Medical AI Encodes a"Feeling of Error": Verifying Cancer Segmentation via Internal Concepts

2026-09-08

AI Summary

研究通过内部概念检测癌症分割模型的错误，使用稀疏自编码器分解神经激活以识别失败案例的独特特征，从而准确预测模型错误。

Abstract

Cancer segmentation models can fail silently, generating plausible but incorrect masks that risk missed findings or unnecessary biopsies. A critical question arises: Do AI models"know"when they are wrong, and if so, can we use the signal to predict their own failures? Humans do have a"Feeling of Error"(FOE): a spontaneous sense of unease that flags a potential error during thinking. We investigate whether cancer segmentation models exhibit an analogous internal signal. Unlike output-level cues (e.g., prediction confidence or uncertainty), which offer no insight into why a failure occurs and suffer from a sensitivity-quality tradeoff where high detection sensitivity could degrade overall segmentation quality. We instead propose to capture the model's FOE from its inner workings. Using mechanistic interpretability tools, specifically Sparse Autoencoders, we decompose internal neural activations into a dictionary of human-interpretable concepts and show that failure cases exhibit a distinct latent signature: fewer active concepts with lower activation magnitudes compared to successful segmentation. By training a classifier on these concept activations, we achieve accurate failure detection along with explanations for the model's mistakes. Experiments on prostate, pancreatic, and brain cancer segmentation demonstrate that our approach outperforms output-based methods in failure detection while preserving segmentation quality.

Problem

• Cancer Segmentation
• Feeling of Error
• Model Failure

Innovation

• Feeling of Error
• Sparse Autoencoders
• Mechanistic Interpretability
• Cancer Segmentation
• Failure Detection