Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Region-specific Risk Quantification for Interpretable Prognosis of COVID-19

Zhusi Zhong, Jie Li, Zhuoqi Ma, Scott Collins, Harrison Bai, Paul Zhang, Terrance Healey, Xinbo Gao, Michael K. Atalay, Zhicheng Jiao

TL;DR

This paper tackles the need for interpretable prognosis of COVID-19 from chest X-ray images by marrying a large-scale pretrained encoder with region-aware explanations. The method uses risk-specific Grad-CAM to localize prognostic regions and a Faster R-CNN–based anatomical detector with a Region Completer to produce 29-region regional risk outputs. It reports improved concordance index and time-dependent AUC on multicenter data and shows qualitative regional interpretability that highlights clinically meaningful areas. The work advances trustworthy AI in radiology by translating global risk into region-level, actionable prognostic insights.

Abstract

The COVID-19 pandemic has strained global public health, necessitating accurate diagnosis and intervention to control disease spread and reduce mortality rates. This paper introduces an interpretable deep survival prediction model designed specifically for improved understanding and trust in COVID-19 prognosis using chest X-ray (CXR) images. By integrating a large-scale pretrained image encoder, Risk-specific Grad-CAM, and anatomical region detection techniques, our approach produces regional interpretable outcomes that effectively capture essential disease features while focusing on rare but critical abnormal regions. Our model's predictive results provide enhanced clarity and transparency through risk area localization, enabling clinicians to make informed decisions regarding COVID-19 diagnosis with better understanding of prognostic insights. We evaluate the proposed method on a multi-center survival dataset and demonstrate its effectiveness via quantitative and qualitative assessments, achieving superior C-indexes (0.764 and 0.727) and time-dependent AUCs (0.799 and 0.691). These results suggest that our explainable deep survival prediction model surpasses traditional survival analysis methods in risk prediction, improving interpretability for clinical decision making and enhancing AI system trustworthiness.

Region-specific Risk Quantification for Interpretable Prognosis of COVID-19

TL;DR

This paper tackles the need for interpretable prognosis of COVID-19 from chest X-ray images by marrying a large-scale pretrained encoder with region-aware explanations. The method uses risk-specific Grad-CAM to localize prognostic regions and a Faster R-CNN–based anatomical detector with a Region Completer to produce 29-region regional risk outputs. It reports improved concordance index and time-dependent AUC on multicenter data and shows qualitative regional interpretability that highlights clinically meaningful areas. The work advances trustworthy AI in radiology by translating global risk into region-level, actionable prognostic insights.

Abstract

The COVID-19 pandemic has strained global public health, necessitating accurate diagnosis and intervention to control disease spread and reduce mortality rates. This paper introduces an interpretable deep survival prediction model designed specifically for improved understanding and trust in COVID-19 prognosis using chest X-ray (CXR) images. By integrating a large-scale pretrained image encoder, Risk-specific Grad-CAM, and anatomical region detection techniques, our approach produces regional interpretable outcomes that effectively capture essential disease features while focusing on rare but critical abnormal regions. Our model's predictive results provide enhanced clarity and transparency through risk area localization, enabling clinicians to make informed decisions regarding COVID-19 diagnosis with better understanding of prognostic insights. We evaluate the proposed method on a multi-center survival dataset and demonstrate its effectiveness via quantitative and qualitative assessments, achieving superior C-indexes (0.764 and 0.727) and time-dependent AUCs (0.799 and 0.691). These results suggest that our explainable deep survival prediction model surpasses traditional survival analysis methods in risk prediction, improving interpretability for clinical decision making and enhancing AI system trustworthiness.
Paper Structure (11 sections, 4 equations, 4 figures, 1 table)

This paper contains 11 sections, 4 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. COVID-19 Survival Prediction Model
  4. Risk-specific Grad-CAM
  5. Anatomical region detection and completion
  6. Regional Risk Prediction
  7. Experiment
  8. Datasets
  9. Implementation Details
  10. Results
  11. Conclusions

Figures (4)

  • Figure 1: Overview of the Survival Prediction (SP) model with regional interpretable prediction. The SP model first generates risk scores based on input CXR (Green). The risk Grad-CAM is calculated by backpropagating the global risk scores to locate disease areas (Blue). A branch detects the 29 anatomical bounding boxes for computing regional risk Grad-CAM and scores, arranging the region names in risk levels for a interpretable and understandable output to healthcare professionals (Orange).
  • Figure 2: The illustration of the anatomical region detector with Faster R-CNN and the proposed Region Completer, which corrects bounding box of the undetected regions with the learned spatial coordinate pattern.
  • Figure 3: Survival curves visualization of survival probabilities for patients of multi-center testing sets, classified into high and low risk groups.
  • Figure 4: Visualizations of the regional interpretable results.