Beyond Standard Sampling: Metric-Guided Iterative Inference for Radiologists-Aligned Medical Counterfactual Generation

Elham Ghelichkhan, Tolga Tasdizen

Keywords: Diffusion Models, Counterfactual, Evaluation Metrics, Inference, Explainable AI, Medical AI

Abstract:

Generative counterfactuals offer a promising avenue for explainable AI in medical imaging, yet ensuring these synthesized images are both anatomically faithful and clinically effective remains a significant challenge. This work presents a domain-specific diffusion framework for generating "healthy" counterfactuals from chest X-rays with cardiomegaly, underpinned by a systematic metric-guided inference strategy. In contrast to methods relying on static sampling parameters, our approach iteratively explores the inference hyperparameter space to maximize our composite selection criterion, $\mathrm{CF\_Score}$, that integrates our novel Faithfulness-Effectiveness Trade-off ($FET$) metric.

We extend the evaluation of counterfactual utility beyond simple classification shifts by conducting the simultaneous validation against radiologist annotations and eye-tracking data. Using the REFLACX dataset, we demonstrate that difference maps derived from our counterfactuals exhibit strong spatial alignment with expert visual attention and annotation. Quantified by Normalized Cross-Correlation, Hit Rate, pixel-wise ROC-AUC, and AUC-IoU, our results confirm that metric-guided counterfactuals provide dense and clinically relevant localizations of pathology that closely mirror human diagnostic reasoning.

Primary Subject Area: Generative Models

Secondary Subject Area: Interpretability and Explainable AI

Registration Requirement: Yes

Visa & Travel: Yes

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 381