Safety
Risk-Calibrated Learning: Minimizing Fatal Errors in Medical AI
arXiv:2604.12693v1 Announce Type: new Abstract: Deep learning models often achieve expert-level accuracy in medical image classification but suffer from a critical flaw: semantic incoherence. These hi
arXiv:2604.12693v1 Announce Type: new Abstract: Deep learning models often achieve expert-level accuracy in medical image classification but suffer from a critical flaw: semantic incoherence. These high-confidence mistakes that are semantically incoherent (e.g., classifying a malignant tumor as benign) fundamentally differ from acceptable errors which stem from visual ambiguity. Unlike safe, fine-grained disagreements, these fatal failures erode clinical trust. To address this, we propose Risk-Calibrated Learning, a technique that explicitly distinguishes between visual ambiguity (fine-grained errors) and catastrophic structural errors. By embedding a confusion-aware clinical severity matrix M into the optimization landscape, our method suppresses critical errors (false negatives) without requiring complex architectural changes. We validate our approach in four different imaging modalities: Brain Tumor MRI, ISIC 2018 (Dermoscopy), BreaKHis (Breast Histopathology), and SICAPv2 (Prostate Histopathology). Extensive experiments demonstrate that our Risk-Calibrated Loss consistently reduces the Critical Error Rate (CER) for all four datasets, achieving relative safety improvements ranging from 20.0% (on breast histopathology) to 92.4% (on prostate histopathology) compared to state-of-the-art baselines such as Focal Loss. These results confirm that our method offers a superior safety-accuracy trade-off across both CNN and Transformer architectures.
Related
- Uncertainty-Aware Image Classification In Biomedical Imaging Using Spectral-normalized Neural Gaussian Processes
- MedVR: Annotation-Free Medical Visual Reasoning via Agentic Reinforcement Learning
- Towards Responsible Multimodal Medical Reasoning via Context-Aligned Vision-Language Models
- Event-Level Detection of Surgical Instrument Handovers in Videos with Interpretable Vision Models
- UniSemAlign: Text-Prototype Alignment with a Foundation Encoder for Semi-Supervised Histopathology Segmentation
Source: arXiv cs.CV | 2026-04-15