• 1. The First Clinical Medical College of Lanzhou University, Lanzhou 730000, P. R. China;
  • 2. Department of Gastrointestinal Surgery, Hernia and Abdominal Wall Surgery, The First Hospital of Lanzhou University, Lanzhou 730000, P. R. China;
YU Yongjiang, Email: ylongy@163.com
Export PDF Favorites Scan Get Citation

Objective To systematically summarize the research progress in risk prediction models for postoperative anastomotic leakage in gastric cancer, and to explore the advantages and limitations of models constructed using traditional statistical methods and machine learning, thereby providing a theoretical basis for clinical precision prediction and early intervention. Method By analyzing domestic and international literature, the construction strategies of logistic regression, least absolute shrinkage and selection operator (LASSO) regression, and machine learning models (support vector machine, random forest, deep learning) were systematically reviewed, and their predictive performance and clinical applicability were compared. Results The traditional logistic regression and LASSO regression models performed excellently in terms of interpretability and in small-sample scenarios but were limited by linear assumptions. The machine learning models significantly enhanced predictive capabilities for complex data through non-linear modeling and automatic feature extraction, but required larger data scales and had higher demands for interpretability. Conclusions Different prediction models have their own advantages and limitations; in practical clinical applications, they should be flexibly selected or complementarily applied based on specific scenarios. Current anastomotic leakage prediction models are evolving from single factor analysis to multi-modal dynamic integration. Future efforts should combine artificial intelligence and multi-center prospective clinical studies to validate, so advancing the development of precise and individualized anastomotic leakage predictive tools for patients after gastric cancer resection.

Citation: DU Nan, MA Wenxiang, XIE Jiajing, YANG Yang, YU Yongjiang. Research progress on risk prediction model of anastomotic leakage after gastric cancer resection. CHINESE JOURNAL OF BASES AND CLINICS IN GENERAL SURGERY, 2025, 32(7): 916-924. doi: 10.7507/1007-9424.202504071 Copy

Copyright © the editorial department of CHINESE JOURNAL OF BASES AND CLINICS IN GENERAL SURGERY of West China Medical Publisher. All rights reserved