Statistical Management of the False Discovery Rate in Medical Instance Segmentation Based on Conformal Risk Control
This addresses the problem of misdiagnosis risk in medical image analysis for clinicians and patients, though it represents an incremental improvement by applying conformal prediction to an existing bottleneck.
The paper tackles confidence calibration issues in medical instance segmentation by proposing a conformal prediction-based framework that adaptively adjusts segmentation thresholds to control false discovery rates. The method ensures expected FDR remains below a user-defined risk level α with high probability while maintaining compatibility with existing models and datasets.
Instance segmentation plays a pivotal role in medical image analysis by enabling precise localization and delineation of lesions, tumors, and anatomical structures. Although deep learning models such as Mask R-CNN and BlendMask have achieved remarkable progress, their application in high-risk medical scenarios remains constrained by confidence calibration issues, which may lead to misdiagnosis. To address this challenge, we propose a robust quality control framework based on conformal prediction theory. This framework innovatively constructs a risk-aware dynamic threshold mechanism that adaptively adjusts segmentation decision boundaries according to clinical requirements.Specifically, we design a \textbf{calibration-aware loss function} that dynamically tunes the segmentation threshold based on a user-defined risk level $α$. Utilizing exchangeable calibration data, this method ensures that the expected FNR or FDR on test data remains below $α$ with high probability. The framework maintains compatibility with mainstream segmentation models (e.g., Mask R-CNN, BlendMask+ResNet-50-FPN) and datasets (PASCAL VOC format) without requiring architectural modifications. Empirical results demonstrate that we rigorously bound the FDR metric marginally over the test set via our developed calibration framework.