Key result
Pos4CHD with FE-Gen improves synthesized fetal echocardiography alignment by ~4 percentage points.
Why the study?
The scarcity of positive training samples is a major bottleneck for developing automated diagnostic systems for fetal congenital heart defects, and existing conditional generative models remain task-specific and still require positive samples for training.
Does the Pos4CHD positive-free mask-guided counterfactual synthesis framework improve image synthesis fidelity and downstream automated diagnosis of fetal congenital heart defects (EbA, TAPVC, VSD) using only normal fetal echocardiography samples?
Does the Pos4CHD positive-free mask-guided counterfactual synthesis framework improve image synthesis fidelity and downstream automated diagnosis of fetal congenital heart defects (EbA, TAPVC, VSD) using only normal fetal echocardiography samples?
Pos4CHD enables the synthesis of anatomically accurate fetal congenital heart defect ultrasound images solely from normal cases, overcoming severe positive-case scarcity to significantly improve automated prenatal CHD classification, detection, and segmentation.
Answers are grounded in this paper and cite their sources.
May augment scarce fetal CHD training data; leaves open prospective clinical validation before adoption.
The scarcity of positive cases remains a major bottleneck in developing automated diagnostic systems for fetal Congenital Heart Defects~(CHDs), particularly for rare subtypes. Although normal cases are accessible with effort, positive cases are hard to collect due to low incidence rates and privacy concerns, causing severe class imbalance or absence of positive training samples. Existing works resort to synthesizing positives with conditional generative models, but they are task-specific and require positive training samples, hindering application for fetal CHDs. To this end, we propose a positive-free mask-guided counterfactual generation framework for fetal echocardiography, namely Pos4CHD, which enables realistic, high-quality synthesis of pseudo-positive samples from semantic masks, with only normal samples at hand. Technically, we first develop FE-Gen, a mask-guided generative model tailored for fetal heart ultrasound synthesis that mitigates histogram drift and enhances image-mask spatial alignment. Building on FE-Gen, Pos4CHD employs a mask-to-image generation paradigm for CHD synthesis. It first manipulates semantic masks to simulate fetal-heart structural anomalies and then renders anatomically consistent images. Extensive experiments on in-house datasets, together with expert evaluations, demonstrate the visual fidelity~($-12.97$ FID) and anatomical consistency~($+3.96%$ DSC) of synthesized images. Further validations across diverse real-world CHD diagnosis tasks exhibit promising results and consistent increments, including zero-shot classification~(achieving $93.25%$ ACC on EbA and $72.92%$ ACC on TAPVC), lesion detection~($+6.95%$ mAP50 on VSD), and anatomy segmentation~($+10.49%$ DSC on fetal heart) compared with baselines. Codes and models are publicly available on {https://github.com/DREAMXFAR/Pos4CHD}{GitHub}.
No takes yet. Share an insight, caveat, or question.
Xuan et al. (2026) studied Fetal Congenital Heart Defects (Ebstein's Anomaly, Total Anomalous Pulmonary Venous Connection, Ventricular Septal Defect) (n=2,855). Pos4CHD (with FE-Gen mask-guided counterfactual synthesis) vs. ControlNet baseline and real data alone was evaluated on Image-mask spatial alignment measured by Dice Similarity Coefficient (DSC) on synthesized fetal echocardiography. Pos4CHD with FE-Gen improved fetal ultrasound synthesis fidelity and spatial alignment, achieving 93.25% zero-shot accuracy on EbA and +6.95% mAP50 on VSD detection.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: