search button
Open Access

Development and application of an early warning model for predicting early mortality following stent placement in malignant biliary obstruction: A comparative analysis of logistic regression and artificial neural network approaches

  • Authors:
    • Yongxin Ma
    • Jiaojiao Qi
    • Xusheng Zhang
    • Kejun Liu
    • Yimin Liu
    • Xuehai Yu
    • Yang Bu
    • Bendong Chen

  • View Affiliations

  • Published online on: March 20, 2025     https://doi.org/10.3892/ol.2025.14983
  • Article Number: 237

  • Copyright: © Ma et al. This is an open access article distributed under the terms of Creative Commons Attribution License.

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


Abstract

Patients with malignant biliary obstruction (MBO) are often treated with endoscopic retrograde cholangiopancreatography (ERCP) combined with biliary stent placement for tumor progression. However, certain patients die within 30 days after the procedure, increasing healthcare resource consumption and patient burden. Therefore, the development of early mortality prediction models is important for optimizing treatment decisions. The present study retrospectively analyzed the clinical data of 285 patients with MBO, including demographic information, laboratory indicators and tumor‑related factors. Logistic regression and artificial neural network (ANN) models were used to construct a prediction tool, and the model performance was evaluated using area under the curve (AUC), accuracy, sensitivity and specificity. The logistic regression model, which identified the cancer antigen 19‑9 (CA19‑9) level and a history of previous ERCP surgery as independent risk factors, had an AUC of 0.727 and an accuracy of 65.0%. The ANN model, which combined five variables, namely CA19‑9, history of previous ERCP surgery, neutrophil‑lymphocyte ratio (NLR), liver metastasis and carcinoembryonic antigen, demonstrated that NLR was the most weighted predictor. Furthermore, the ANN model had an AUC of 0.813, an accuracy of 88.2% and a specificity that was markedly higher than that of the logistic regression model (95.5 vs. 83.3%). However, the ANN model was revealed to be slightly less sensitive compared with the logistic regression model (61.1 vs. 61.2%). In conclusion, compared with logistic regression, the ANN model had a greater performance level in terms of predictive power and specificity, and is suitable for capturing complex non‑linear relationships. However, its complexity and risk of overfitting need to be further optimized. The present study provides a new tool for the accurate prediction of the risk of early death after ERCP in patients with MBO, which could help improve individualized treatment strategies.
View Figures
View References

Related Articles

Journal Cover

May-2025
Volume 29 Issue 5

Print ISSN: 1792-1074
Online ISSN:1792-1082

Sign up for eToc alerts

Recommend to Library

  • Journal Metrics
  • Impact Factor: 2.5
  • Ranked Oncology
    (total number of cites)
  • CiteScore: 5.7
  • CiteScore Rank: Q2: #139/404 Oncology
  • Source Normalized Impact per Paper (SNIP): 0.574
  • SCImago Journal Rank (SJR): 0.644
International Congress on Molecular Medicine Conference Banner
International Congress on Molecular Medicine Conference Banner
The Cancer Story
World Congress on Advances in Oncology and International Symposium on Molecular Medicine Conference Banner
World Academy Of Sciences Banner
Heyday Road Banner

0

Copy and paste a formatted citation
x
Spandidos Publications style
Ma Y, Qi J, Zhang X, Liu K, Liu Y, Yu X, Bu Y and Chen B: Development and application of an early warning model for predicting early mortality following stent placement in malignant biliary obstruction: A comparative analysis of logistic regression and artificial neural network approaches. Oncol Lett 29: 237, 2025.
APA
Ma, Y., Qi, J., Zhang, X., Liu, K., Liu, Y., Yu, X. ... Chen, B. (2025). Development and application of an early warning model for predicting early mortality following stent placement in malignant biliary obstruction: A comparative analysis of logistic regression and artificial neural network approaches. Oncology Letters, 29, 237. https://doi.org/10.3892/ol.2025.14983
MLA
Ma, Y., Qi, J., Zhang, X., Liu, K., Liu, Y., Yu, X., Bu, Y., Chen, B."Development and application of an early warning model for predicting early mortality following stent placement in malignant biliary obstruction: A comparative analysis of logistic regression and artificial neural network approaches". Oncology Letters 29.5 (2025): 237.
Chicago
Ma, Y., Qi, J., Zhang, X., Liu, K., Liu, Y., Yu, X., Bu, Y., Chen, B."Development and application of an early warning model for predicting early mortality following stent placement in malignant biliary obstruction: A comparative analysis of logistic regression and artificial neural network approaches". Oncology Letters 29, no. 5 (2025): 237. https://doi.org/10.3892/ol.2025.14983