Machine learning‑based radiomics models accurately predict Crohn's disease‑related anorectal cancer

Horio,Yuki; Ikeda,Jota; Matsumoto,Kentaro; Okada,Shinichiro; Nagano,Kentaro; Kusunoki,Kurando; Kuwahara,Ryuichi; Kimura,Kei; Kataoka,Kozo; Beppu,Naohito; Uchino,Motoi; Ikeda,Masataka; Okadome,Takeshi; Yamakado,Koichiro; Ikeuchi,Hiroki

doi:10.3892/ol.2024.14553

Machine learning‑based radiomics models accurately predict Crohn's disease‑related anorectal cancer

Authors:
- Yuki Horio
- Jota Ikeda
- Kentaro Matsumoto
- Shinichiro Okada
- Kentaro Nagano
- Kurando Kusunoki
- Ryuichi Kuwahara
- Kei Kimura
- Kozo Kataoka
- Naohito Beppu
- Motoi Uchino
- Masataka Ikeda
- Takeshi Okadome
- Koichiro Yamakado
- Hiroki Ikeuchi
View Affiliations

Affiliations: Department of Gastroenterological Surgery, Hyogo Medical University, Nishinomiya, Hyogo 663‑8501, Japan, Department of Radiology, Hyogo Medical University, Nishinomiya, Hyogo 663‑8501, Japan, Department of Science and Engineering, Kwansei Gakuin University, Sanda, Hyogo 669‑1330, Japan
Published online on: July 3, 2024 https://doi.org/10.3892/ol.2024.14553
Article Number: 421

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

The radiological diagnosis of Crohn's disease (CD)‑related anorectal cancer is difficult; it is often found in advanced stages and has a poor prognosis because of the difficulty of curative surgery. However, there are no studies on predicting the diagnosis of CD‑related cancer. The present study aimed to develop a predictive model to diagnose CD cancerous lesions more accurately in a way that can be interpreted by clinicians. Patients with CD who developed anorectal CD lesions at Hyogo Medical University (Nishinomiya, Japan) between March 2009 and June 2022 were included in the present study. T2‑weighted and T1‑weighted magnetic resonance (MR) images were utilized for our analysis. Images of anorectal lesions were segmented using open‑source 3D Slicer software, and radiomic features were extracted using PyRadiomics. Six machine learning models were investigated and compared: i) Support vector machine; ii) naive Bayes; iii) random forest; iv) light gradient boosting machine; v) extremely randomized trees; vi) and regularized greedy forest (RGF). SHapley Additive exPlanations (SHAP) values were calculated to assess the extent to which each radiomic feature contributed to the model's predictions compared to baseline, represented as the average of the model's predictions for all test data. The T2‑weighted images of 28 patients with anorectal cancer and 40 non‑cancer patients were analyzed and the contrast‑enhanced T1‑weighted images of 22 cancer and 40 non‑cancer patients. The model with the highest area under the curve (AUC) was the RGF‑based model constructed using T2‑weighted image features, achieving an AUC of 0.944 (accuracy, 0.862; recall, 0.830). The SHAP‑based model explanation suggested a strong association between the diagnosis of CD‑related anorectal cancer and features such as complex lesion texture; greater pixel separation within the same coronal cross‑section; larger, randomly distributed clumps of pixels with the same signal intensity; and a more spherical lesion shape on T2‑weighted images. The MRI radiomics‑based RGF model demonstrated outstanding performance in predicting CD‑related anorectal cancer. These results may affect the diagnosis and surveillance strategies of CD‑related colorectal cancer.

