GEN-Guard: Correcting Generalization Failures for Deployable Federated Surgical AI

2026-06-18

AI Summary

This study addresses the critical issue of generalization failure—termed “performance leakage”—in federated surgical AI, where models selected solely based on validation data from participating hospitals often underperform on unseen institutions. The work systematically identifies and quantifies this problem and introduces GEN-Guard, a novel framework that, after standard federated convergence, employs Client-Isolated Evaluation (CBE) to detect generalization failures and applies Disagreement-Aware Distillation (DAD) for feature-level correction without requiring retraining. Evaluated on cholecystectomy phase recognition and colonoscopy polyp segmentation tasks, GEN-Guard improves performance on unseen institutions by up to 3 percentage points and boosts the worst-performing sites by 3–9 points, substantially mitigating the over 80% model selection failure rate previously observed.

Abstract

Federated Learning (FL) in surgical video AI enables collaborative model training without sharing sensitive data. However, standard evaluation practices - selecting the "best" global model based only on validation data from participating hospitals - can lead to suboptimal deployment choices. We identify this critical failure mode as performance leakage, where the selected model overfits internal federation data and fails to generalize to unseen institutions. We propose GEN-Guard, a practical post-hoc framework to detect and correct generalization failures in federated surgical AI. It integrates Generalization Detection via Client-Blocked Evaluation (CBE), which validates performance on isolated client distributions to prevent performance leakage, and Generalization Correction through Disagreement-Aware Distillation (DAD), which learns adaptive feature-level corrections for cross-institutional robustness. Both components operate after standard FL convergence while providing robust support for zero-shot adaptation to unseen environments. We first quantify the severity of performance leakage, observing Model Selection Failures (MSFs) exceeding 80% under standard evaluation. GEN-Guard is evaluated on two multi-center clinical challenges: surgical phase recognition in laparoscopic cholecystectomy and polyp segmentation in colonoscopy. Across both datasets, GEN-Guard consistently corrects these failures, improving in-federation F1 scores by up to 2 points, unseen-institution performance by up to 3 points, and worst-case institutional performance by 3-9 points. Performance leakage represents a systematic and previously under-recognized risk in federated surgical AI. GEN-Guard provides a practical solution for detecting and correcting such failures. By improving cross-institutional robustness and zero-shot generalization, it strengthens the reliability of FL for real-world surgical deployment.

Problem

• federated learning
• generalization failure
• performance leakage
• surgical AI
• model selection

Innovation

• Performance Leakage
• Client-Blocked Evaluation
• Disagreement-Aware Distillation
• Federated Surgical AI
• Zero-Shot Generalization