colorectal cancer (CRC), prostate cancer (PCa) and stomach cancer, and more than half of cancer deaths globally are attributed to these cancer types (World Health Organization. Cancer Today. 2020; http://gco.iarc.fr/). In China, a growing cancer burden was unexpectedly observed, and nearly 3 million cancer deaths and >4 million individuals were diagnosed with cancer in 2020. Accurate diagnosis and personalized cancer therapy are the keys to improving the cancer prognosis for every patient. With the promise of an estimated numerical prognosis for patients with cancer, AI has been proposed as a prominent way to improve pathological diagnostic accuracy (DA) and cancer prognostication (CP). Over the past decades, although the exploration and exploitation of AI in cancer pathology has evolved substantially, their generations, interpretations and impact on patients have remained incompletely understood by oncologists. In the present study, the published literature on Google Scholar and PubMed were searched using the following terms: ‘AI’ AND ‘pathology’ AND ‘cancer’, or ‘machine learning (ML)’ AND ‘pathology’ AND ‘cancer’, or ‘deep learning’ AND ‘pathology’ AND ‘cancer’. The present review includes an overview of the roles of AI-integrated cancer pathology in precision medicine by summarizing the rationale for its use, clarifying recent innovations and grouping the main applications, as well as discussing the challenges for AI-integrated cancer pathology-based individualized treatment.

Artificial intelligence (AI)

AI is defined as the ability of a machine to simulate, extend and expand human cognition, and has the most features suggestive of human intelligence, to decide upon an action to achieve the desired goal. Initially proposed in 1950 by Turing with ‘Can Machines Think’, and subsequently coined in 1956 by McCarthy (Fig. 1), AI was originally described as being able to ‘simulate intelligence’. With the gradual advent of ML based on computational algorithms and the following computational intelligence, the more power of the decision-making pattern AI was endued with, and the more diverse, complex or esoteric fields AI was involved in, a marked impact was made on modern civilization, namely on Industrial Revolution 4.0 (1). Deep learning (DL) introduced in 1986 by Dechter is by far the most common subtype of ML, and represents computational models originating from artificial neural networks (ANNs) to derive progressively higher-order features from data in the form of multiple non-linear layers (Fig. 2) (1). Inspired by neurobiology, DL is composed of units that compute a weighted sum of the multiple inputs (referred to as the pre-activation) and transforms the results non-linearly, which includes automatic learning and hierarchical representation on multiple levels (2). DL has been employed in nearly all scientific areas, especially in medical imaging, including cancer pathology (the cornerstone of cancer medicine), which has yielded important results that paved the way for the development of the convolutional neural network (CNN), generative adversarial network (GAN), auto encode and so on, to meet the demands of specific biomedical applications for precision oncology.

Figure 1.

(A) Timeline of the terms used and hallmarks in the field of AI, which influenced the development of cancer pathology. (B) The number of papers retrieved according the four categories of principal tasks of integration of AI and cancer pathology in precision medicine, including: DA, CP, DB and DPr. AI, artificial intelligence; ANNs, artificial neural networks; CNN, convolutional neural network; CP, cancer prognostication; DA, diagnostic accuracy; DB, drug benefit; DBN, deep belief network; DGMate, DiGital tuMor pArameTErs; DLBCL, diffuse large B-cell lymphoma; DP, digital pathology; DPr, disease prevention; GANs, generative adversarial networks; LLSM, lattice light sheet microscopy; LSTM, long short-term memory; ML, machine learning; NSCLC, non-small cell lung cancer; rDL, rationalized DL; SL, swarm learning; SIM, structured illumination microscopy; TILs, tumor-infiltrating lymphocytes; UL, unsupervised learning; WSIs, whole slide images.

Figure 2.

General framework of AI integrating into cancer pathology. AI encompasses a wide variety of strategies (such as, expert system, robotics, fuzzy logic and nature language processing), in which ML is used to make a prediction emulating what an intelligent human may do in the same situation. DL is a subset of a larger group of ML methods developed from artificial neural networks, which consist of algorithms with hierarchical processing layers that perform non-linear transformations to represent and learn data characteristics effectively. Integrating DL and other ML approaches into cancer pathology, which is mainly composed of conventional pathology and MP, can reinforce the ability of oncologists to understand and address the role of cancer pathology in precision medicine pertaining to diagnostic accuracy, cancer prognostication, drug benefit and disease prevention. AI, artificial intelligence; DL, deep learning; ML, machine learning; WSI, whole slide image.

Digital pathology (DP)

Defined as an image-based approach, DP aims to acquire, manage, interpret and distribute pathology information in a computer-empowered manner, to extract and analyze pathological visual data. Since the introduction of telepathology in 1986 by Weinstein, the emergence of whole slide images (WSIs) in the 1990s, and other digital images (Electronic microscope digital images), has boosted the prosperity of the DP era (Fig. 1A) (3). Efforts have been devoted to improving the scanning speed and the accuracy of images, which are the two major issues affecting the throughout performance of WSIs. Notably, GAN, an unsupervised technique proposed by Goodfellow et al (4) in 2014, was composed of two competing models: i) A generative model G capturing the data distribution; and ii) a discriminative model D, which made likelihood estimations on the training data rather than G. With respect to this, Fanous and Popescu (5) reported a new method in 2022 (referred to as GANscan) to significantly increase the scanning speed of the whole slide scanner and to correct any defocusing by integrating the Pix2Pix image translation framework and GAN-based model.

Notably, WSIs provide an enabling AI platform for generating a rich variety of novel ecosystems in DP, which is known as Pathomics. Pathomics is an AI-based multi-systems integration methodology to understand cellular interactions and signaling by analyzing relevant data from tissues and cells. With the emergence of Pathomics, integration of AI into the DP workflow has achieved some breakthroughs in cancer pathology. For example, in MICCAI 2014 (Table I), the brain tumor digital pathology challenge, including two sub-challenges (classification and automated segmentation), was posted. In this regard, Barker et al (6) demonstrated a novel method using local representative tiles, decided diagnostically by an elastic net classifier, for automated classification in brain cancer cases, which exhibited high accuracy for diagnosis as well as structural stability and a robustness for varying parameters, implying that it may be useful for automatic differentiation of the two subtypes of brain cancer (glioblastoma multiforme and lower grade glioma) (Table II). In another example, multiple-instance learning (MIL), first proposed by Dietterich et al in 1997, is a variation on supervised learning. In 2019, Campanella et al (7) presented the MIL-RNN models in the analysis of three datasets (including PCa, basal cell carcinoma of the skin and BC metastases to axillary lymph nodes datasets) of 44,732 WSIs from 15,187 patients, in which MIL-based residual neural network (ResNet)34 models were used to classify tiles. Then, semantically rich tile-level feature representations were generated, which integrated the information across the WSIs through the RNN models and achieved the final classification result. This system was shown to be able to train accurate classification models at an unprecedented scale, providing the cornerstone for the evolution of computational decision support systems in clinical practice (Fig. 1).

Table I.

Overview of the principal challenges related to digital pathology (as of May 2023).

Table II.

Overview of the principal papers AI-integrated cancer pathology for four main tasks, including DA, CP, DB and DPr (as of May 2023).

Table II.

Overview of the principal papers AI-integrated cancer pathology for four main tasks, including DA, CP, DB and DPr (as of May 2023).

