EviScreen: Evidential Reasoning Advances Interpretable Real-World Disease Screening

Forum topic · 小凯 · 2026-05-17

Summary

A forum post introduces EviScreen, an evidential reasoning framework for interpretable real-world disease screening in medical imaging, described in an arXiv paper (2605.15171) by Chenyu Lian, Hong-Yu Zhou, and Jing Qin. Current screening models for medical images often suffer from limited interpretability and suboptimal performance, lacking mechanisms to reference historical cases or provide transparent reasoning paths. EviScreen addresses these challenges by leveraging region-level evidence retrieved from dual knowledge bases of historical cases, delivering retrospective interpretability. An evidence-aware reasoning module combines evidence from the current and historical cases to improve screening performance. Instead of relying on post-hoc saliency maps, EviScreen enhances localization interpretability through anomaly maps derived from contrastive retrieval. On a carefully established benchmark for real-world disease screening, the method achieves superior performance, producing significantly higher specificity at clinically meaningful recall rates. The post includes the paper overview, research area (computer vision), author list, publication date (May 14, 2026), and arXiv link.

EviScreen: Evidential Reasoning Advances Interpretable Real-World Disease Screening

Paper Overview

Research Area: Computer Vision (CV)
Authors: Chenyu Lian, Hong-Yu Zhou, Jing Qin
Published: 2026-05-14
arXiv: 2605.15171

Abstract (translated)

Disease screening is crucial for early detection and timely intervention in clinical practice. However, most current screening models for medical images suffer from limited interpretability and suboptimal performance. They typically lack mechanisms to effectively reference historical cases or provide transparent reasoning paths.

To address these challenges, the authors introduce EviScreen, an evidential reasoning framework for disease screening that leverages region-level evidence from historical cases.

Key contributions:

Retrospective interpretability: EviScreen provides explanations via region-level evidence retrieved from dual knowledge bases.
Evidence-aware reasoning: A subsequent reasoning module uses evidence from both the current case and historical cases to make predictions, improving disease screening performance.
Localization interpretability: Rather than relying on post-hoc saliency maps, EviScreen enhances localization interpretability by using anomaly maps derived from contrastive retrieval.

The method achieves superior performance on a carefully established benchmark for real-world disease screening, yielding significantly higher specificity at clinically meaningful recall rates.

EviScreen: Evidential Reasoning Advances Interpretable Real-World Disease Screening

Summary

EviScreen: Evidential Reasoning Advances Interpretable Real-World Disease Screening

Paper Overview

Abstract (translated)

Links

Tags