View Figures

View References

1	Higashi D, Katsuno H, Kimura H, Takahashi K, Ikeuchi H, Kono T, Nezu R, Hatakeyama K, Kameyama H, Sasaki I, et al: Current state of and problems related to cancer of the intestinal tract associated with Crohn's disease in Japan. Anticancer Res. 36:3761–3766. 2016.PubMed/NCBI
2	Sasaki H, Ikeuchi H, Bando T, Hirose K, Hirata A, Chohno T, Horio Y, Tomita N, Hirota S, Ide Y, et al: Clinicopathological characteristics of cancer associated with Crohn's disease. Surg Today. 47:35–41. 2017. View Article : Google Scholar : PubMed/NCBI
3	Kim J, Lee HS, Park SH, Yang SK, Ye BD, Yang DH, Kim KJ, Byeon JS, Yoon YS, Yu CS and Kim J: Pathologic features of colorectal carcinomas associated with Crohn's disease in Korean population. Pathol Res Pract. 213:250–255. 2017. View Article : Google Scholar : PubMed/NCBI
4	Uchino M, Ikeuchi H, Hata K, Minagawa T, Horio Y, Kuwahara R, Nakamura S, Watanabe K, Saruta M, Fujii T, et al: Intestinal cancer in patients with Crohn's disease: A systematic review and meta-analysis. J Gastroenterol Hepatol. 36:329–336. 2021. View Article : Google Scholar : PubMed/NCBI
5	Yano Y, Matsui T, Hirai F, Okado Y, Sato Y, Tsurumi K, Ishikawa S, Beppu T, Koga A, Yoshizawa N, et al: Cancer risk in Japanese Crohn's disease patients: Investigation of the standardized incidence ratio. J Gastroenterol Hepatol. 28:1300–1305. 2013. View Article : Google Scholar : PubMed/NCBI
6	Thomas M, Bienkowski R, Vandermeer TJ, Trostle D and Cagir B: Malignant transformation in perianal fistulas of Crohn's disease: A systematic review of literature. J Gastrointest Surg. 14:66–73. 2010. View Article : Google Scholar : PubMed/NCBI
7	Hirsch D, Wangsa D, Zhu YJ, Hu Y, Edelman DC, Meltzer PS, Heselmeyer-Haddad K, Ott C, Kienle P, Galata C, et al: Dynamics of genome alterations in Crohn's disease-associated colorectal carcinogenesis. Clin Cancer Res. 24:4997–5011. 2018. View Article : Google Scholar : PubMed/NCBI
8	Palmieri C, Müller G, Kroesen AJ, Galata C, Rink AD, Morgenstern J and Kruis W: Perianal fistula-associated carcinoma in Crohn's disease: A multicentre retrospective case control study. J Crohns Colitis. 15:1686–1693. 2021. View Article : Google Scholar : PubMed/NCBI
9	Shwaartz C, Munger JA, Deliz JR, Bornstein JE, Gorfine SR, Chessin DB, Popowich DA and Bauer JJ: Fistula-associated anorectal cancer in the setting of Crohn's disease. Dis Colon Rectum. 59:1168–1173. 2016. View Article : Google Scholar : PubMed/NCBI
10	Beaugerie L, Carrat F, Nahon S, Zeitoun JD, Sabaté JM, Peyrin-Biroulet L, Colombel JF, Allez M, Fléjou JF, Kirchgesner J, et al: High risk of anal and rectal cancer in patients with anal and/or perianal Crohn's disease. Clin Gastroenterol Hepatol. 16:892–899.e2. 2018. View Article : Google Scholar : PubMed/NCBI
11	Galata C, Hirsch D, Reindl W, Post S, Kienle P, Boutros M, Gaiser T and Horisberger K: Clinical and histopathologic features of colorectal adenocarcinoma in Crohn's disease. J Clin Gastroenterol. 52:635–640. 2018. View Article : Google Scholar : PubMed/NCBI
12	Ogino T, Mizushima T, Fujii M, Sekido Y, Eguchi H, Nezu R, Ikeuchi H, Motoi U, Futami K, Okamoto K, et al: Crohn's disease-associated anorectal cancer has a poor prognosis with high local recurrence: A subanalysis of the nationwide Japanese study. Am J Gastroenterol. 118:1626–1637. 2023. View Article : Google Scholar : PubMed/NCBI