First author/s, year	Characteristic	Staining	Tissue/cell	Task	(Refs.)
Shipp et al, 2002	The supervised learning classification techniques based on genomic profiling could effectively predict outcome for patients with DLBCL and identify rational targets for intervention.	H&E, IHC, genomic microarrays	DLBCL	CP & DB	(55)
Khan et al, 2008	wFDT demonstrated comparative performance to predict the prognosis of BC from the SEER BC dataset.	H&E	BC	CP	(56)
Fatakdawala et al, 2010	EMaGACO was applied to evaluate lymphocytic infiltration on HER2⁺ BC WSIs with comparative performance.	H&E	BC	DA	(57)
Sertel et al, 2010	A computer-aided detection system for automated identification of CB cells from H&E stained FL tissue samples was presented with a promising 80.7% detection accuracy.	H&E	FL	DA	(58)
Dundar et al, 2011	A prototype system for automatically classifying breast microscopic tissues to distinguish between UDH and actionable subtypes (ADH and DCIS) was observed with competitive performance.	H&E	IBL	DA	(59)
Tuominen et al, 2012	The ImmunoMembrane method based on AI technology was used to improve the diagnostic accuracy of HER2 IHC from BC.	IHC	BC	DA	(60)
Gertych et al, 2012	CF-SVM classifiers were delineated to improve the risk stratification of CC by evaluating dual p16/Ki67 nuclear ICC based on PAP.	PAP, ICC	CC	CP	(61)
Doyle et al, 2012	A cascaded (CAS) approach was proposed to classify prostate biopsy tissue samples based on H&E stained WSIs with comparative performance.	H&E	PCa	DA	(62)
Wang et al, 2014	A light CNN model based on a combination of a CNN model and handcrafted features (morphology, color and texture features) was proposed to detect mitosis in H&E stained WSIs to predict disease aggressiveness and patient with BC outcome.	H&E	BC	CP	(63)
Lewis et al, 2014	The classifier (QuHbIC) was validated to predict outcomes of patients using the information from a H&E stained microarray cohort of p16⁺ OSCC cases.	H&E, TMA	OSCC	CP	(64)
Sirinukunwattana et al, 2015	Outperformance was observed in an experiment using a spatially constrained CNN to detect the malignant epithelial nuclei in H&E stained WSIs of BC and CRC cases.	H&E	BC & CRC	DA	(65)
Yuan et al, 2015	An unbiased methodology was proposed to model the spatial distribution of TILs in TNBC to predict prognosis and treatment options.	H&E	TNBC	CP & DB	(66)
Veta et al, 2015	The results from the Assessment of Mitosis Detection Algorithms 2013 (AMIDA13) challenge were presented. The challenge was based on a dataset consisting of 12 training and 11 testing subjects, with >1,000 annotated mitotic figures by multiple observers.	H&E	BC	DA	(52)
Paul, 2015	A fast and accurate approach for automatic mitosis detection from BC histopathological images was proposed with a favorable result.	H&E	BC	DA	(51)
Xie et al, 2015	Used a CNN-based voting method to localize nucleus centroids with heavy cluttering and morphologic variations in microscopy images.	H&E, IHC	Neuroen-docrine tumor	DA	(67)
Sirinukunwattana et al, 2016	Joint detection and classification of the proposed SC-CNN and NEP coupled with CNN could detect and classify cell nuclei in histopathology images of cancerous tissue more accurately.	H&E	CRC	DA	(68)
Romo-Bucheli et al, 2016	Developed a DL classifier to automatically identify tubule nuclei and predict Oncotype DX risk from WSIs of ER⁺ BC.	H&E	BC	DA & DB	(69)
Xu et al, 2016	A DL strategy, SSAE, was presented to efficiently detect nuclei on high-resolution histopathological images of BC.	H&E	BC	DA	(70)
Turkki et al, 2016	A CNN-based approach was applied to quantify immune cell infiltration in BC samples using only basic morphology staining.	H&E, IHC	BC	DA	(71)
Ali et al, 2016	An image-processing pipeline based on SVM was used to accurately diagnose and classify lymphocyte cells from H&E stained WSIs to predict the response to neoadjuvant chemotherapy and the outcome of the patient.	H&E	BC	DA, CP & DB	(72)
Bartsch et al, 2016	A genetic programming algorithm was applied to evolve classifier mathematical models for outcome prediction, which predicted recurrence in non-muscle invasive UC of the bladder.	Gene profiling	BLCA	DB & CP	(73)
Yu et al, 2016	Regularized machine learning methods were used to analyze information from H&E stained and TMA-based WSIs of LC to predict prognosis of patients.	H&E, TMA	LC	CP	(74)
Barker et al, 2016	A novel method, using local representative tiles determined diagnostically by an Elastic Net classifier for automated classification in brain cancer cases, demonstrated high accuracy for diagnosis as well as being structurally stable and robust for varying parameters, implying that it could be used to automatically differentiate between GBM and LGG.	H&E	GBM, LGG	DA	(6)
Ali et al, 2017	Lymphocyte density in pre-treatment biopsies was validated as an independent predictor of pCR in BC by analyzing the H&E stained WSIs using an SVM-based approach.	H&E	BC	DA, CP & DB	(75)
Ehteshami Bejnordi et al, 2017	7 automatic DL algorithms showed outperformance in the CAMELYON16 researcher challenge competition, which could be used to improve diagnostic accuracy and efficiency in detecting metastasis in female BC lymph node tissue sections.	H&E	BC	DA	(53)
Lu et al, 2017	The OHbIC was constructed to predict outcomes of patients with OSCC from H&E-based WSIs.	H&E	OSCC	CP	(76)
Gecer et al, 2018	A system based on the combination of four FCNs and one CNN was presented to detect and classify BCs in WSIs with outperformance.	H&E	BC	DA	(77)
Haider et al, 2018	SIMMS algorithm was observed to be useful for predicting phenotypes in five cancer types.	H&E	Five cancer types, including BC	CP	(45)
Yoshida et al, 2018	The e-Pathologist image analysis software was proposed to automatically classify GC on H&E stained WSIs with comparative performance to pathologists.	H&E	GC	DA	(78)
Bo et al, 2018	A deep residual network with 50 layers was used auto-classify GC, resulting in improving the generalization performance.	H&E	GC	DA	(79)
Ichimasa et al, 2018	An artificial intelligence model was generated to help in deciding the additional surgery after endoscopic resection of T1 CRC with the presence of lymph node metastasis.	H&E	CRC	CP & DPr	(80)
Mezheyeuski et al, 2018	A fluorescence-based multiplexed immunohistochemical method in combination with a multispectral imaging system was used to quantify immune infiltrates in situ in an NSCLC TMA.	H&E, IHC	LC	DA	(81)
Yi et al, 2018	A microvessel prediction method using fully CNNs was proposed and evaluated on H&E stained WSIs from LUAD.	H&E, IHC	LUAD	DA	(11)
Mobadersany et al, 2018	A computational approach combining the power of adaptive ML algorithms with traditional survival models was proposed to predict glioma outcomes by integrating information from WSIs and genomic profiling, which surpassed current-level performance.	H&E, genomic profiling	Gliomas	CP	(82)
Saltz et al, 2018	A comprehensive methodology (termed Computational Staining) based on CNNs was framed to map TILs and analyze the molecular and clinical correlation with TILs from H&E-stained WSIs containing 13 TCGA cancer types. TIL map structural patterns were generated with a competitive performance, which revealed that TILs patterns were associated with tumor and immune molecular features, cancer type and survival, which laid a new milestone for TME research.	H&E	13 TCGA cancer types	DA, DB & CP	(21)
Ching et al, 2018	A new neural network framework called Cox-nnet was applied to predict patient outcomes accurately and efficiently from 10 high-throughput transcriptome TGCA datasets.	Gene profiling	10 TCGA cancer datasets	CP	(41)
Katzman et al, 2018	DeepSurv, a Cox proportional hazards deep neural network and state-of-the-art survival method, was used to explore, understand and predict the effects of patient characteristics on the risk of failure.	H&E	Cancer	CP	(42)
Chaudhary et al, 2018	A DL-based model of HCC was proposed to predict HCC survival from 6 patient cohorts by integrating multiomics data (RNAseq, miRNA sequencing, and methylation data from the TCGA) with outperformance.	Gene profiling	HCC	CP	(83)
Günakan et al, 2019	A Naive Bayesian ML algorithm was used to construct a model for predicting lymph node metastasis in patients with endometrial cancer.	H&E	UCEC	CP	(84)
Tabl et al, 2019	A hierarchical ML system could predict the 5-year survival of patients who underwent specific therapy and identify gene biomarkers to guide the treatment of BC.	Gene profiling	BC	DA, CP & DB	(46)
Turkki et al, 2019	An ML method was used to predict the outcome of patients with BC by analyzing the information from TMAs.	TMA	BC	CP	(85)
Fang et al, 2019	A CNN with deep generalized multi-instance learning was used to classify differentiation degree (poorly and well/moderately) and lauren type (intestinal, diffuse and mixed) in GC with a favorable result.	H&E	GC	DA	(86)
Mori and Miwa, 2019	VGG16 based on CNN was observed to predict the invasive progression in gastric signet-ring cell carcinoma component to distinguish between intra-mucosal or advanced disease.	H&E, IHC	GC	DA	(87)
Leon et al, 2019	Two approaches based on CNN were proposed to classify benign or malignant cancer from stomach biopsy specimens.	H&E	GC	DA	(88)
Wang et al, 2019	An automated cell type classification pipeline, ConvPath, was proposed to identify TME-related features.	H&E	LUAD	DA	(89)
Aprupe et al, 2019	A CNN-based immune cell detection and quantification method was applied on IHC WSIs from LC, which demonstrated a similar performance to humans.	IHC	LC	DB	(90)
Dihge et al, 2019	Seven machine-learning models were proposed to predict lymph node meta-stasis of BC from the combined information of clinicopathological characteristics, gene expression data and mixed features, as well as SLNB, which demonstrated comparable accuracy and identified more node-negative patients, resulting in reducing the rates of SLNB for patients at low risk of nodal involvement.	H&E	BC	CP & DPr	(91)
Nir et al, 2019	Classifier performance was evaluated by its accuracy, sensitivity and specificity in the detection of PCa in H&E stained TMAs.	H&E, TMA	PCa	DA & CP	(92)
Campanella et al, 2019	A multiple instance learning-based DL system (MIL-RNN models) was presented, with accurate classification of cancer types from 3 datasets of 44,732 H&E stained WSIs from 15,187 patients (PCa, BCC and BC metastasizing to axillary lymph nodes).	H&E,	PCa, BCC and BC	DA	(7)
Courtiol et al, 2019	A new approach based on DCNN, termed MesoNet, was developed to accurately predict the overall survival of patients with MM from WSIs.	H&E	MM	DA	(93)
Achi et al, 2019	An attempted build of a lymphoma diagnostic model for four diagnostic categories using DL with a CNN algorithm: Benign lymph node, DLBCL, BL and SLL.	H&E	Lymphoma	DA	(94)
Gao et al, 2019	A novel DL-based framework deep cancer subtype classification (DeepCC) was used to classify cancer molecular subtypes using TCGA RNA-Seq datasets.	Gene profiling	CRC and BC	DA & DB	(36)
Rodner et al, 2019	An FCN-based automated image analysis algorithm was proposed to discriminate between HNCA and non-cancerous epithelium with a significant performance.	H&E, IHC	HNCA	DA	(95)
Hu et al, 2019	A DL-based visual evaluation algorithm was developed to automatically recognize cervical precancer or cancer with a favorable result.	H&E	CC	DA & DPr	(96)
Tian et al, 2019	An ML algorithm (Random Forest) was applied to build a risk stratification model for cervical precursor lesions based on CIN2⁺ enriched biomarkers from the capture-based NGS of 34 paired samples (including exfoliated cervical samples, cervical tissues and others).	Gene profiling	CC	CP	(97)
Newman et al, 2019	CIBERSORTx, an ML method, was introduced to infer cell-type-specific gene expression profiles from bulk tissue transcriptomes, which revealed cell-type-specific phenotypic states linked to distinct driver mutations and response to immune checkpoint blockade.	Gene profiling	Multiple cancer types, including malignant melanoma	DB	(98)
Oskal et al, 2019	A U-net based approach to epidermal tissue segmentation in H&E stained WSIs from malignant melanoma was proposed with a superior performance compared with existing techniques.	H&E	Malignant melanoma	DA	(15)
van IJzendoorn et al, 2019	Various ML methods were used to identify novel diagnostic and prognostic markers and therapeutic targets for STS from multiple STS gene expression datasets.	Gene profiling	STS	DA, CP& DB	(44)
Bulik-Sullivan et al, 2019	A DL-based computational model of antigen presentation named EDGE was proposed to predict neoantigens in a large HLA peptide and genomic dataset from various human tumors, resulting in increasing the positive predictive value of HLA antigen prediction by up to ninefold, which suggested that EDGE could enable an improved ability to develop neoantigen-targeted immunotherapies for patients with cancer.	Gene profiling	Various cancer types	DB	(47)
Huang et al, 2020	A set of DL-based models, including Cox-nnet, DeepSurv and a method proposed by Huang et al named AECOX (AutoEncoder with Cox regression network) were applied to capture the information from RNA-seq data to predict the survival prognosis of patients across 12 cancer types with superior performances.	Gene profiling	12 cancer types	CP	(43)
Choi et al, 2020	A robust GSC based on whole transcriptome RNAseq was designed to validate gene expressions values predicting clinicopathological features using several LC cohorts.	Gene profiling	LC	CP & DB	(20)
Saillard et al, 2020	Two DL algorithm models based on H&E stained WSIs outperformed a prediction of the survival of patients with HCC treated by surgical resection.	H&E	HCC	CP	(99)
Dietz et al, 2020	Integration of the ImageJ system in the KNIME analytics platform was used in single-cell analysis of AR⁺ LNCaP cells, which could be used to test future therapeutic agents targeting the AR pathway and AR activation in PCa.	H&E, Gene profiling	PCa	DA & DB	(100)
Bulten et al, 2020	A newly developed DL system was introduced to delineate individual glands, assign Gleason growth patterns and determine the biopsy-level grade in H&E stained WSIs from prostate biopsies, with a similar performance to pathologists.	H&E	PCa	DA	(101)
Han et al, 2020	A total of seven ML techniques were applied to detect and grade PCa based on H&E stained WSIs.	H&E	PCa	DB	(102)
Cascianelli et al, 2020	A strategy, named AWCA, was proposed to be integrated with PAM50, which could be applied as a single-sample method to improve classification robustness with high concordance and prognostic ability in BC. A total of five supervised learners were explored to build robust, single-sample intrinsic subtype callers via RNA-seq, which regularized mLR and displayed the best performance with classifications, robustness and prognostic ability, resulting in strengthening BC subtyping.	Gene profiling	BC	DA	(33)
Yu et al, 2020	A GO enrichment based on gene co-expression networks (GOEGCN) was proposed to study how the modulated co-expressed gene couples impacted biological functions at a GO level among all the BC subtypes, to some extent.	Gene profiling	BC	DA	(34)
Valle et al, 2020	A topic modeling was proposed to analyze TCGA transcriptomic data from BC and LC to reconstruct cancer sub-types and predict prognosis.	Gene profiling	BC&LC	DA & CP	(35)
Jaber et al, 2020	A DL approach for approximating PAM-50 intrinsic subtyping was developed using only H&E stained WSIs of breast biopsy tissue sections.	H&E	BC	DA	(103)
Valieris et al, 2020	An efficient ML algorithm was used to predict DRD from H&E stained WSIs.	H&E	BC and GC	DB & DPr	(104)
Woerl et al, 2020	DL was applied to predict the molecular subtype of MIBC samples from H&E stained WSIs.	H&E	MIBC	DA	(17)
Chen et al, 2020	scRNA-seq was used to understand the TME of NPC for developing targeted therapy and predicting patient prognosis.	Gene profiling	NPC	CP & DB	(37)
Zhou et al, 2020	scRNA-seq was used to detect intra-tumor heterogeneity and TIME in advanced OS.	Gene profiling	OS	DB	(38)
Kong et al, 2020	An ML framework was observed to predict drug response in patients with CRC or BLCA by using organoid modelderived drug genome data.	Gene profiling	CRC and BLCA	DB & CP	(40)
Skrede et al, 2020	A total of 10 CNNs, proposed for classifying supersized heterogeneous images, were used to predict patient outcomes by analyzing a prognostic biomarker in CRC H&E stained WSIs.	H&E	CRC	CP	(105)
Reichling et al, 2020	Software was developed to detect colon tumor, healthy mucosa, stroma and immune cells on CD3 and CD8 stained slides to automatically quantify the lymphocyte density and surface area of the tumor core and invasive margin. Digital parameters within the tumor cells related to patient outcomes were analyzed using a LASSO algorithm, DGMate (DiGital tuMor pArame-TErs). The results suggested that AI could potentially improve patient care by assisting pathologists in better defining the prognosis of patients with stage III colon cancer.	H&E, IHC	CRC	CP	(106)
Bao et al, 2020	A validated AI-assisted cytology system was used to conduct a perspective cohort study within a population-based CC screening program for 0.7 million women, with the results supporting the AI-based cytology system for primary CC screening in large-scale populations.	PAP	CC	DA & DPr	(107)
Xu-Monette et al, 2020	An NGS-COO model was built to classify DLBCL from genetic and transcriptional data obtained by RNA-seq with a similar performance to COO classification by either Affymetrix GeneChip microarray or the NanoString Lymph2Cx assay. Moreover, the NGS survival models were shown to stratify 30% high-risk patients with poor prognosis.	Gene profiling	DLBCL	DA & CP	(108)
Foersch et al, 2021	DL was used to accurately diagnose frequent subtypes of STS using H&E stained WSIs.	H&E	STS	DA	(109)
Lagree et al, 2021	A CAD pipeline using AI was introduced to assess the biomarkers for grading tumors in BC.	H&E, IHC	BC	DA	(110)
Bychkov et al, 2021	ML algorithms were used not only to predict the amplification of ERBB2 based on tumor morphological features from TMA, but also to predict the survival and efficacy of anti-ERBB2 treatment.	H&E, CISH, v	BC	DB & CP	(111)
Li et al, 2021	A novel DL-based biomarker that pre-dicted pCR from images of H&E stained WSIs was proposed and its predictive performance in BC was evaluated, resulting in a superior performance compared with other conventional biomarkers including sTILs and tumor subtype.	H&E	BC	DB	(112)
Chakraborty et al, 2021	XAI models were used to retrieve immune cell composition estimates from bulk RNA-seq data from TCGA breast invasive carcinoma data from the cbioPortal compared with TIMER2.0 based on EPIC, CIBERSORT, TIMER and xCell computational methods. Novel insights derived from the XAI model showed that B cells, CD8⁺ T cells, M0 macrophage, and NK T cells are the most critical TME features for enhanced prognosis of patients with BC.	Gene profiling	BC	CP	(113)
Fitzgerald et al, 2021	DL platforms for capturing information from the image characteristics of H&E images could offer new possibilities for better modelling of disease appearance and possibly improve the prediction of disease stage and patient outcomes.	H&E	BC and PCa	DA, CP & DB	(114)
Ayyad et al, 2021	CAD systems were applied to aid in the diagnosis of PCa by capturing information from WSIs.	H&E	PCa	DA	(115)
Haggenmüller et al, 2021	A total of 19 readers, including 11CNN-based approaches addressing the classification of dermoscopic images, six concentrating on the classification of clinical images, and two utilized for WSIs, were used to classify tumors and assess their potential clinical relevance in melanoma, with a superior or at least equivalent performance compared with human experts.	H&E	Melanoma	DA & CP	(116)
Kiehl et al, 2021	A CNN-based image analysis could help predict LNM of patients with CRC using H&E stained WSIs.	H&E	CRC	DA & CP	(117)
Yamashita et al, 2021	A DL model (MSINet) was developed to predict MSI on H&E stained WSIs of CRC, which showed exceeded performance.	H&E	CRC	DA & DB	(118)
Krause et al, 2021	A ‘histology CGAN’ was proposed to detect genetic alterations (MSI) from routine histology images in CRC.	H&E	CRC	DA & DB	(119)
Cancian et al, 2021	A DL pipeline based on the DeepLab-v3 architecture and semantic segmentation technique warranted the separation of TAMs from the background and the identification of single TAMs on H&E stained WSIs from CRC liver metastasis	H&E	CRC liver metastasis	CP & DB	(24)
Bian et al, 2021	A DL-based computational framework (ImmunoAIzer) was proposed to characterize cell distribution and gene mutations in the TME of CRC.	H&E, IHC	CRC	CP & DB	(25)
Saito et al, 2021	An ML-based SVM was proposed to predict the early recurrence of HCC after resection from H&E stained WSIs.	H&E	HCC	CP	(120)
Yamashita et al, 2021	A DL-based system (HCC-SurvNet) provided recurrence risk scores, which might augment current patient stratification methods and help refine the clinical management of patients undergoing primary surgical resection for HCC.	H&E	HCC	DA/DB	(121)
Yang et al, 2021	The first DL-based six-type classifier for histopathological WSI classification of LUAD, LUSC, SCLC, pulmonary tuberculosis, organizing pneumonia and normal lung was developed with a similar performance to experienced pathologists, and superior performance to other existing computational methods.	H&E	LC	DA	(122)
Chen et al, 2021	A method based on training neural networks on entire WSIs using only slide-level diagnostics was applied to classify LC subtypes.	H&E	LC	DA	(123)
Park et al, 2021	An algorithm based on CNN was observed to classify WSI-based gastric biopsies into three diagnostic categories (NFD, TA or carcinoma) with an outperformance.		GC	DA	(124)
Wang et al, 2021	Total WSIs of lymph nodes were analyzed using a DL framework to identify lymph nodes and tumor regions, and then assess the ratio of tumor area to LNM area with a similar performance to experienced pathologists.	H&E	GC	CP	(125)
Muti et al, 2021	A DL-based classifier was developed and validated to detect MSI and EBV status from H&E stained WSIs of GC.	H&E	GC	DB	(126)
Yan et al, 2021	A DTI-derived DLS was proposed to improve prediction accuracy of overall survival in patients with invasive glioma and classification accuracy of risk groups with distinct molecular pathway activities.	Gene profiling	Glioma	CP & DB	(127)
Bao et al, 2022	A robust ML model, an integrated stacking ML approach, was used for the early detection of the most prevalent and lethal cancer types using multiple plasma cfDNA fragmentomic features.	Gene profiling	HCC, CRC and LUAD	DPr	(128)
Duchmann et al, 2022	A mechanistic learning framework was used to predict the response to chemotherapy using dysplastic features and the oncogenetics of patients with AML treated in ALFA clinical trials.	Bone marrow smears, gene profiling	AML	CP & DB	(129)
Gimeno et al, 2022	A novel XAI method (MOM) was applied to an AML cohort of 319 ex vivo tumor samples with 122 screened drugs and whole-exome suquencing for guaranteeing that predictions were interpretable and robust. The predictive performance of MOM was successfully validated in three different large-scale screening experiments with a therapeutic strategy based on the FLT3, CBFβ-MYH11 and NRAS status, which predicted patient response to quizartinib, trametinib, selumetinib and crizotinib.	Gene profiling	AML	DB	(49)
Yen et al, 2022	Cox proportional hazard analysis was performed, and ML algorithms were applied to generate multivariate miRNA panels to predict nilotinib response at treatment-naïve or post-treatment time points. A combination of miR-145 and miR-708 were observed to be effective predictors of NL response in treatment-naïve patients, whereas miR-150 and miR-185 were significant classifiers at 1-month and 3-month post-NL therapy.	IHC, gene profiling	CML	DB & CP	(130)
Laukhtina et al, 2022	An ML-based variable selection method (LASSO regression) was used to determine the prognostic value with a set of preoperative SIR-biomarkers relative to standard clinicopathological variables to improve the selection of patients with mRCC for cell-reducing nephrectomy.	H&E	RCC	CP & DPr	(131)
Cheng et al, 2022	A total of four DNNs (ResNet50, InceptionV3, Xception and Ensemble) were introduced to improve diagnostic accuracy using H&E stained WSIs from patients with HNLs.	H&E	HCC	DA	(132)
Kleppe et al, 2022	A clinical decision support system based on the combination of DoMore-v1-CRC and pathological staging markers was developed to facilitate individualized choice of adjuvant therapy.	H&E	CRC	DA & DB	(133)
Byeon et al, 2022	A DL model was developed to automatically classify digital pathology images of colon lesions obtained from colonoscopy-related specimen with excellent performances for discriminating adenocarcinoma from non-adenocarcinoma lesions.	H&E	CRC	DA	(134)
Schrammen et al, 2022	SLAM was proposed to simultaneously detect tumor and predict genetic alterations from H&E stained WSIs.	H&E	CRC	DA& DB	(18)
Feng et al, 2022	An AI radiopathomics integrated model (RAPIDS) was proposed and validated to predict pCR in patients with locally advanced CRC using pretreatment MRI and H&E stained WSIs, which demonstrated an outperformance.	H&E, MRI	CRC	CP	(135)
Ding et al, 2022	A graph neural network approach was proposed to predict the comprehensive evaluation cross-level molecular profiles of genetic mutations, copy number alterations and functional protein expression from WSIs with favorable results.	H&E	CRC	DA & CP	(136)
Wang et al, 2022	A nomogram model combining ML pathomics, radiomics features, Immunoscore and clinical factors was constructed to predict the postoperative outcome of patients with CRC with lung metastasis, which demonstrated an outperformance.	H&E, MRI, IHC	CRC lung metastasis	CP	(137)
Sundar et al, 2022	A random forest ML model was applied on NanoString profiles to identify a gene signature to select patient benefits from paclitaxel treatment.	Gene profiling	GC	DB	(138)
Pfob et al, 2022	The feasibility of an ML algorithm (intelligent VAB) to identify exceptional responders to NST was evaluated in patients with BC with a favorable result.	H&E	BC	DB & CP	(139)
Prat et al, 2022	A supervised learning algorithm, HER2DX, was used to predict pCR and survival outcome from the gene profiling of early-stage HER2⁺ BC, with an outperformance.	Gene profiling	BC	CP	(140)
Farinella et al, 2022	A DSS developed by an ML-based pipe-line was applied to analyze an open-source HGSOC proteomic dataset, and it displayed high discerning ability on a dataset of HGSOC biopsies with the resulting output consisting of a combination of three highly discriminating proteins (TOP1, PDIA4 and OGN), which could be of interest for further clinical and experimental validation.	Gene profiling	HGSOC	DB	(141)
Meena and Hasija 2022	The XGBoost ML models (applications of XAI) were trained on binary classification datasets comprising expression data of 40 SCC, 38 AK and 46 normal healthy skin samples, resulting in 23 significant genes being identified and found to be associated with the progression of SCC.	Gene profiling	SCC of the skin	DA & CP	(50)
Park et al, 2022	An AI-powered WSI analyzer of TILs in the TMA that could define three immune phenotypes (inflamed, immune-excluded and immune-desert) was proposed, with an outperformance.	H&E, TMA	LC	DA, CP & DB	(142)
Cheng et al, 2022	A DL-based AI model was developed to automatically analyze the IHC expression of PD-L1 in patients with LC, with a similar performance to pathologists.	IHC	LC	DA & DB	(143)
Vanguri et al, 2022	An ML-based multimodal model was developed to predict immunotherapy response to PD-L1 blockade by capturing the integrated information of medical imaging, histopathologic and genomic features as well as the clinical data from patients with advanced NSCLC.	H&E, IHC, gene profiling	LC	CP & DB	(144)
Liu et al, 2022	A novel processing pipeline for nuclear segmentation, cell-level image feature extraction, and patient-level feature aggregation was introduced to facilitate the evaluation of pathological prognosis in patients with neuroblastoma.	H&E	NB	CP	(145)
Altini et al, 2022	The performance of NDG-CAM in nuclei detection was demonstrated to be in-line with state-of-the-art methods in five datasets based on histopathology WSIs.	H&E	Five datasets, including BC	DA	(48)
Sorin et al, 2023	A highly multiplexed IMC was applied to characterize the cellular landscape of he LUAD TIME, which demonstrated an outperformance in patient prognosis analysis using multiplex immunofluorescence staining techniques to label immune cells in the TIME and combining it with AI.	Gene profiling	LUAD	CP & DB	(39)
Rakaee et al, 2023	An ML-based TIL scoring approach was developed to evaluate TILs associated with clinical outcomes in patients with advanced NSCLC who received anti-PD-L1 monotherapy. The result was the best evidence to date that TILs are closely associated with a benefit from ICI.	H&E	NSCLC	CP & DB	(22)
Foersch et al, 2023	A multi-stain DL model utilizing AI was applied to determine the AImmunoscore in >1,000 patients with CRC for the prediction of prognosis and therapy, with a favorable result.	H&E, IHC	CRC	CP & DB	(146)
Liu et al, 2023	An ML-based pipeline was introduced to identify the CMBs from the TCGA-LGG cohort and assess the association between CMS and the prognosis and treatment response in LGG, with an outperformance.	H&E	LGG	CP & DB	(147)
Cao et al, 2023	An unsupervised generative adversarial network was applied to validate ultraviolet-photoacoustic-microscopy images o allow for rapidly pathologic diagnoses of bone tissue and aid the intraoperative determination of tumor margins.	H&E, frozen sections	Bony sarcomas	DA	(148)
Wu et al, 2023	A lymph node metastases diagnostic mode was developed to assess the clinical effect of an AI-assisted workflow.	H&E	BLCA	DA	(149)

