Radiologist-Guided Causal Concept Bottleneck Models for Chest X-Ray Interpretation

2026-05-08

AI Summary

This work addresses the disconnect between model explanations and clinical reasoning in existing concept bottleneck models for medical imaging, which lack explicit modeling of the clinical processes through which diseases generate radiological findings. To bridge this gap, the authors propose XpertCausal—a radiologist-guided causal concept bottleneck model that, for the first time, integrates expert-defined causal structures into the concept bottleneck framework. The model employs a noisy-OR probabilistic mechanism to capture the causal generation of imaging signs from underlying pathologies and uses Bayesian inference to back-propagate predicted concepts into pathology probabilities, thereby constraining the reasoning pathway to align with clinical knowledge. Experiments on MIMIC-CXR demonstrate that XpertCausal significantly outperforms non-causal baselines and ablation variants in terms of AUROC, calibration, explanation quality, and alignment with expert reasoning pathways.

Abstract

Concept Bottleneck Models (CBMs) in medical imaging aim to improve model interpretability by predicting intermediate clinical concepts before final diagnoses. However, most existing CBMs treat concepts as discriminative predictors of pathology labels, without explicitly modelling the underlying clinical generative process where diseases produce observable radiographic findings. We propose XpertCausal, a radiologist-guided causal CBM for chest X-ray interpretation which models pathology-to-concept relationships using a probabilistic noisy-OR framework. This generative model is then inverted via Bayesian inference to estimate pathology probabilities from predicted concepts. Radiologist-curated concept-pathology associations are used to constrain model structure to radiologist-defined clinically plausible reasoning pathways. We evaluate XpertCausal on MIMIC-CXR across pathology classification performance, calibration, explanation quality, and alignment with radiologist-defined reasoning pathways. Compared with both a non-causal CBM baseline and a causal ablation with unconstrained learned associations, XpertCausal achieves improved AUROC, calibration, and clinically relevant explanation quality, while learning concept-pathology relationships that more closely align with expert knowledge. These results demonstrate that incorporating clinically motivated causal structure and expert domain knowledge into CBMs can lead to more accurate, interpretable, and clinically aligned models for CXR interpretation.

Problem

• Concept Bottleneck Models
• Chest X-Ray Interpretation
• Causal Modeling
• Radiologist-Guided Reasoning
• Clinical Interpretability

Innovation

• Causal Concept Bottleneck Model
• Noisy-OR
• Bayesian Inference
• Radiologist-Guided
• Chest X-Ray Interpretation