13	Hirano Y, Futami K, Higashi D, Mikami K and Maekawa T: Anorectal cancer surveillance in Crohn's disease. J Anus Rectum Colon. 2:145–154. 2018. View Article : Google Scholar : PubMed/NCBI
14	Horvat N, Carlos Tavares Rocha C, Clemente Oliveira B, Petkovska I and Gollub MJ: MRI of rectal cancer: Tumor staging, imaging techniques, and management. Radiographics. 39:367–387. 2019. View Article : Google Scholar : PubMed/NCBI
15	Lad SV, Haider MA, Brown CJ and Mcleod RS: MRI appearance of perianal carcinoma in Crohn's disease. J Magn Reson Imaging. 26:1659–1662. 2007. View Article : Google Scholar : PubMed/NCBI
16	Devon KM, Brown CJ, Burnstein M and McLeod RS: Cancer of the anus complicating perianal Crohn's disease. Dis Colon Rectum. 52:211–216. 2009. View Article : Google Scholar : PubMed/NCBI
17	Xu Y, Liu X, Cao X, Huang C, Liu E, Qian S, Liu X, Wu Y, Dong F, Qiu CW, et al: Artificial intelligence: A powerful paradigm for scientific research. Innovation (Camb). 2:1001792021.PubMed/NCBI
18	Lundberg SM and Lee SI: A unified approach to interpreting model predictions. Adv Neural Inf Process Syst. 30:4765–4774. 2017.PubMed/NCBI
19	Shapley LS: A value for n-person games. Contributions to the Theory of Games. 2:307–317. 1953.
20	Rodríguez-Pérez R and Bajorath J: Interpretation of compound activity predictions from complex machine learning models using local approximations and shapley values. J Med Chem. 63:8761–8777. 2020. View Article : Google Scholar : PubMed/NCBI
21	Li R, Shinde A, Liu A, Glaser S, Lyou Y, Yuh B, Wong J and Amini A: Machine learning-based interpretation and visualization of nonlinear interactions in prostate cancer survival. JCO Clin Cancer Inform. 4:637–646. 2020. View Article : Google Scholar : PubMed/NCBI
22	Li W, Liu Y, Liu W, Tang ZR, Dong S, Li W, Zhang K, Xu C, Hu Z, Wang H, et al: Machine learning-based prediction of lymph node metastasis among osteosarcoma patients. Front Oncol. 12:7971032022. View Article : Google Scholar : PubMed/NCBI
23	Li W, Dong S, Wang H, Wu R, Wu H, Tang ZR, Zhang J, Hu Z and Yin C: Risk analysis of pulmonary metastasis of chondrosarcoma by establishing and validating a new clinical prediction model: A clinical study based on SEER database. BMC Musculoskelet Disord. 22:5292021. View Article : Google Scholar : PubMed/NCBI
24	Kora P, Ooi CP, Faust O, Raghavendra U, Gudigar A, Chan WY, Meenakshi K, Swaraja K, Plawiak P and Acharya UR: Transfer learning techniques for medical image analysis: A review. Biocybern Biomed Eng. 42:79–107. 2022. View Article : Google Scholar : PubMed/NCBI
25	Singh D, Kumar V and Vaishali Kaur M: Classification of COVID-19 patients from chest CT images using multi-objective differential evolution-based convolutional neural networks. Eur J Clin Microbiol Infect Dis. 39:1379–1389. 2020. View Article : Google Scholar : PubMed/NCBI
26	Kwon H, Park J and Lee Y: Stacking ensemble technique for classifying breast cancer. Healthc Inform Res. 25:283–288. 2019. View Article : Google Scholar : PubMed/NCBI
27	Zwanenburg A, Vallières M, Abdalah MA, Aerts HJWL, Andrearczyk V, Apte A, Ashrafinia S, Bakas S, Beukinga RJ, Boellaard R, et al: The image biomarker standardization initiative: Standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology. 295:328–338. 2020. View Article : Google Scholar : PubMed/NCBI