[i] ADH, atypical ductal hyperplasia; AECOX, AutoEncoder with Cox regression network; AI, artificial intelligence;AML, acute myeloid leukemia; ANN, artificial neural network; BC, breast cancer; BCCs, basal cell carcinomas; BLCA, bladder cancer; BL, Burkitt lymphoma; CC, cervical cancer; CAD, computer-aided diagnostic; cfDNA, cell-free DNA; CISH, chromogenic in situ hybridization; CMBs, cellular morphometric biomarkers; CMS, cellular morphometric subtypes; CML, Chronic myeloid leukemia; CNN, convolutional neural network; CRC, colorectal cancer; DA, diagnostic accuracy; DB, drug benefit; DCNN, deep convolutional neural networks; DLBCL, diffuse large B-cell lymphoma; DLS, deep learning signature; DPr, disease prevention; DSS, decision support system; DTI, diffusion tensor imaging; CP, cancer prognostication; DCIS, ductal carcinoma in situ; DRD, DNA repair deficiency; EMaGACO, Expectation-maximization-driven geodesic active contour with overlap resolution; FCN, fully convolutional neural networks; FISH, fluorescence in-situ hybridization; FL, follicular lymphoma; GBM, glioblastoma multiforme; GC, gastric carcinoma; GO, gene ontology; GSC, genomic sequencing classifier; HCC, hepatocellular carcinoma; HGSOC, High-grade serous ovarian cancer; HNCA, head and neck carcinoma; HNLs, hepatocellular nodular lesions; IBL, intraductal breast lesions; ICC, immunocytochemistry; IHC, immunohistochemistry; LC, lung cancer; LGG, lower grade glioma; LNM, lymph node metastasis; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; MIBC, muscle-invasive bladder cancer; miRNA, microRNA; miR, microRNA; ML, machine learning; mLR, multiclass logistic regression; MM, malignant mesothelioma; MOM, multi-dimensional module optimization; mRCC, metastatic renal cell carcinoma; MSI, microsatellite instability; NB, neuroblastoma; NEP, neighboring ensemble predictor; NFD, negative for dysplasia; NGS, next-generation sequencing; NPC, nasopharyngeal carcinoma; NSCLC, non-small cell lung cancer; NST, neoadjuvant systemic treatment; OC, ovarian cancer; OHbIC, oral cavity histomorphometric-based imageclassifier; OS, osteosarcoma; OSCC, oropharyngeal squamous cell carcinoma; PCa, prostate cancer; PAP, Papanicolau; pCR, pathological complete response; RNA-seq, RNA sequencing; SC-CNN, spatially constrained convolutional neural network; SCLC, small cell lung carcinoma; scRNA-seq, single cell RNA sequencing; SIMMS, Subnetwork Integration for Multi-Modal Signatures; SLAM, slide-level assessment model; SLNB, sentinel lymph node biopsy; SLL, small lymphocytic lymphoma; SSAE, stacked sparse autoencoder; SCC, squamous cell carcinoma; sTILs, stromal tumor-infiltrating lymphocytes; STS, soft tissue sarcomas; SVM, support-vector-machine; TA, tubular adenoma; TAMs, tumor-associated macrophages; TCGA, Cancer Genome Atlas; TILs, tumor-infiltrating lymphocytes; TIME, tumor immune microenvironment; TMA, tissue microarray; TME, tumor microenvironment; TNBC, triple-negative breast cancer; UC, urothelial carcinoma; UCEC, uterine corpus endometrial cancer; UDH, usual ductal hyperplasia; VAB, vacuum-assisted biopsy; wFDT, weighted fuzzy decision trees; WSIs, whole-slide images; XAI, explainable artificial intelligence.

Notably, one of the biggest challenges in neuroanatomy [the automatic segmentation of neuronal structures presented in stacks of electron microscopy (EM) images] at ISBI 2012 was won by Ciresan et al (8) by employing a special type of deep ANN as a pixel classifier (Table I). However, there were two key disadvantages to the strategy raised by Ciresan et al, which were that it was slow and required a trade-off between localization accuracy and the use of context. Fortunately, a noteworthy finding was reported by Shelhamer et al (9) in 2017, where the fully convolutional networks (FCN) model could convert semantic-level images into pixel-level images using a convolution layer, instead of the full connection layer of the original segmentation network, which promoted the rapid development of image segmentation. For instance, Signaevsky et al (10) revealed that the FCN implemented in PyTorch was efficient and well suited for the practical application of WSIs derived from 22 autopsy brains from patients with tauopathies, which produced high precision and recall in naïve WSI semantic segmentation. Yi et al (11) also demonstrated that a finely-tuned FCN could be applied to analyze the microvessels of H&E stained histology images from patients with lung adenocarcinoma (LUAD) for prognostication. As a result of this FCN development, the U-Net model was presented in 2015 by Ronneberger et al (12) who, through modifying and extending the FCN, which supported automatic recognition and precise segmentation of characteristic lesions in EM stacks, beat the network presented by Ciresan et al (8) at ISBI 2015 (Table I). Furthermore, building on past efforts, Ronneberger et al developed an ImageJ plugin that enabled non-ML practitioners to solve problems with U-Net on either a local computer or a remote server/cloud service, which provided a generic DL-based software package for 2D and 3D cell detection and segmentation (12,13). A case in point, Lee et al (14) in 2020 used the modified U-Net to perform spatial analysis for the tumor microenvironment (TME) in WSIs from breast invasive carcinoma cases to replenish cancer classification and prediction, which demonstrated the effectiveness of the U-Net approach and the quantitative estimates derived from the spatial analysis. In addition, U-Net was proposed to aid the epidermis segmentation task on WSIs from malignant melanoma by Oskal et al (15), in which a superior performance of U-Net compared with existing techniques was observed.

In short, AI has been involved in the growth of DP, which has indicated that AI may bring a future of intelligent diagnostic robots.

Molecular pathology (MP)

In addition to the use of conventional pathology as the basis for DP, AI has also demonstrated unparalleled advantages and potential in MP, which is a sub-microscopic discipline of pathology. MP predicts risk, facilitates diagnosis and improves prognostication based on a complete understanding of the biological impact of specific molecular variants, mutations and dysregulations in diseases. MP techniques, which were rooted in fundamental molecular biology discoveries of the 1940s to 1980s (for example, PCR developed by Mullis in the 1980s), have moved from genomics to proteomics [such as next-generation sequencing, single cell RNA sequencing, fluorescence in situ hybridization, NMR spectroscopy and iTRAQ-MALDI-MS/MS], which has been driven by an increasing number of druggable targets and predictive biomarkers. Undeniably, AI fostered the development of MP, which was mainly encompassed by molecular profiles for risk prediction and molecular diagnostics, especially genomic diagnostics (16). For example, Woerl et al (17) formulated a novel ResNet-based approach (termed mibCNN) to predict the molecular subtype of muscle-invasive bladder cancer (MIBC) by analyzing two datasets of WSIs from H&E staining alone. The results demonstrated that DL could be trained to predict the significant molecular characteristics of MIBC from digital images derived from H&E slides only, potentially resulting in an improvement in managing this disease. Similarly, Schrammen et al (18) described a comparative performance with a state-of-the-art method (referred to as the slide-level assessment model), which simultaneously detected tumors and predicted the genetic alterations (including microsatellite instability/mismatch repair deficiency status and BRAF mutational status) of CRC from H&E-stained histopathological WSIs.

Numerous studies have focused on the risk stratification for cancer, especially for the globally leading cause of cancer deaths, LC. For instance, Choi and Na (19) proposed a novel DL-based risk stratification model to predict prognosis for patients with LUAD based on a gene co-expression network, which exhibited a significant association with patient prognosis independent of other clinicopathological features. Moreover, Choi et al (20) designed a robust Genomic Sequencing Classifier (GSC; a second-generation risk stratification algorithm) based on whole transcriptome RNA sequencing to validate the gene expression values predicting clinicopathological features of several LC cohorts. Compared with the Bronchial Genomic Classifier (the first generation), the GSC presented an optimized risk stratification across several independent LC cohorts, which aided physicians in capturing the optimal actionable information for the precision medicine of patients.

