Iterative Explainability for Weakly Supervised Segmentation in Medical PE Detection
This addresses the challenge of limited annotations for AI-based diagnostic assistance in detecting pulmonary embolisms, a leading cause of cardiovascular death, with an incremental improvement over existing weakly supervised methods.
The paper tackles the problem of scarce fine-grained annotations for pulmonary embolism (PE) detection in medical imaging by introducing iExplain, a weakly supervised learning algorithm that iteratively transforms coarse image-level annotations into detailed pixel-level PE masks. The method achieves PE detection performance comparable to strongly supervised methods and outperforms existing weakly supervised methods on the RSPECT dataset.
Pulmonary Embolism (PE) are a leading cause of cardiovascular death. Computed tomographic pulmonary angiography (CTPA) is the gold standard for PE diagnosis, with growing interest in AI-based diagnostic assistance. However, these algorithms are limited by scarce fine-grained annotations of thromboembolic burden. We address this challenge with iExplain, a weakly supervised learning algorithm that transforms coarse image-level annotations into detailed pixel-level PE masks through iterative model explainability. Our approach generates soft segmentation maps used to mask detected regions, enabling the process to repeat and discover additional embolisms that would be missed in a single pass. This iterative refinement effectively captures complete PE regions and detects multiple distinct embolisms. Models trained on these automatically generated annotations achieve excellent PE detection performance, with significant improvements at each iteration. We demonstrate iExplain's effectiveness on the RSPECT augmented dataset, achieving results comparable to strongly supervised methods while outperforming existing weakly supervised methods.