28	Wibmer A, Hricak H, Gondo T, Matsumoto K, Veeraraghavan H, Fehr D, Zheng J, Goldman D, Moskowitz C, Fine SW, et al: Haralick texture analysis of prostate MRI: Utility for differentiating non-cancerous prostate from prostate cancer and differentiating prostate cancers with different Gleason scores. Eur Radiol. 25:2840–2850. 2015. View Article : Google Scholar : PubMed/NCBI
29	Liang C, Huang Y, He L, Chen X, Ma Z, Dong D, Tian J, Liang C and Liu Z: The development and validation of a CT-based radiomics signature for the preoperative discrimination of stage I–II and stage III–IV colorectal cancer. Oncotarget. 7:31401–31412. 2016. View Article : Google Scholar : PubMed/NCBI
30	Matsuno H, Mizushima T, Nezu R, Nakajima K, Takahashi H, Haraguchi N, Nishimura J, Hata T, Yamamoto H, Doki Y and Mori M: Detection of anorectal cancer among patients with Crohn's disease undergoing surveillance with various biopsy methods. Digestion. 94:24–29. 2016. View Article : Google Scholar : PubMed/NCBI
31	Ky A, Sohn N, Weinstein MA and Korelitz BI: Carcinoma arising in anorectal fistulas of Crohn's disease. Dis Colon Rectum. 41:992–996. 1998. View Article : Google Scholar : PubMed/NCBI
32	Park YW, Eom J, Kim D, Ahn SS, Kim EH, Kang SG, Chang JH, Kim SH and Lee SK: Correction to: A fully automatic multiparametric radiomics model for differentiation of adult pilocytic astrocytomas from high-grade gliomas. Eur Radiol. 32:57842022. View Article : Google Scholar : PubMed/NCBI
33	Du R, Lee VH, Yuan H, Lam KO, Pang HH, Chen Y, Lam EY, Khong PL, Lee AW, Kwong DL and Vardhanabhuti V: Radiomics model to predict early progression of nonmetastatic nasopharyngeal carcinoma after intensity modulation radiation therapy: A multicenter study. Radiol Artif Intell. 1:e1800752019. View Article : Google Scholar : PubMed/NCBI
34	Wang Y, Lang J, Zuo JZ, Dong Y, Hu Z, Xu X, Zhang Y, Wang Q, Yang L, Wong STC, et al: The radiomic-clinical model using the SHAP method for assessing the treatment response of whole-brain radiotherapy: a multicentric study. Eur Radiol. 32:8737–8747. 2022. View Article : Google Scholar : PubMed/NCBI
35	LeCun Y, Bengio Y and Hinton G: Deep learning. Nature. 521:436–444. 2015. View Article : Google Scholar : PubMed/NCBI
36	Hussein S, Kandel P, Bolan CW, Wallace MB and Bagci U: Lung and pancreatic tumor characterization in the deep learning era: Novel supervised and unsupervised learning approaches. IEEE Trans Med Imaging. 38:1777–1787. 2019. View Article : Google Scholar : PubMed/NCBI
37	Hirsch D and Gaiser T: Crohn's disease-associated colorectal carcinogenesis: TP53 mutations and copy number gains of chromosome arm 5p as (early) markers of tumor progression. Pathologe. 39 (Suppl 2):S253–S261. 2018.(In German). View Article : Google Scholar
38	Fujita M, Matsubara N, Matsuda I, Maejima K, Oosawa A, Yamano T, Fujimoto A, Furuta M, Nakano K, Oku-Sasaki A, et al: Genomic landscape of colitis-associated cancer indicates the impact of chronic inflammation and its stratification by mutations in the Wnt signaling. Oncotarget. 9:969–981. 2017. View Article : Google Scholar : PubMed/NCBI
39	Ronneberger O, Fischer P and Brox T: U-net: Convolutional networks for biomedical image segmentation. Medical image computing and computer-assisted intervention-MICCAI 2015: 18th international conference, Munich, Germany, October 5–9, 2015, proceedings, part III 18. Springer; Cham: pp. 234–241. 2015