Growing evidence has shown that the TME, which is a complex physical and biochemical system comprising tumor cells, tumor stromal cells, vascular cells, immune cells and cytokines, plays an important role in tumor initiation, progression, metastasis and drug resistance. One of the most notable advances facilitated by AI has been in the understanding of the tumor-infiltrating lymphocytes (TILs) of the TME. Based on information regarding the TME from multiple studies in which immune content response data from The Cancer Genome Atlas (TCGA) were comprehensively analyzed, Saltz et al (21) framed a comprehensive methodology (termed Computational Staining) based on CNNs to map TILs and analyze the molecular and clinical correlation of TILs from H&E-stained WSIs containing 13 TCGA cancer types (Table II). In the present study, TIL map structural patterns with competitive performance were generated and it was revealed that the TIL patterns were associated with tumor and immune molecular features, cancer type and patient survival, which laid a new milestone for TME research (Fig. 3). Rakaee et al (22) described an attempt to associate an ML-based assessment of TILs from standard histological images with the outcomes of anti-PD-L1 monotherapy in patients with advanced non-small cell LC (NSCLC), which provided evidence that AI-based patient TIL assessment may enhance precision therapy.

Figure 3.

Overview of the AI-assisted analysis of the TME in cancer pathology. Current TME analysis approaches in AI-integrated cancer pathology generally rely on WSIs and gene profiling. Classically, numerous efforts have been devoted to capturing information about the TME from H&E, IHC or TMA based WSIs, such as H&E of 13 The Cancer Genome Atlas cancer types (21), H&E of BC (72), H&E and IHC of BC (71), IHC of BC (14), TMA of LC (142) and IHC of LC (90,143). Although these efforts have demonstrated an outperformance in the study of the TME, the platform of WSIs relying on images has limited the available information of TME for the AI-based approaches. Alternatively, with the emergence of cancer biomarkers and corresponding technologies, more foci from the TME from gene profiling by AI-based single cell RNA sequencing or next-generation sequencing have been reported, such as for BC (23,113), CRC liver metastasis (24,28), CRC (25), ovarian cancer (30), nasopharyngeal carcinoma (37) and osteosarcoma (38). These AI-based methods have provided new perspectives for research on the TME and made it prone to catching the information linked to the cross-talk between elements of the immune system with tumor cells in the TME. AI, artificial intelligence; BC, breast cancer; CRC, colorectal cancer; IHC, immunohistochemistry; LC, lung cancer; TMA, tissue microarray; TME, tumor microenvironment; WSIs, whole slide images.

As another key component of the TME, tumor-associated macrophages (TAMs) were studied by Bao et al (23) in triple-negative BC through an integrated analysis of single-cell and bulk RNA-seq. A DL approach based on the neural network, PyTorch, for the TAM-related gene signature was observed to display high accuracy in predicting the immunotherapy response (Fig. 3). Moreover, Cancian et al (24) demonstrated that CNN-based DeepLab-v3 could accurately recognize TAMs from the background, and separate different TAMs in marked contrast to the other two CNN-based models, which may satisfy the requirements of clinical practice for the characterization of TAM-related metrics in human colorectal liver metastases. In addition, Bian et al (25) designed a DL-based computational framework (termed ImmunoAIzer) to analyze the spatial distribution of immune cells and cancer cells in the TME and to detect gene mutations in colon cancer, which provided comprehensive information of cell distribution and tumor gene mutation status more efficiently and less expensively.

Moreover, tumor heterogeneity (including the contribution of the TME), an unneglectable factor affecting tumor evolution and therapy resistance, has also rapidly become a hot field of tumor phenotype research facilitated by AI. Undoubtedly, AI has made it easier to capture information from heterogeneous datasets to achieve accurate predictions and recognition patterns (26). For instance, since single-cell transcriptome sequencing technology was first reported by Tang et al (27) in 2009, and it has been widely applied in understanding molecular mechanisms and cellular properties of numerous biological processes, especially the tumor heterogeneity in various tumor types, such as CRC (28), ovarian cancer (29,30), head and neck cancer (31) and hepatocellular carcinoma (HCC) (32). In the light of this, Del Giudice et al (26) attempted to comprehensively review AI-based single-cell RNA-seq experiments in cancer from six network databases including the Genomic Data Commons Data Portal, ENCODE, Gene Expression Omnibus, Sequence Read Archive, Human Cell Atlas and Single Cell Portal, and sorted these experiments according to the tasks of AI models. These experiments, as well as the latest data, were categorized as defining cancer subtypes and cell clones (33–36), parsing tumor immune microenvironment (37–39), identifying new drug biomarkers (40), assessing and predicting disease recurrence and patient survival (41–45), detecting new putative actionable vulnerabilities (44,46) and predicting tumor immunoprofiling (47) (Table II) (26). More notably, the contributions of AI to the needs of cancer genomics for improving patient management were highlighted (26).

With the promise of continuing advances in the autonomous systems, explainable AI (XAI; the emerging generation of AI), which modified DL techniques to learn explainable features, has come to the fore in precision medicine, guaranteeing that predictions were more comprehensible and robust both in DP and MP. For example, Altini et al (48) tested the performance of NDG class activation mapping (NDG-CAM), a gradient-weighted CAM (one of the XAI applications) based method for nuclei detection in histopathology WSIs. NDG-CAM was observed to outperform in five datasets compared with other state-of-the-art methods. Moreover, Gimeno et al (49) developed a novel XAI method (multi-dimensional module optimization; MOM), which was applied to an acute myeloid leukemia (AML) cohort of 319 ex vivo tumor samples with 122 screened drugs and whole-exome sequencing. The predictive performance of MOM in AML was successfully validated in three different large-scale screening experiments with a therapeutic strategy based on the FLT3, CBFβ-MYH11 and NRAS status, which predicted patient response to quizartinib, trametinib, selumetinib and crizotinib (49). In addition, Meena and Hasija (50) described the outperformance of XGBoost ML models in identifying 23 significant genes of squamous cell cancer (SCC) of the skin, which may be the diagnostic and prognostic biomarkers of patients with SCC.

Taken together, AI has empowered MP by challenging itself, which mitigated various broken promises of precision medicine to improve the poor prognosis of different cancer types.

Application

To date, integration of AI and the cancer pathology workflow has achieved some breakthroughs for precision medicine, these applications were summarized and classified into four groups according to their principal tasks in precision medicine: DA, CP, drug benefit (DB) and disease prevention (DPr) (Figs. 1B and 2). A review of the literature from Google scholar and PubMed was conducted using the following terms: ‘AI’ AND ‘pathology’ AND ‘cancer’ AND ‘diagnosis’, or ‘AI’ AND ‘pathology’ AND ‘cancer’ AND ‘therapy’, or ‘AI’ AND ‘pathology’ AND ‘cancer’ AND ‘prognosis’. or ‘AI’ AND ‘pathology’ AND ‘cancer’ AND ‘prevention’. In total, ~20,000 available texts (Fig. 1B) were retrieved (most published studies had an impact factor of ≥5) and these principal applications of AI-integrated cancer pathology were listed chronologically for the first time (to the best of our knowledge) in Table II.

While reducing the workload of oncologists, the vast majority of these studies were concerned with integration of AI methods into cancer pathology to improve DA, including identification, classification, detection and discrimination of cancer and other malignancies. This means that integration of AI and cancer pathology has been used primarily as an aid to cancer diagnosis, and a great deal of effort has been devoted to address the bottlenecks in improving DA, such as low speed and efficiency of scanning slides, image distortion, spatial distribution complexities, limited recognition or auto-recognition ability and inefficient analytical capabilities, as well as the shortfalls for process integration (i.e., tradeoffs between performance and new features), etc. (Table II). Thus, these reasonable efforts have served to enhance the roles of cancer pathology in precision medicine, as far as possible, but they are not sufficient to keep pace with the ever-growing needs of personalized treatment.

Although the body of literature in the field of integration of AI into cancer pathology for CP is relatively small, the prediction of cancer prognosis has gained more attention, which is mainly attributed to the growth of MP. It is well-known that a cancer prognosis typically involves multiple physicians from different specialties using different subsets of biomarkers and multiple clinical factors, including the age and general health of the patient, the location, type, grade and size of the cancer, as well as the tumor-node-metastasis staging of the tumor. There are three predictive foci concerned with cancer prognostication, including the prediction of cancer susceptibility (i.e., risk assessment), the prediction of cancer recurrence and the prediction of cancer survivability. Differentiating these cancer predictive foci from biomarkers and factors may pose a problem. In this regard, as a consequence of the outperformance of AI integrated into MP driven by the rapid development of cancer biomarkers, which cover a broad range of biochemical entities, such as nucleic acids, proteins, sugars, small metabolites and cytogenetic and cytokinetic parameters, as well as entire tumor cells found in bodily fluids, improving predictions of prognosis via these biomarkers were observed in 10 types of TCGA cancer (41), soft tissue sarcomas (44), BC (46), cervical cancer (CC) (61), bladder cancer (73), glioma (82,127), AML (129) and chronic myelogenous leukemia (130) (Table II). Therefore, AI has helped clinicians to accurately stratify risk in patients via the assessment of various cancer biomarkers.

Similarly, encouraged by the emergence of cancer biomarkers and corresponding technologies, there has been a gradual increase in the studies related to the integrations of AI and cancer pathology in predicting drug benefits by evaluating novel cancer biomarkers, including TIGIT (38), activation of BH3-only proteins (39), amino acid synthesis and interconversion (39), tumor human leukocyte antigen peptide (47), CTLA4 (66), oncotype DX and other gene expression (69), 21 genes (73), dysplasia (129), TOP1, PDIA4 and OGN (141) (Table II). Accordingly, these studies have provided compelling evidence for the feasibility of developing targeted therapies against these molecules for patients to obtain optimized medical treatment.

Regardless of the number of papers focusing on DPr in terms of integration of AI and cancer pathology is the least among the four main tasks, the influence of these studies on cancer precision medicine remains vast. As a part of precision medicine, cancer prevention, which is generalized as being ‘founded on describing the burden of cancer, identifying the causes, and evaluating and implementing preventive interventions’, means there is ‘a significant amount of work to be conducted’. Cancer prevention has achieved some success in reducing the global cancer incidence. The most convincing evidence is that tobacco control efforts and smoking cessation efforts as well as additional efforts, such as computed tomography for LC screening, since the 1960′s will continue to reduce global LC incidence (150). Due to the advances in understanding cancer biology, precision prevention approaches have been made feasible in various types of cancer with the assistance of AI, such as in CRC (80,128), BC (91), CC (96,107), liver cancer (128), LUAD (128) and renal cell cancer (131) (Table II). It is hypothesized that these attempts will reduce global cancer burden in the future.

The integration of AI into these practices has allowed technologies to handle large amounts of data related to cancer pathology, which has not only alleviated the workload pressure of oncologists but has also improved the accuracy of diagnosis and prediction of cancer prognosis and drug benefits, as well as promoted DPr for precision medicine.

Discussion

At present, cancer remains a major public health problem worldwide, and cancer death was adversely affected by the coronavirus 2019 (COVID-19) global pandemic. Thus, there is a pressing need to develop novel and effective strategies for cancer precision medicine. Although growing evidence indicates that a combination of cancer pathology and AI can bring exciting changes to cancer precision medicine, a large number of technical, ethical and legal challenges still need to be appropriately addressed.

Firstly, there are four main limitations in terms of technology level in the development of AI in cancer pathology, including: The quality and standardization of pathological images and WSIs (especially the images from 2D to 3D, if possible, or even 4D), the optimization of data, the interpretability of data and the verification of algorithms. Therefore, more efforts than could possibly be reviewed in the present review have been devoted to address these problems. For example, Zheng et al (151) created a Graph-Transformer fusing a graph-based representation of a WSI, and developed a graph-based digital pathology visual transducer to predict disease grade, which outperformed compared with current state-of-the-art methods for WSI classification. It is hypothesized that these behaviors could benefit cancer pathology ultimately with faster and more accurate diagnoses, as well as faster screening. Moreover, Qiao et al (152) described the rationalized DL-structured illumination microscopy and lattice light sheet microscopy system to build a minimally invasive 2D/3D live cell imaging system for following intracellular dynamics and trajectory, which provided the possibility that a novel system may be created in the future by extending the aforementioned capabilities to monitor the therapeutic effect of the treatments to tumor cells in vivo safely. In addition, owing to the fact that the effectiveness of these systems was limited by the inability of the machine to explain its thoughts and actions to human users, XAI, a suite of ML techniques, was created to promote safety and clarity by showing how decisions are made in AI models, especially in critical tasks, such as drug screening with genetic events (49,50) and drug-drug interaction predictions (153).

Nevertheless, cancer is a highly complex disease involving a cascade of microscopic and macroscopic changes with mechanisms and interactions that are not yet fully understood. With the continuous discovery of novel cancer biomarkers and the innovation of technologies, cancer biomarkers that provide insights into the state and course of disease in the form of quantitative or qualitative measurements have made the fusion of multiple medical perspectives to guide patient management feasible, which AI has promoted in multi-modal patterns. Since Chaudhary et al (83) first employed a DL-based model to identify multi-omics features to predict HCC survival risk, AI-guided research linked to multi-omics has emerged. For instance, Feng et al (135) proposed and evaluated an AI radiopath-omics integrated model (termed RAPIDS) to predict pathological complete response (pCR) to neoadjuvant chemoradiotherapy in patients with locally advanced rectal cancer via pretreatment MRI and H&E stained WSIs. The results suggested that RAPIDS was able to predict pCR to neoadjuvant chemoradiotherapy, based on pretreatment radiopath-omics images, with high accuracy and robustness, which demonstrated a successful fusion of imagen-omics, path-omics and clinical data to predict patient prognosis (Fig. 4). Furthermore, Wang et al (137) built a nomogram model combining path-omics, radio-omics, protein-omics and clinical data to predict the postoperative outcome of patients with CRC and lung metastasis, which demonstrated an outperformance in risk prediction. Moreover, Vanguri et al (144) recently developed an ML-based multimodal model to predict immunotherapy response to PD-L1 blockade by capturing the information through a combination of imagen-omics, path-omics, genomics and clinical data from patients with advanced NSCLC. The model provided a quantitative rationale for using AI-guided multimodal integration to improve prediction of the drug benefits and outcomes in patients with NSCLC. Hence, synergies of cancer pathology with other medical tools could provide more information to the clinic to make an accurate and rapid decision in personalized treatments for patients. However, the well-known failed implementation of AI in oncology (the IBM Watson program at MD Anderson) is a reminder that multiple deployment challenges must be overcome prior to the integration of a pathology AI system into a clinical work environment (154,155). It is hypothesized that exploring the potential advantages of multimodal integration of path-omics, genomics, imagen-omics, protein-omics, epigenomics and other omics, as well as clinical data to make appropriate management decisions and improve patient outcomes may be the most challenging issue of cancer precision medicine in the future.

Figure 4.

Multimodal integration of AI into precision medicine for patients with cancer. Since numerous issues stemming from the complexity of adopting AI in medical practice have been presented, multimodal features to improve the predictive capacity of AI-based medical approaches has successfully been applied in certain cancer types. For instance, the fusion of multi-omics and clinical data was used to determine an appropriate treatment management and predict the prognosis of patients with cancer, which was verified with outperformances in colorectal (135,137) and lung (144) cancer. However, there are still numerous challenges related to technology, ethics and the law of AI-based medical practice that need to be appropriately addressed. Thus, exploring the potential advantages of multimodal integration of multi-omics and clinical data to make appropriate treatment management decisions and improve patient outcomes may be the most challenging issue of cancer precision medicine in the future. AI, artificial intelligence.

