Uncertainty-Calibrated Explainable AI for Fetal Ultrasound Plane Classification
This work addresses the need for trustworthy AI in prenatal screening, though it is incremental as it synthesizes existing methods into a clinical workflow.
The paper tackled the problem of unreliable fetal ultrasound plane classification due to domain shift and poor calibration by developing a framework that integrates uncertainty estimation with explainable AI, achieving improved calibration and selective prediction metrics on the FETAL_PLANES_DB benchmark.
Fetal ultrasound standard-plane classification underpins reliable prenatal biometry and anomaly screening, yet real-world deployment is limited by domain shift, image noise, and poor calibration of predicted probabilities. This paper presents a practical framework for uncertainty-calibrated explainable AI in fetal plane classification. We synthesize uncertainty estimation methods (Monte Carlo dropout, deep ensembles, evidential learning, and conformal prediction) with post-hoc and uncertainty-aware explanations (Grad-CAM variants, LIME-style local surrogates, and uncertainty-weighted multi-resolution activation maps), and we map these components to a clinician-facing workflow. Using FETAL_PLANES_DB as a reference benchmark, we define a reporting protocol that couples accuracy with calibration and selective prediction, including expected calibration error, Brier score, coverage-risk curves, and structured error analysis with explanations. We also discuss integration points for quality control and human-in-the-loop review, where uncertainty flags trigger re-acquisition or expert confirmation. The goal is a reproducible, clinically aligned blueprint for building fetal ultrasound classifiers whose confidence and explanations remain trustworthy under noisy acquisition conditions.