What question did this study set out to answer?

This study aims to assess the feasibility of using machine learning to quantitatively analyze ICG perfusion metrics for predicting hypoperfusion during colorectal cancer surgery.

May 28, 2026European Journal of Surgical Oncology0 citationsOpen Access

Interpretable and Non-Linear Machine Learning Models for Quantitative ICG Perfusion Assessment in Colorectal Cancer Anastomosis: A Feasibility Study

PDPaulina Daniluk-MarsyGdańsk Medical University JCJohannes CasteleinUniversity Medical Center Groningen KPKarol PołomSpołeczna Akademia Nauk

Key Points

This study aims to assess the feasibility of using machine learning to quantitatively analyze ICG perfusion metrics for predicting hypoperfusion during colorectal cancer surgery.
Single-centre retrospective feasibility study involving 81 patients undergoing colorectal cancer resection.
Intraoperative fluorescence videos processed to extract four perfusion metrics and combined with eight clinical variables.
Training of logistic regression and three non-linear algorithms (random forest, XGBoost, support vector machine) evaluated using 5-fold cross-validation.
Elastic-net logistic regression with perfusion metrics achieved a ROC-AUC of 0.76 (95% CI 0.62–0.88).
Random forest using clinical variables alone yielded a ROC-AUC of 0.77 (95% CI 0.59–0.93).
The combined model reached the highest ROC-AUC of 0.81 (95% CI 0.63–0.94), with SHAP analysis identifying latency and PIR as strong hypoperfusion predictors.

Abstract

AbstractBackground Anastomotic leakage (AL) remains a major complication after colorectal cancer surgery (CCS). Intraoperative indocyanine green (ICG) fluorescence angiography is widely used to assess bowel perfusion, but interpretation is largely subjective. Quantitative analysis of fluorescence kinetics combined with machine learning (ML) may enable more objective, physiology-based risk stratification. Materials and Methods In this single-centre retrospective feasibility study, 81 patients undergoing colorectal cancer resection with ICG perfusion assessment were analysed. Intraoperative fluorescence videos were post-processed to extract four quantitative perfusion metrics. Eight clinical variables with plausible relevance to microvascular health were included. Intraoperative hypoperfusion (n=12, 15%) was defined by delayed fluorescence propagation judged by the operating surgeon. Logistic regression with elastic-net regularisation and three non-linear algorithms (random forest, XGBoost, support vector machine) were trained using perfusion metrics-only, clinical variables-only, and combined feature sets. Performance was evaluated with repeated stratified 5-fold cross-validation and bootstrap confidence intervals. Results Using perfusion-only features, elastic-net logistic regression achieved a ROC-AUC of 0.76 (95% CI 0.62–0.88). Clinical-only models reached a ROC-AUC of 0.77 (95% CI 0.59–0.93) with random forest. The combined model (perfusion + clinical) yielded the highest discrimination (ROC-AUC 0.81, 95% CI 0.63–0.94) with good calibration (Brier score 0.14). SHAP analysis identified increased latency and lower plateau intensity ratio (PIR) as the strongest predictors of hypoperfusion. Conclusion Quantitative ICG perfusion indices, particularly latency and PIR, combined with interpretable ML models, can predict surgeon-assessed intraoperative hypoperfusion during colorectal cancer surgery with good discrimination. These findings support further development of explainable, perfusion-guided decision-support tools to reduce anastomotic risk.

AIに質問

Bookmark

View Full Paper