Notably, recent attempts to verify the capability of ChatGPT in effecting precision medicine have been reported since ChatGPT was released at the end of 2022, as it was reasonable to assume that using larger language models to enhance relationships through better communication would have a beneficial impact on patients (156,157). Sinha et al (158) assessed the capability of ChatGPT in solving higher-order reasoning regarding pathology and found that ChatGPT may offer meaningful responses but also current limitations.

Secondly, the vulnerability of ethics in the development of AI in cancer pathology cannot be ignored. With the introduction of WSI scanners, slides (as high-resolution digital images) can be captured, stored and transmitted electronically, and oncologists can annotate and label images as part of their reports. When the images and reports are stored in the cloud, the process allows third parties to share the information and associated data for research or other purposes. Moreover, oncologists have an obligation to keep identifiable patient information confidential (except with expressed consent), however, we consider that more could be done to ensure patient privacy and data security in this process. For instance, Chauhan and Gullapalli (159) posed three key foundational principles of ethical AI in the context of cancer pathology: Transparency, accountability and governance, to guide future practice. This viewpoint is appreciated, and it deserves popularizing in the clinic.

Thirdly, it has been recognized that the adoption of a new technology cannot be accomplished without some changes to the human and social aspects of the organization. The lack of global standardization has prevented the reusing of AI-integrated cancer pathology data to meet the broad range of patient safety and quality reporting requirements, which has been attributed to the following potential factors: i) The standardization for storage and retrieval of medical images; ii) the high standards and regulations that surround the cancer pathology laboratory; iii) the adoption of a global diagnostic system (i.e., as with the World Health Organisation); iv) the normalization of validation systems; v) the establishment of corresponding legal liabilities related to AI; and vi) the oncologist, etc. To facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, Warnat-Herresthal et al (160) proposed a decentralized ML approach termed Swarm Learning (SL), which united edge computing, blockchain-based peer-to-peer networking and coordination, while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. They chose four cases of heterogeneous diseases (COVID-19, tuberculosis, leukemia and lung pathologies) with >16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as >95,000 chest X-ray images to validate the feasibility of using SL to develop disease classifiers using distributed data. The results demonstrated that SL classifiers outperform those developed at individual sites, which allows for further study in more cancer types (160). It is hoped that in addition to establishing new rules and laws to provide a comprehensive guarantee for the development of AI in oncology pathology, more options could be explored to integrate the medical data for precision medicine to the extent allowed by laws (Fig. 4).

In conclusion, the 21st century has witnessed tremendous breakthroughs in AI and cancer pathology, and most of these creations and innovations have improved and advanced understanding of these systems. As such, it is hoped to improve understanding of the concepts related to these systems and to help others realize the goals of precision medicine.

Acknowledgements

Not applicable.

Funding

The present study was partly supported by 2017 Elite Talents Training Plan of the Third Affiliated Hospital of Guangzhou Medical University and 2023 Plan on enhancing scientific research in GMU.

Availability of data and materials

Not applicable.

Authors' contributions

JP designed the review, wrote and revised the manuscript with input from BL, JF, QZ, QJ. BL, JF, QZ, ND, QJ and JP performed and analyzed the literature search. JF and BL made the figures. QZ, BL and JP completed the major part of tables. QJ participated in revising the review critically for the important intellectual content. Data authentication is not applicable. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

AI	artificial intelligence
ANN	artificial neural network
BC	breast cancer
CC	cervical cancer
CNN	convolutional neural network
CP	cancer prognostication
CRC	colorectal cancer
DA	diagnostic accuracy
DL	deep learning
DP	digital pathology
DPr	disease prevention
EM	electron microscopy
FCN	fully CNNs
GSC	genomic sequencing classifier
HCC	hepatocellular carcinoma
LC	lung cancer
LUAD	lung adenocarcinoma
MIBC	muscle-invasive bladder cancer
MIL	multiple-instance learning
ML	machine learning
MP	molecular pathology
NSCLC	non-small cell LC
PCa	prostate cancer
pCR	pathological complete response
SL	swarm learning
TCGA	The Cancer Genome Atlas
TILs	tumor-infiltrating lymphocytes
WSIs	whole-slide images

References

1	LeCun Y, Bengio Y and Hinton G: Deep learning. Nature. 521:436–444. 2015. View Article : Google Scholar : PubMed/NCBI
2	Kriegeskorte N and Golan T: Neural network models and deep learning. Curr Biol. 29:R231–R236. 2019. View Article : Google Scholar : PubMed/NCBI
3	Barisoni L, Lafata KJ, Hewitt SM, Madabhushi A and Balis UGJ: Digital pathology and computational image analysis in nephropathology. Nat Rev Nephrol. 16:669–685. 2020. View Article : Google Scholar : PubMed/NCBI
4	Goodfellow IJ, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A and Bengio Y: Generative Adversarial Networks. 10–Jun;2014.doi: 10.48550/arXiv.1406.2661.
5	Fanous MJ and Popescu G: GANscan: Continuous scanning microscopy using deep learning deblurring. Light Sci Appl. 11:2652022. View Article : Google Scholar : PubMed/NCBI
6	Barker J, Hoogi A, Depeursinge A and Rubin DL: Automated classification of brain tumor type in whole-slide digital pathology images using local representative tiles. Med Image Anal. 30:60–71. 2016. View Article : Google Scholar : PubMed/NCBI
7	Campanella G, Hanna MG, Geneslaw L, Miraflor A, Werneck Krauss Silva V, Busam KJ, Brogi E, Reuter VE, Klimstra DS and Fuchs TJ: Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med. 25:1301–1309. 2019. View Article : Google Scholar : PubMed/NCBI
8	Ciresan D, Giusti A, Gambardella L and Schmidhuber J: Deep neural net-works segment neuronal membranes in electron microscopy images. NIPS. 2852–2860. 2012.
9	Shelhamer E, Long J and Darrell T: Fully Convolutional networks for semantic segmentation. IEEE Trans Pattern Anal Mach Intell. 39:640–651. 2017. View Article : Google Scholar : PubMed/NCBI
10	Signaevsky M, Prastawa M, Farrell K, Tabish N, Baldwin E, Han N, Iida MA, Koll J, Bryce C, Purohit D, et al: Artificial intelligence in neuropathology: Deep learning-based assessment of tauopathy. Lab Invest. 99:1019–1029. 2019. View Article : Google Scholar : PubMed/NCBI
11	Yi F, Yang L, Wang S, Guo L, Huang C, Xie Y and Xiao G: Microvessel prediction in H&E Stained Pathology Images using fully convolutional neural networks. BMC Bioinformatics. 19:642018. View Article : Google Scholar : PubMed/NCBI
12	Ronneberger O, Fischer P and Brox T: U-net: Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer Assisted Intervention (MICCAI). Springer, LNCS; pp. 9351pp. 234–241. 2015
13	Falk T, Mai D, Bensch R, Çiçek Ö, Abdulkadir A, Marrakchi Y, Böhm A, Deubner J, Jäckel Z, Seiwald K, et al: U-Net: Deep learning for cell counting, detection, and morphometry. Nat Methods. 16:67–70. 2019. View Article : Google Scholar : PubMed/NCBI
14	Lee S, Zhao Y, Masoud M and Belkasim S: Quantitative spatial analysis on whole slide images using U-net. Computational Biol Bioinform. 8:90–96. 2020. View Article : Google Scholar
15	Oskal KRJ, Risdal M, Janssen EAM, Undersrud ES and Gulsrud TO: A U-net based approach to epidermal tissue segmentation in whole slide histopathological images. SN Applied Sciences. 1:6722019. View Article : Google Scholar
16	Echle A, Rindtorff NT, Brinker TJ, Luedde T, Pearson AT and Kather JN: Deep learning in cancer pathology: A new generation of clinical biomarkers. Br J Cancer. 124:686–696. 2021. View Article : Google Scholar : PubMed/NCBI
17	Woerl AC, Eckstein M, Geiger J, Wagner DC, Daher T, Stenzel P, Fernandez A, Hartmann A, Wand M, Roth W, et al: Deep learning predicts molecular subtype of muscle-invasive bladder cancer from conventional Histopathological Slides. Eur Urol. 78:256–264. 2020. View Article : Google Scholar : PubMed/NCBI
18	Schrammen PL, Ghaffari Laleh N, Echle A, Truhn D, Schulz V, Brinker TJ, Brenner H, Chang-Claude J, Alwers E, Brobeil A, et al: Weakly supervised annotation-free cancer detection and prediction of genotype in routine histopathology. J Pathol. 256:50–60. 2022. View Article : Google Scholar : PubMed/NCBI
19	Choi H and Na KJ: A Risk stratification model for lung cancer based on gene coexpression network and deep learning. Biomed Res Int. 16:29142802018.PubMed/NCBI
20	Choi Y, Qu J, Wu S, Hao Y, Zhang J, Ning J, Yang X, Lofaro L, Pankratz DG, Babiarz J, et al: Improving lung cancer risk stratification leveraging whole transcriptome RNA sequencing and machine learning across multiple cohorts. BMC Med Genomics. 13:1512020. View Article : Google Scholar : PubMed/NCBI
21	Saltz J, Gupta R, Hou L, Kurc T, Singh P, Nguyen V, Samaras D, Shroyer KR, Zhao T, Batiste R, et al: Spatial organization and molecular correlation of tumor-infiltrating lymphocytes using deep learning on pathology images. Cell Rep. 23:181–193. 2018. View Article : Google Scholar : PubMed/NCBI
22	Rakaee M, Adib E, Ricciuti B, Sholl LM, Shi W, Alessi JV, Cortellini A, Fulgenzi CAM, Viola P, Pinato DJ, et al: Association of machine learning-based assessment of tumor-infiltrating lymphocytes on standard histologic images with outcomes of immunotherapy in patients with NSCLC. JAMA Oncol. 9:51–60. 2023. View Article : Google Scholar : PubMed/NCBI
23	Bao X, Shi R, Zhao T, Wang Y, Anastasov N, Rosemann M and Fang W: Integrated analysis of single-cell RNA-seq and bulk RNA-seq unravels tumour heterogeneity plus M2-like tumour-associated macrophage infiltration and aggressiveness in TNBC. Cancer Immunol Immunother. 70:189–202. 2021. View Article : Google Scholar : PubMed/NCBI
24	Cancian P, Cortese N, Donadon M, Di Maio M, Soldani C, Marchesi F, Savevski V, Santambrogio MD, Cerina L, Laino ME, et al: Development of a deep-learning pipeline to recognize and characterize macrophages in colo-rectal liver metastasis. Cancers (Basel). 13:33132021. View Article : Google Scholar : PubMed/NCBI
25	Bian C, Wang Y, Lu Z, An Y, Wang H, Kong L, Du Y and Tian J: ImmunoAIzer: A deep learning-based computational framework to characterize cell distribution and gene mutation in tumor microenvironment. Cancers (Basel). 13:16592021. View Article : Google Scholar : PubMed/NCBI
26	Del Giudice M, Peirone S, Perrone S, Priante F, Varese F, Tirtei E, Fagioli F and Cereda M: Artificial intelligence in bulk and single-cell RNA-sequencing data to foster precision oncology. Int J Mol Sci. 22:45632021. View Article : Google Scholar : PubMed/NCBI
27	Tang F, Barbacioru C, Wang Y, Nordman E, Lee C, Xu N, Wang X, Bodeau J, Tuch BB, Siddiqui A, et al: mRNA-Seq whole-transcriptome analysis of a single cell. Nat Methods. 6:377–382. 2009. View Article : Google Scholar : PubMed/NCBI
28	Wu Y, Yang S, Ma J, Chen Z, Song G, Rao D, Cheng Y, Huang S, Liu Y, Jiang S, et al: Spatiotemporal immune landscape of colorectal cancer liver metastasis at single-cell level. Cancer Discov. 12:134–153. 2022. View Article : Google Scholar : PubMed/NCBI
29	Geistlinger L, Oh S, Ramos M, Schiffer L, LaRue RS, Henzler CM, Munro SA, Daughters C, Nelson AC, Winterhoff BJ, et al: Multiomic analysis of subtype evolution and heterogeneity in high-grade serous ovarian carcinoma. Cancer Res. 80:4335–4345. 2020. View Article : Google Scholar : PubMed/NCBI
30	Hao Q, Li J, Zhang Q, Xu F, Xie B, Lu H, Wu X and Zhou X: Single-cell transcriptomes reveal heterogeneity of high-grade serous ovarian carcinoma. Clin Transl Med. 11:e5002021. View Article : Google Scholar : PubMed/NCBI
31	Puram SV, Tirosh I, Parikh AS, Patel AP, Yizhak K, Gillespie S, Rodman C, Luo CL, Mroz EA, Emerick KS, et al: Single-Cell transcriptomic analysis of primary and metastatic tumor ecosystems in head and neck cancer. Cell. 171:1611–1624. 2017. View Article : Google Scholar : PubMed/NCBI
32	Zhang Q, He Y, Luo N, Patel SJ, Han Y, Gao R, Modak M, Carotta S, Haslinger C, Kind D, et al: Landscape and dynamics of single immune cells in hepatocellular carcinoma. Cell. 179:829–845.e20. 2019. View Article : Google Scholar : PubMed/NCBI
33	Cascianelli S, Molineris I, Isella C, Masseroli M and Medico E: Machine learning for RNA sequencing-based intrinsic subtyping of breast cancer. Sci Rep. 10:140712020. View Article : Google Scholar : PubMed/NCBI
34	Yu Z, Wang Z, Yu X and Zhang Z: RNA-Seq-Based breast cancer subtypes classification using machine learning approaches. Comput Intell Neurosci. 2020:47379692020. View Article : Google Scholar : PubMed/NCBI
35	Valle F, Osella M and Caselle M: A Topic modeling analysis of TCGA breast and lung cancer transcriptomic data. Cancers (Basel). 12:37992020. View Article : Google Scholar : PubMed/NCBI
36	Gao F, Wang W, Tan M, Zhu L, Zhang Y, Fessler E, Vermeulen L and Wang X: DeepCC: A novel deep learning-based framework for cancer molecular subtype classification. Oncogenesis. 8:442019. View Article : Google Scholar : PubMed/NCBI
37	Chen YP, Yin JH, Li WF, Li HJ, Chen DP, Zhang CJ, Lv JW, Wang YQ, Li XM, Li JY, et al: Single-cell transcriptomics reveals regulators underlying immune cell diversity and immune subtypes associated with prognosis in nasopharyngeal carcinoma. Cell Res. 30:1024–1042. 2020. View Article : Google Scholar : PubMed/NCBI
38	Zhou Y, Yang D, Yang Q, Lv X, Huang W, Zhou Z, Wang Y, Zhang Z, Yuan T, Ding X, et al: Single-cell RNA landscape of intratumoral heterogeneity and immunosuppressive microenvironment in advanced osteosarcoma. Nat Commun. 11:63222020. View Article : Google Scholar : PubMed/NCBI
39	Sorin M, Rezanejad M, Karimi E, Fiset B, Desharnais L, Perus LJM, Milette S, Yu MW, Maritan SM, Doré S, et al: Single-cell spatial landscapes of the lung tumour immune microenvironment. Nature. 614:548–554. 2023. View Article : Google Scholar : PubMed/NCBI
40	Kong J, Lee H, Kim D, Han SK, Ha D, Shin K and Kim S: Network-based machine learning in colorectal and bladder organoid models predicts anti-cancer drug efficacy in patients. Nat Commun. 11:54852020. View Article : Google Scholar : PubMed/NCBI
41	Ching T, Zhu X and Garmire LX: Cox-nnet: An artificial neural network method for prognosis prediction of high-throughput omics data. PLoS Comput Biol. 14:e10060762018. View Article : Google Scholar : PubMed/NCBI
42	Katzman JL, Shaham U, Cloninger A, Bates J, Jiang T and Kluger Y: DeepSurv: Personalized treatment recommender system using a Cox proportional hazards deep neural network. BMC Med Res Methodol. 18:242018. View Article : Google Scholar : PubMed/NCBI
43	Huang Z, Johnson TS, Han Z, Helm B, Cao S, Zhang C, Salama P, Rizkalla M, Yu CY, Cheng J, et al: Deep learning-based cancer survival prognosis from RNA-seq data: Approaches and evaluations. BMC Med Genomics. 13:412020. View Article : Google Scholar : PubMed/NCBI
44	van IJzendoorn DGP, Szuhai K, Briaire-de Bruijn IH, Kostine M, Kuijjer ML and Bovée JVMG: Machine learning analysis of gene expression data reveals novel diagnostic and prognostic biomarkers and identifies therapeutic targets for soft tissue sarcomas. PLoS Comput Biol. 15:e10068262019. View Article : Google Scholar : PubMed/NCBI
45	Haider S, Yao CQ, Sabine VS, Grzadkowski M, Stimper V, Starmans MHW, Wang J, Nguyen F, Moon NC, Lin X, et al: Pathway-based subnetworks enable cross-disease biomarker discovery. Nat Commun. 9:47462018. View Article : Google Scholar : PubMed/NCBI
46	Tabl AA, Alkhateeb A, ElMaraghy W, Rueda L and Ngom A: A machine learning approach for identifying gene biomarkers guiding the treatment of breast cancer. Front Genet. 10:2562018. View Article : Google Scholar : PubMed/NCBI
47	Bulik-Sullivan B, Busby J, Palmer CD, Davis MJ, Murphy T, Clark A, Busby M, Duke F, Yang A, Young L, et al: Deep learning using tumor HLA peptide mass spectrometry datasets improves neoantigen identification. Nat Biotechnol. 37:55–63. 2019. View Article : Google Scholar
48	Altini N, Brunetti A, Puro E, Taccogna MG, Saponaro C, Zito FA, De Summa S and Bevilacqua V: NDG-CAM: Nuclei detection in histopathology images with semantic segmentation networks and Grad-CAM. Bioengineering (Basel). 9:4752022. View Article : Google Scholar : PubMed/NCBI
49	Gimeno M, San José-Enériz E, Villar S, Agirre X, Prosper F, Rubio A and Carazo F: Explainable artificial intelligence for precision medicine in acute myeloid leukemia. Front Immunol. 13:9773582022. View Article : Google Scholar : PubMed/NCBI
50	Meena J and Hasija Y: Application of explainable artificial intelligence in the identification of Squamous Cell Carcinoma biomarkers. Comput Biol Med. 146:1055052022. View Article : Google Scholar : PubMed/NCBI
51	Paul A and Prasad Mukherjee D: Mitosis detection for invasive breast cancer grading in histopathological images. IEEE Transactions on Image Processing. 24:4041–4054. 2015. View Article : Google Scholar : PubMed/NCBI
52	Veta M, van Diest PJ, Willems SM, Wang H, Madabhushi A, Cruz-Roa A, Gonzalez F, Larsen AB, Vestergaard JS, Dahl AB, et al: Assessment of algorithms for mitosis detection in breast cancer histopathology images. Med Image Anal. 20:237–248. 2015. View Article : Google Scholar : PubMed/NCBI
53	Ehteshami Bejnordi B, Veta M, Johannes van Diest P, van Ginneken B, Karssemeijer N, Litjens G, van der Laak JAWM; the CAMELYON16 Consortium, ; Hermsen M, Manson QF, et al: Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer. JAMA. 318:2199–2210. 2017. View Article : Google Scholar : PubMed/NCBI
54	Wang CW, Lee YC, Calista E, Zhou F, Zhu H, Suzuki R, Komura D, Ishikawa S and Cheng SP: A benchmark for comparing precision medicine methods in thyroid cancer diagnosis using tissue microarrays. Bioinformatics. 34:1767–1773. 2018. View Article : Google Scholar : PubMed/NCBI
55	Shipp MA, Ross KN, Tamayo P, Weng AP, Kutok JL, Aguiar RC, Gaasenbeek M, Angelo M, Reich M, Pinkus GS, et al: Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning. Nat Med. 8:68–74. 2002. View Article : Google Scholar : PubMed/NCBI
56	Khan U, Shin H, Choi JP and Kim M: wFDT-Weighted Fuzzy Decision Trees for prognosis of breast cancer survivability. AusDM ‘08: Proceedings of the 7th Australasian Data Mining Conference. Vol 87:pp141–152. 2008.
57	Fatakdawala H, Xu J, Basavanhally A, Bhanot G, Ganesan S, Feldman M, Tomaszewski JE and Madabhushi A: Expectation-maximization-driven geodesic active contour with overlap resolution (EMaGACOR): application to lymphocyte segmentation on breast cancer histopathology. IEEE Trans Biomed Eng. 57:1676–1689. 2010. View Article : Google Scholar : PubMed/NCBI
58	Sertel O, Lozanski G, Shana'ah A and Gurcan MN: Computer-aided detection of centroblasts for follicular lymphoma grading using adaptive likelihood-based cell segmentation. IEEE Trans Biomed Eng. 57:2613–2616. 2010. View Article : Google Scholar : PubMed/NCBI
59	Dundar MM, Badve S, Bilgin G, Raykar V, Jain R, Sertel O and Gurcan MN: Computerized classification of intraductal breast lesions using histopathological images. IEEE Trans Biomed Eng. 58:1977–1984. 2011. View Article : Google Scholar : PubMed/NCBI
60	Tuominen VJ, Tolonen TT and Isola J: ImmunoMembrane: A publicly available web application for digital image analysis of HER2 immunohistochemistry. Histopathology. 60:758–767. 2012. View Article : Google Scholar : PubMed/NCBI
61	Gertych A, Joseph AO, Walts AE and Bose S: Automated detection of dual p16/Ki67 nuclear immunoreactivity in liquid-based Pap tests for improved cervical cancer risk stratification. Ann Biomed Eng. 40:1192–1204. 2012. View Article : Google Scholar : PubMed/NCBI
62	Doyle S, Feldman MD, Shih N, Tomaszewski J and Madabhushi A: Cascaded discrimination of normal abnormal, and confounder classes in histopathology: Gleason grading of prostate cancer. BMC Bioinformatics. 13:2822012. View Article : Google Scholar : PubMed/NCBI
63	Wang H, Cruz-Roa A, Basavanhally A, Gilmore H, Shih N, Feldman M, Tomaszewski J, Gonzalez F and Madabhushi A: Mitosis detection in breast cancer pathology images by combining handcrafted and convolutional neural network features. J Med Imaging (Bellingham). 1:0340032014. View Article : Google Scholar : PubMed/NCBI
64	Lewis JS Jr, Ali S, Luo J, Thorstad WL and Madabhushi A: A quantitative histomorphometric classifier (QuHbIC) identifies aggressive versus indolent p16-positive oropharyngeal squamous cell carcinoma. Am J Surg Pathol. 38:128–137. 2014. View Article : Google Scholar : PubMed/NCBI
65	Sirinukunwattana K, Raza SEA, Tsang YW, Snead DR, Cree IA and Rajpoot NM: A Spatially Constrained Deep Learning Framework for Detection of Epithelial Tumor Nuclei in Cancer Histology Images. Conference: 1st International Workshop on Patch-based Techniques in Medical Imaging, Held in Conjunction with MICCAI. 2015.doi: 10.1007/978-3-319-28194-0_19.
66	Yuan Y: Modelling the spatial heterogeneity and molecular correlates of lymphocytic infiltration in triple-negative breast cancer. J R Soc Interface. 12:201411532015. View Article : Google Scholar : PubMed/NCBI
67	Xie Y, Kong X, Xing F, Liu F, Su H and Yang L: Deep Voting: A robust approach toward nucleus localization in microscopy images. Med Image Comput Comput Assist Interv. 9351:374–382. 2015.PubMed/NCBI
68	Sirinukunwattana K, Ahmed Raza SE, Yee-Wah Tsang, Snead DR, Cree IA and Rajpoot NM: Locality sensitive deep learning for detection and classification of nuclei in routine colon cancer histology images. IEEE Trans Med Imaging. 35:1196–1206. 2016. View Article : Google Scholar : PubMed/NCBI
69	Romo-Bucheli D, Janowczyk A, Gilmore H, Romero E and Madabhushi A: Automated tubule nuclei quantification and correlation with oncotype DX risk categories in ER+ breast cancer whole slide images. Sci Rep. 6:327062016. View Article : Google Scholar : PubMed/NCBI
70	Xu J, Xiang L, Liu Q, Gilmore H, Wu J, Tang J and Madabhushi A: Stacked Sparse Autoencoder (SSAE) for nuclei detection on breast cancer histopathology images. IEEE Trans Med Imaging. 35:119–130. 2016. View Article : Google Scholar : PubMed/NCBI
71	Turkki R, Linder N, Kovanen PE, Pellinen T and Lundin J: Antibody-supervised deep learning for quantification of tumor-infiltrating immune cells in hematoxylin and eosin stained breast cancer samples. J Pathol Inform. 7:382016. View Article : Google Scholar : PubMed/NCBI
72	Ali HR, Dariush A, Provenzano E, Bardwell H, Abraham JE, Iddawela M, Vallier AL, Hiller L, Dunn JA, Bowden SJ, et al: Computational pathology of pre-treatment biopsies identifies lymphocyte density as a predictor of response to neoadjuvant chemotherapy in breast cancer. Breast Cancer Res. 18:212016. View Article : Google Scholar : PubMed/NCBI
73	Bartsch G Jr, Mitra AP, Mitra SA, Almal AA, Steven KE, Skinner DG, Fry DW, Lenehan PF, Worzel WP and Cote RJ: Use of artificial intelligence and machine learning algorithms with gene expression profiling to predict recurrent nonmuscle invasive urothelial carcinoma of the bladder. J Urol. 195:493–498. 2016. View Article : Google Scholar : PubMed/NCBI
74	Yu KH, Zhang C, Berry GJ, Altman RB, Ré C, Rubin DL and Snyder M: Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features. Nat Commun. 7:124742016. View Article : Google Scholar : PubMed/NCBI
75	Ali HR, Dariush A, Thomas J, Provenzano E, Dunn J, Hiller L, Vallier AL, Abraham J, Piper T, Bartlett JMS, et al: Lymphocyte density determined by computational pathology validated as a predictor of response to neoadjuvant chemotherapy in breast cancer: Secondary analysis of the ARTemis trial. Ann Oncol. 28:1832–1835. 2017. View Article : Google Scholar : PubMed/NCBI
76	Lu C, Lewis JS Jr, Dupont WD, Plummer WD Jr, Janowczyk A and Madabhushi A: An oral cavity squamous cell carcinoma quantitative histomorphometric-based image classifier of nuclear morphology can risk stratify patients for disease-specific survival. Mod Pathol. 30:1655–1665. 2017. View Article : Google Scholar : PubMed/NCBI
77	Gecer B, Aksoy S, Mercan E, Shapiro LG, Weaver DL and Elmore JG: Detection and classification of cancer in whole slide breast histopathology images using deep convolutional networks. Pattern Recognit. 84:345–356. 2018. View Article : Google Scholar : PubMed/NCBI
78	Yoshida H, Shimazu T, Kiyuna T, Marugame A, Yamashita Y, Cosatto E, Taniguchi H, Sekine S and Ochiai A: Automated histological classification of whole-slide images of gastric biopsy specimens. Gastric Cancer. 21:249–257. 2018. View Article : Google Scholar : PubMed/NCBI
79	Bo L, Huang M, Zhang J, Li Y and Li R: Gastric Pathology Image Recognition Based on Deep Residual Networks. 2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC). 408–412. 2018.
80	Ichimasa K, Kudo SE, Mori Y, Misawa M, Matsudaira S, Kouyama Y, Baba T, Hidaka E, Wakamura K, Hayashi T, et al: Artificial intelligence may help in predicting the need for additional surgery after endoscopic resection of T1 colorectal cancer. Endoscopy. 50:230–240. 2018. View Article : Google Scholar : PubMed/NCBI
81	Mezheyeuski A, Bergsland CH, Backman M, Djureinovic D, Sjöblom T, Bruun J and Micke P: Multispectral imaging for quantitative and compartment-specific immune infiltrates reveals distinct immune profiles that classify lung cancer patients. J Pathol. 244:421–431. 2018. View Article : Google Scholar : PubMed/NCBI
82	Mobadersany P, Yousefi S, Amgad M, Gutman DA, Barnholtz-Sloan JS, Velázquez Vega JE, Brat DJ and Cooper LAD: Predicting cancer outcomes from histology and genomics using convolutional networks. Proc Natl Acad Sci USA. 115:E2970–E2979. 2018. View Article : Google Scholar : PubMed/NCBI
83	Chaudhary K, Poirion OB, Lu L and Garmire LX: Deep learning-based multi-omics integration robustly predicts survival in liver cancer. Clin Cancer Res. 24:1248–1259. 2018. View Article : Google Scholar : PubMed/NCBI
84	Günakan E, Atan S, Haberal AN, Küçükyıldız İA, Gökçe E and Ayhan A: A novel prediction method for lymph node involvement in endometrial cancer: Machine learning. Int J Gynecol Cancer. 29:320–324. 2019. View Article : Google Scholar : PubMed/NCBI
85	Turkki R, Byckhov D, Lundin M, Isola J, Nordling S, Kovanen PE, Verrill C, von Smitten K, Joensuu H, Lundin J, et al: Breast cancer outcome prediction with tumour tissue images and machine learning. Breast Cancer Res Treat. 177:41–52. 2019. View Article : Google Scholar : PubMed/NCBI
86	Fang M, Zhang W, Dong D, Zhou J and Tian J: Predicting histopathological findings of gastric cancer via deep generalized multi-instance learning. Proceedings of the SPIE. Vol 10949:id. 109491Q. pp62019.
87	Mori H and Miwa H: A histopathologic feature of the behavior of gastric signet-ring cell carcinoma; an image analysis study with deep learning. Pathol Int. 69:437–439. 2019. View Article : Google Scholar : PubMed/NCBI
88	Leon F, Gelvez M, Jaimes Z, Gelvez T and Arguello H: Supervised Classification of Histopathological Images Using Convolutional Neuronal Networks for Gastric Cancer Detection. STSIVA; 2019, View Article : Google Scholar
89	Wang S, Wang T, Yang L, Yang DM, Fujimoto J, Yi F, Luo X, Yang Y, Yao B, Lin S, et al: ConvPath: A software tool for lung adenocarcinoma digital pathological image analysis aided by a convolutional neural network. EBioMedicine. 50:103–110. 2019. View Article : Google Scholar : PubMed/NCBI
90	Aprupe L, Litjens G, Brinker TJ, van der Laak J and Grabe N: Robust and accurate quantification of biomarkers of immune cells in lung cancer micro-environment using deep convolutional neural networks. PeerJ. 7:e63352019. View Article : Google Scholar : PubMed/NCBI
91	Dihge L, Vallon-Christersson J, Hegardt C, Saal LH, Häkkinen J, Larsson C, Ehinger A, Loman N, Malmberg M, Bendahl PO, et al: Prediction of lymph node metastasis in breast cancer by gene expression and clinicopathological models: Development and validation within a population-based cohort. Clin Cancer Res. 25:6368–6381. 2019. View Article : Google Scholar : PubMed/NCBI
92	Nir G, Davood K, Goldenberg SL, Fazli L, Skinnider BF, Tavassoli P, Turbin D, Villamil CF, Wang G..Thompson DJS, et al: Comparison of artificial intelligence techniques to evaluate performance of a classifier for automatic grading of prostate cancer from digitized histopathologic images. JAMA Netw Open. 2:e1904422019. View Article : Google Scholar : PubMed/NCBI
93	Courtiol P, Maussion C, Moarii M, Pronier E, Pilcer S, Sefta M, Manceron P, Toldo S, Zaslavskiy M, Le Stang N, et al: Deep learning-based classification of mesothelioma improves prediction of patient outcome. Nat Med. 25:1519–1525. 2019. View Article : Google Scholar : PubMed/NCBI
94	Achi HE, Belousova T, Chen L, Wahed A, Wang I, Hu Z, Kanaan Z, Rios A and Nguyen AND: Automated diagnosis of lymphoma with digital pathology images using deep learning. Ann Clin Lab Sci. 49:153–160. 2019.PubMed/NCBI
95	Rodner E, Bocklitz T, von Eggeling F, Ernst G, Chernavskaia O, Popp J, Denzler J and Guntinas-Lichius O: Fully convolutional networks in multimodal nonlinear microscopy images for automated detection of head and neck carcinoma: Pilot study. Head Neck. 41:116–121. 2019.PubMed/NCBI
96	Hu L, Bell D, Antani S, Xue Z, Yu K, Horning MP, Gachuhi N, Wilson B, Jaiswal MS, Befano B, et al: An observational study of deep learning and automated evaluation of cervical images for cancer screening. J Natl Cancer Inst. 111:923–932. 2019. View Article : Google Scholar : PubMed/NCBI
97	Tian R, Cui Z, He D, Tian X, Gao Q, Ma X, Yang JR, Wu J, Das BC, Severinov K, et al: Risk stratification of cervical lesions using capture sequencing and machine learning method based on HPV and human integrated genomic profiles. Carcinogenesis. 40:1220–1228. 2019. View Article : Google Scholar : PubMed/NCBI
98	Newman AM, Steen CB, Liu CL, Gentles AJ, Chaudhuri AA, Scherer F, Khodadoust MS, Esfahani MS, Luca BA, Steiner D, et al: Determining cell type abundance and expression from bulk tissues with digital cytometry. Nat Biotechnol. 37:773–782. 2019. View Article : Google Scholar : PubMed/NCBI
99	Saillard C, Schmauch B, Laifa O, Moarii M, Toldo S, Zaslavskiy M, Pronier E, Laurent A, Amaddeo G, Regnault H, et al: Predicting survival after hepatocellular carcinoma resection using deep learning on histological slides. Hepatology. 72:2000–2013. 2020. View Article : Google Scholar : PubMed/NCBI
100	Dietz C, Rueden CT, Helfrich S, Dobson ETA, Horn M, Eglinger J, Evans EL III, McLean DT, Novitskaya T, Ricke WA, et al: Integration of the ImageJ Ecosystem in the KNIME Analytics Platform. Front Comput Sci. 2:82020. View Article : Google Scholar : PubMed/NCBI
101	Bulten W, Pinckaers H, van Boven H, Vink R, de Bel T, van Ginneken B, van der Laak J, Hulsbergen-van de Kaa C and Litjens G: Automated deep-learning system for Gleason grading of prostate cancer using biopsies: A diagnostic study. Lancet Oncol. 21:233–241. 2020. View Article : Google Scholar : PubMed/NCBI
102	Han W, Johnson C, Gaed M, Gómez JA, Moussa M, Chin JL, Pautler S, Bauman GS and Ward AD: Histologic tissue components provide major cues for machine learning-based prostate cancer detection and grading on prostatectomy specimens. Sci Rep. 10:99112020. View Article : Google Scholar : PubMed/NCBI
103	Jaber MI, Song B, Taylor C, Vaske CJ, Benz SC, Rabizadeh S, Soon-Shiong P and Szeto CW: A deep learning image-based intrinsic molecular subtype classifier of breast tumors reveals tumor heterogeneity that may affect survival. Breast Cancer Res. 22:122020. View Article : Google Scholar : PubMed/NCBI
104	Valieris R, Amaro L, Osório CABT, Bueno AP, Rosales Mitrowsky RA, Carraro DM, Nunes DN, Dias-Neto E and Silva ITD: Deep learning predicts underlying features on pathology images with therapeutic relevance for breast and gastric cancer. Cancers (Basel). 12:36872020. View Article : Google Scholar : PubMed/NCBI
105	Skrede OJ, De Raedt S, Kleppe A, Hveem TS, Liestøl K, Maddison J, Askautrud HA, Pradhan M, Nesheim JA, Albregtsen F, et al: Deep learning for prediction of colorectal cancer outcome: A discovery and validation study. Lancet. 395:350–360. 2020. View Article : Google Scholar : PubMed/NCBI
106	Reichling C, Taieb J, Derangere V, Klopfenstein Q, Le Malicot K, Gornet JM, Becheur H, Fein F, Cojocarasu O, Kaminsky MC, et al: Artificial intelligence-guided tissue analysis combined with immune infiltrate assessment predicts stage III colon cancer outcomes in PETACC08 study. Gut. 69:681–690. 2020. View Article : Google Scholar : PubMed/NCBI
107	Bao H, Sun X, Zhang Y, Pang B, Li H, Zhou L, Wu F, Cao D, Wang J, Turic B, et al: The artificial intelligence-assisted cytology diagnostic system in large-scale cervical cancer screening: A population-based cohort study of 0.7 million women. Cancer Med. 9:6896–6906. 2020. View Article : Google Scholar : PubMed/NCBI
108	Xu-Monette ZY, Zhang H, Zhu F, Tzankov A, Bhagat G, Visco C, Dybkaer K, Chiu A, Tam W, Zu Y, et al: A refined cell-of-origin classifier with targeted NGS and artificial intelligence shows robust predictive value in DLBCL. Blood Adv. 4:3391–3404. 2020. View Article : Google Scholar : PubMed/NCBI
109	Foersch S, Eckstein M, Wagner DC, Gach F, Woerl AC, Geiger J, Glasner C, Schelbert S, Schulz S, Porubsky S, et al: Deep learning for diagnosis and survival prediction in soft tissue sarcoma. Ann Oncol. 32:1178–1187. 2021. View Article : Google Scholar : PubMed/NCBI
110	Lagree A, Shiner A, Alera MA, Fleshner L, Law E, Law B, Lu FI, Dodington D, Gandhi S, Slodkowska EA, et al: Assessment of digital pathology imaging biomarkers associated with breast cancer histologic grade. Curr Oncol. 28:4298–4316. 2021. View Article : Google Scholar : PubMed/NCBI
111	Bychkov D, Linder N, Tiulpin A, Kücükel H, Lundin M, Nordling S, Sihto H, Isola J, Lehtimäki T, Kellokumpu-Lehtinen PL, et al: Deep learning identifies morphological features in breast cancer predictive of cancer ERBB2 status and trastuzumab treatment efficacy. Sci Rep. 11:40372021. View Article : Google Scholar : PubMed/NCBI
112	Li F, Yang Y, Wei Y, He P, Chen J, Zheng Z and Bu H: Deep learning-based predictive biomarker of pathological complete response to neoadjuvant chemotherapy from histological images in breast cancer. J Transl Med. 19:3482021. View Article : Google Scholar : PubMed/NCBI
113	Chakraborty D, Ivan C, Amero P, Khan M, Rodriguez-Aguayo C, Başağaoğlu H and Lopez-Berestein G: Explainable artificial intelligence reveals novel insight into tumor microenvironment conditions linked with better prognosis in patients with breast cancer. Cancers (Basel). 13:34502021. View Article : Google Scholar : PubMed/NCBI
114	Fitzgerald J, Higgins D, Mazo Vargas C, Watson W, Mooney C, Rahman A, Aspell N, Connolly A, Aura Gonzalez C and Gallagher W: Future of biomarker evaluation in the realm of artificial intelligence algorithms: Application in improved therapeutic stratification of patients with breast and prostate cancer. J Clin Pathol. 74:429–434. 2021. View Article : Google Scholar : PubMed/NCBI
115	Ayyad SM, Shehata M, Shalaby A, Abou El-Ghar M, Ghazal M, El-Melegy M, Abdel-Hamid NB, Labib LM, Ali HA and El-Baz A: Role of AI: histopathological images in detecting prostate cancer: A survey. Sensors (Basel). 21:25862021. View Article : Google Scholar : PubMed/NCBI
116	Haggenmüller S, Maron RC, Hekler A, Utikal JS, Barata C, Barnhill RL, Beltraminelli H, Berking C, Betz-Stablein B, Blum A, et al: Skin cancer classification via convolutional neural networks: Systematic review of studies involving human experts. Eur J Cancer. 156:202–216. 2021. View Article : Google Scholar : PubMed/NCBI
117	Kiehl L, Kuntz S, Höhn J, Jutzi T, Krieghoff-Henning E, Kather JN, Holland-Letz T, Kopp-Schneider A, Chang-Claude J, Brobeil A, et al: Deep learning can predict lymph node status directly from histology in colorectal cancer. Eur J Cancer. 157:464–473. 2021. View Article : Google Scholar : PubMed/NCBI
118	Yamashita R, Long J, Longacre T, Peng L, Berry G, Martin B, Higgins J, Rubin DL and Shen J: Deep learning model for the prediction of microsatellite instability in colorectal cancer: A diagnostic study. Lancet Oncol. 22:132–141. 2021. View Article : Google Scholar : PubMed/NCBI
119	Krause J, Grabsch HI, Kloor M, Jendrusch M, Echle A, Buelow RD, Boor P, Luedde T, Brinker TJ, Trautwein C, et al: Deep learning detects genetic alterations in cancer histology generated by adversarial networks. J Pathol. 254:70–79. 2021.PubMed/NCBI
120	Saito A, Toyoda H, Kobayashi M, Koiwa Y, Fujii H, Fujita K, Maeda A, Kaneoka Y, Hazama S, Nagano H, et al: Prediction of early recurrence of hepatocellular carcinoma after resection using digital pathology images assessed by machine learning. Mod Pathol. 34:417–425. 2021. View Article : Google Scholar : PubMed/NCBI
121	Yamashita R, Long J, Saleem A, Rubin DL and Shen J: Deep learning predicts postsurgical recurrence of hepatocellular carcinoma from digital histopathologic images. Sci Rep. 11:20472021. View Article : Google Scholar : PubMed/NCBI
122	Yang H, Chen L, Cheng Z, Yang M, Wang J, Lin C, Wang Y, Huang L, Chen Y, Peng S, et al: Deep learning-based six-type classifier for lung cancer and mimics from histopathological whole slide images: A retrospective study. BMC Med. 19:802021. View Article : Google Scholar : PubMed/NCBI
123	Chen CL, Chen CC, Yu WH, Chen SH, Chang YC, Hsu TI, Hsiao M, Yeh CY and Chen CY: An annotation-free whole-slide training approach to pathological classification of lung cancer types using deep learning. Nat Commun. 12:11932021. View Article : Google Scholar : PubMed/NCBI
124	Park J, Jang BG, Kim YW, Park H, Kim BH, Kim MJ, Ko H, Gwak JM, Lee EJ, Chung YR, et al: A prospective validation and observer performance study of a deep learning algorithm for pathologic diagnosis of gastric tumors in endoscopic biopsies. Clin Cancer Res. 27:719–728. 2021. View Article : Google Scholar : PubMed/NCBI
125	Wang X, Chen Y, Gao Y, Zhang H, Guan Z, Dong Z, Zheng Y, Jiang J, Yang H, Wang L, et al: Predicting gastric cancer outcome from resected lymph node histopathology images using deep learning. Nat Commun. 12:16372021. View Article : Google Scholar : PubMed/NCBI
126	Muti HS, Heij LR, Keller G, Kohlruss M, Langer R, Dislich B, Cheong JH, Kim YW, Kim H, Kook MC, et al: Development and validation of deep learning classifiers to detect Epstein-Barr virus and microsatellite instability status in gastric cancer: A retrospective multicentre cohort study. Lancet Digit Health. 3:e654–e664. 2021. View Article : Google Scholar : PubMed/NCBI
127	Yan J, Zhao Y, Chen Y, Wang W, Duan W, Wang L, Zhang S, Ding T, Liu L, Sun Q, et al: Deep learning features from diffusion tensor imaging improve glioma stratification and identify risk groups with distinct molecular pathway activities. EBioMedicine. 72:1035832021. View Article : Google Scholar : PubMed/NCBI
128	Bao H, Wang Z, Ma X, Guo W, Zhang X, Tang W, Chen X, Wang X, Chen Y, Mo S, et al: Letter to the Editor: An ultra-sensitive assay using cell-free DNA fragmentomics for multi-cancer early detection. Mol Cancer. 21:1292022. View Article : Google Scholar : PubMed/NCBI
129	Duchmann M, Wagner-Ballon O, Boyer T, Cheok M, Fournier E, Guerin E, Fenwarth L, Badaoui B, Freynet N, Benayoun E, et al: Machine learning identifies the independent role of dysplasia in the prediction of response to chemotherapy in AML. Leukemia. 36:656–663. 2022. View Article : Google Scholar : PubMed/NCBI
130	Yen R, Grasedieck S, Wu A, Lin H, Su J, Rothe K, Nakamoto H, Forrest DL, Eaves CJ and Jiang X: Identification of key microRNAs as predictive biomarkers of Nilotinib response in chronic myeloid leukemia: A sub-analysis of the ENESTxtnd clinical trial. Leukemia. 36:2443–2452. 2022. View Article : Google Scholar : PubMed/NCBI
131	Laukhtina E, Schuettfort VM, D'Andrea D, Pradere B, Quhal F, Mori K, Sari Motlagh R, Mostafaei H, Katayama S, Grossmann NC, et al: Selection and evaluation of preoperative systemic inflammatory response biomarkers model prior to cytoreductive nephrectomy using a machine-learning approach. World J Urol. 40:747–754. 2022. View Article : Google Scholar : PubMed/NCBI
132	Cheng N, Ren Y, Zhou J, Zhang Y, Wang D, Zhang X, Chen B, Liu F, Lv J, Cao Q, et al: Deep Learning-based classification of hepatocellular nodular lesions on whole-slide histopathologic images. Gastroenterology. 162:1948–1961. 2022. View Article : Google Scholar : PubMed/NCBI
133	Kleppe A, Skrede OJ, De Raedt S, Hveem TS, Askautrud HA, Jacobsen JE, Church DN, Nesbakken A, Shepherd NA, Novelli M, et al: A clinical decision support system optimising adjuvant chemotherapy for colorectal cancers by integrating deep learning and pathological staging markers: A development and validation study. Lancet Oncol. 23:1221–1232. 2022. View Article : Google Scholar : PubMed/NCBI
134	Byeon SJ, Park J, Cho YA and Cho BJ: Automated histological classification for digital pathology images of colonoscopy specimen via deep learning. Sci Rep. 12:128042022. View Article : Google Scholar : PubMed/NCBI
135	Feng L, Liu Z, Li C, Li Z, Lou X, Shao L, Wang Y, Huang Y, Chen H, Pang X, et al: Development and validation of a radiopathomics model to predict pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: A multicentre observational study. Lancet Digit Health. 4:e8–e17. 2022. View Article : Google Scholar : PubMed/NCBI
136	Ding K, Zhou M, Wang H, Zhang S and Metaxas DN: Spatially aware graph neural networks and cross-level molecular profile prediction in colon cancer histopathology: A retrospective multi-cohort study. Lancet Digit Health. 4:e787–e795. 2022. View Article : Google Scholar : PubMed/NCBI
137	Wang R, Dai W, Gong J, Huang M, Hu T, Li H, Lin K, Tan C, Hu H, Tong T, et al: Development of a novel combined nomogram model integrating deep learning-pathomics, radiomics and immunoscore to predict postoperative outcome of colorectal cancer lung metastasis patients. J Hematol Oncol. 15:112022. View Article : Google Scholar : PubMed/NCBI
138	Sundar R, Barr Kumarakulasinghe N, Huak Chan Y, Yoshida K, Yoshikawa T, Miyagi Y, Rino Y, Masuda M, Guan J, Sakamoto J, et al: Machine-learning model derived gene signature predictive of paclitaxel survival benefit in gastric cancer: Results from the randomised phase III SAMIT trial. Gut. 71:676–685. 2022. View Article : Google Scholar : PubMed/NCBI
139	Pfob A, Sidey-Gibbons C, Rauch G, Thomas B, Schaefgen B, Kuemmel S, Reimer T, Hahn M, Thill M, Blohmer JU, et al: Intelligent Vacuum-assisted biopsy to identify breast cancer patients with pathologic complete response (ypT0 and ypN0) after neoadjuvant systemic treatment for omission of breast and axillary surgery. J Clin Oncol. 40:1903–1915. 2022. View Article : Google Scholar : PubMed/NCBI
140	Prat A, Guarneri V, Pascual T, Brasó-Maristany F, Sanfeliu E, Paré L, Schettini F, Martínez D, Jares P, Griguolo G, et al: Development and validation of the new HER2DX assay for predicting pathological response and survival outcome in early-stage HER2-positive breast cancer. EBioMedicine. 75:1038012022. View Article : Google Scholar : PubMed/NCBI
141	Farinella F, Merone M, Bacco L, Capirchio A, Ciccozzi M and Caligiore D: Machine Learning analysis of high-grade serous ovarian cancer proteomic dataset reveals novel candidate biomarkers. Sci Rep. 12:30412022. View Article : Google Scholar : PubMed/NCBI
142	Park S, Ock CY, Kim H, Pereira S, Park S, Ma M, Choi S, Kim S, Shin S, Aum BJ, et al: Artificial Intelligence-powered spatial analysis of tumor-infiltrating lymphocytes as complementary biomarker for immune checkpoint inhibition in Non-Small-Cell lung cancer. J Clin Oncol. 40:1916–1928. 2022. View Article : Google Scholar : PubMed/NCBI
143	Cheng G, Zhang F, Xing Y, Hu X, Zhang H, Chen S, Li M, Peng C, Ding G, Zhang D, et al: Artificial intelligence-assisted score analysis for predicting the expression of the immunotherapy biomarker PD-L1 in lung cancer. Front Immunol. 13:8931982022. View Article : Google Scholar : PubMed/NCBI
144	Vanguri RS, Luo J, Aukerman AT, Egger JV, Fong CJ, Horvat N, Pagano A, Araujo-Filho JAB, Geneslaw L, Rizvi H, et al: Multimodal integration of radiology, pathology and genomics for prediction of response to PD-(L)1 blockade in patients with non-small cell lung cancer. Nat Cancer. 3:1151–1164. 2022. View Article : Google Scholar : PubMed/NCBI
145	Liu Y, Jia Y, Hou C, Li N, Zhang N, Yan X, Yang L, Guo Y, Chen H, Li J, et al: Pathological prognosis classification of patients with neuroblastoma using computational pathology analysis. Comput Biol Med. 149:1059802022. View Article : Google Scholar : PubMed/NCBI
146	Foersch S, Glasner C, Woerl AC, Eckstein M, Wagner DC, Schulz S, Kellers F, Fernandez A, Tserea K, Kloth M, et al: Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer. Nat Med. 29:430–439. 2023. View Article : Google Scholar : PubMed/NCBI
147	Liu XP, Jin X, Seyed Ahmadian S, Yang X, Tian SF, Cai YX, Chawla K, Snijders AM, Xia Y, van Diest PJ, et al: Clinical significance and molecular annotation of cellular morphometric subtypes in lower-grade gliomas discovered by machine learning. Neuro Oncol. 25:68–81. 2023. View Article : Google Scholar : PubMed/NCBI
148	Cao R, Nelson SD, Davis S, Liang Y, Luo Y, Zhang Y, Crawford B and Wang LV: Label-free intraoperative histology of bone tissue via deep-learning-assisted ultraviolet photoacoustic microscopy. Nat Biomed Eng. 7:124–134. 2023. View Article : Google Scholar : PubMed/NCBI
149	Wu S, Hong G, Xu A, Zeng H, Chen X, Wang Y, Luo Y, Wu P, Liu C, Jiang N, et al: Artificial intelligence-based model for lymph node metastases detection on whole slide images in bladder cancer: A retrospective, multicentre, diagnostic study. Lancet Oncol. 24:360–370. 2023. View Article : Google Scholar : PubMed/NCBI
150	Barnett R: Lung cancer. Lancet. 390:9282017. View Article : Google Scholar : PubMed/NCBI
151	Zheng Y, Gindra RH, Green EJ, Burks EJ, Betke M, Beane JE and Kolachalama VB: A Graph-transformer for whole slide image classification. IEEE Trans Med Imaging. 41:3003–3015. 2022. View Article : Google Scholar : PubMed/NCBI
152	Qiao C, Li D, Liu Y, Zhang S, Liu K, Liu C, Guo Y, Jiang T, Fang C, Li N, et al: Rationalized deep learning super-resolution microscopy for sustained live imaging of rapid subcellular processes. Nat Biotechnol. 41:367–377. 2023. View Article : Google Scholar : PubMed/NCBI
153	Vo TH, Nguyen NTK, Kha QH and Le NQK: On the road to explainable AI in drug-drug interactions prediction: A systematic review. Comput Struct Biotechnol J. 20:2112–2123. 2022. View Article : Google Scholar : PubMed/NCBI
154	Cheng JY, Abel JT, Balis UGJ, McClintock DS and Pantanowitz L: Challenges in the development, deployment, and regulation of artificial intelligence in anatomic pathology. Am J Pathol. 191:1684–1692. 2021. View Article : Google Scholar : PubMed/NCBI
155	Schmidt C: M. D. Anderson Breaks with IBM Watson, Raising Questions about artificial intelligence in oncology. J Natl Cancer Inst. 1092017.doi: 10.1093/jnci/djx113.
156	Lecler A, Duron L and Soyer P: Revolutionizing radiology with GPT-based models: Current applications, future possibilities and limitations of ChatGPT. Diagn Interv Imaging. 104:269–274. 2023. View Article : Google Scholar : PubMed/NCBI
157	Will ChatGPT transform healthcare? Nat Med. 29:505–506. 2023. View Article : Google Scholar : PubMed/NCBI
158	Sinha RK, Deb Roy A, Kumar N and Mondal H: Applicability of ChatGPT in assisting to solve higher order problems in pathology. Cureus. 15:e352372023.PubMed/NCBI
159	Chauhan C and Gullapalli RR: Ethics of AI in pathology: Current paradigms and emerging issues. Am J Pathol. 91:1673–1683. 2021. View Article : Google Scholar : PubMed/NCBI
160	Warnat-Herresthal S, Schultze H, Shastry KL, Manamohan S, Mukherjee S, Garg V, Sarveswara R, Händler K, Pickkers P, Aziz NA, et al: Swarm Learning for decentralized and confidential clinical machine learning. Nature. 594:265–270. 2021. View Article : Google Scholar : PubMed/NCBI

September-2023
Volume 63 Issue 3

Print ISSN: 1019-6439
Online ISSN:1791-2423

Recommend to Library

Journal Metrics
Impact Factor: 4.5
Ranked Oncology
(total number of cites)
CiteScore: 9.6
CiteScore Rank: Q1: #66/404 Oncology
Source Normalized Impact per Paper (SNIP): 0.88
SCImago Journal Rank (SJR): 1.099

Year	Name of the challenge	Tissue	Staining	Aims of the challenge	Provided ground-truth
2012	ISBI2012 https://imagej.net/events/isbi-2012-segmentation-challenge	Neuro	-	Automatic segmentation of neural structures	Binary labels
2013	AMIDA13 https://www.isi.uu.nl/research/challenges/	BC	H&E	Assessment of mitosis detection algorithms (51,52)	-
2014	ISBI2014 https://archives.embs.org/wp-content/archives/ISBI/2014/?page_id=17	Cervix, uterus	PAP	Extracting the boundaries of individual cytoplasm and nuclei from overlapping cervical cytology images	Counting mitosis
2014	ICPR https://mitos-atypia-14.grand-challenge.org	BC	H&E	Mitosis detection, nuclear atypia scoring	Centroids of mitosis, nuclear atypia scoring
2015	GlaS (TIA) https://warwick.ac.uk/fac/sci/dcs/research/tia/glascontest/	Gland	H&E	Gland segmentation	Binary masks
2015	BioImaging2015 http://www.bioimaging2015.ineb.up.pt/challenge_overview.html	BC	H&E	Cancer classification	Labels
2016	TUMAC https://tupac.grand-challenge.org/	BC	H&E	Prediction of tumor proliferation scores	Tumor proliferation scoring, molecular proliferation scoring
2016	HER2 Scoring https://warwick.ac.uk/fac/sci/dcs/research/tia/her2contest	BC	IHC	HER2 scoring	HER2 scoring and a percentage score given by pathologists
2016	CAMELYON'16 https://camelyon16.grand-challenge.org/ (Closed)	BC	H&E	Detection of cancer metastasis (53)	Annotated contours, binary masks
2017	CAMELYON'17 https://camelyon17.grand-challenge.org	BC	H&E	Detection of cancer metastasis	Metastases annotations in WSIs, patient pathological node stage labeling
2017	ISBI 2017 https://biomedicalimaging.org/2017/	TC	H&E, IHC	Tissue microarray analysis in TC diagnosis (54)	-
2018	ICIAR 2018 (BACH) https://iciar2018-challenge.grand-challenge.org	BC	H&E	Classification and pixel-wise labelling of WSIs	Pixel-wise labels
2018	PatchCamelyon https://patchcamelyon.grand-challenge.org	Lymph node	H&E	Metastasis detection	Binary label indicating presence of metastatic tissue
2019	ACDC-LungHP https://acdc-lunghp.grand-challenge.org	LC	H&E	Cancer detection, assessment of tumor proliferation	Annotation of cancer regions
2019	ANHIR https://anhir.grand-challenge.org/Intro/	LC, BC	H&E, IHC	Comparing the accuracy and the speed of automatic non-linear registration methods on a set of large images from the same tissue samples but stained with different biomarkers	-
2019	Digestpath2019 https://digestpath2019.grand-challenge.org/	CRC	H&E	Evaluation of signet ring cells and colonoscopy tissue screening	Number of signet ring cell for each patch and tissue segmentation
2019	LYSTO https://lysto.grand-challenge.org/LYSTO	BC, CRC and PCa	IHC	Assessment of lymphocytes	Number of lympho cytes for each patch
2019	PAIP https://paip2019.grand-challenge.org/Home/	LC	H&E	LC segmentation	Tumor area segmentation, viable tumor area
2019	CodaLab (ISBI 2019) https://competitions.codalab.org/competitions/20395#learn_the_details-overview	Blood	-	Cell classification	Marked by the expert oncologist
2019	LYON https://lyon19.grand-challenge.org/Home/	BC, CRC and PCa	IHC	Lymphocyte detection	-
2019	Gleason https://gleason2019.grand-challenge.org	PCa	H&E	Gleason grading (tissue microarray)	Maps and labels
2019	SPIE-AAPM-NCI BreastPathQ https://breastpathq.grand-challenge.org/	BC	H&E	Determination of cancer cellularity	Tumor rating
2020	ECDP2020 https://ecdp2020.grand-challenge.org/Home/	BC	H&E	Identifying HER2⁺ from HER2	-
2021	BCNB https://bcnb.grand-challenge.org/	BC	H&E	Axillary lymph node metastasis status	Annotation of cancer regions
2021	PANDA https://panda.grand-challenge.org/	PCa	H&E	Gleason grading	Tumor regional score grading
2021	MIDOG 2021 https://midog2021.grand-challenge.org/	Various tumors, including BC	H&E	Mitosis detection	Labels
2021	TIGER https://tiger.grand-challenge.org/	BC	H&E	Automatic quantification of TILs in BC	Tumor region segmentation
2021	BCSS https://bcsegmentation.grand-challenge.org/	BC	-	BC semantic segmentation	Image segmentation
2022	AGGC22 https://aggc22.grand-challenge.org/	PCa	H&E	Gleason grading	Identifying the cellular and glandular patterns for Gleason grading
2022	BCI https://bci.grand-challenge.org/	BC	H&E, IHC	Identify HER2⁺	Peak signal to noise ratio and structural similarity
2022	CoNIC https://conic-challenge.grand-challenge.org/	CRC	-	Nuclear segmentation and classification, prediction of cell composition	-
2022	MIDOG2022 https://midog2022.grand-challenge.org/	-	H&E	Mitosis detection	Labels
2022	WSSS4LUAD https://wsss4luad.grand-challenge.org/	LUAD	H&E	Tissue semantic segmentation	Image annotation
2022	Neurips22 https://neurips22-cellseg.grand-challenge.org/	-	H&E	Cell semantic segmentation	Image segmentation
2022	ENDO-AID https://endo-aid.grand-challenge.org/	Endometrial carcinoma	H&E	Prediction of cancer types	Labels
2023	PAIP2023 https://2023paip.grand-challenge.org/	Colon and pancreatic cancer	H&E	Tumor cellularity prediction	Counting tumor cells

International
Journal of
Oncology