Annotation-free deep learning for detection and segmentation of fetal germinal matrix-intraventricular hemorrhage in brain MRI

2026-05-10

AI Summary

This study addresses the challenge of automatic detection and segmentation of germinal matrix–intraventricular hemorrhage (GMH-IVH) in fetal MRI, which is hindered by the scarcity of annotated data. To overcome this limitation, the authors propose FreeHemoSeg, a novel deep learning framework that operates without real hemorrhage annotations. Leveraging medical prior knowledge, the method synthesizes pseudo-hemorrhage images and trains exclusively on normal fetal T2-weighted MRI scans. Comprehensive multi-center internal and external validation demonstrates that FreeHemoSeg achieves strong performance, with Dice similarity coefficients of 0.559 (internal) and 0.512 (external), and superior AUROC compared to existing approaches. Furthermore, the model significantly enhances radiologists’ diagnostic sensitivity and efficiency in identifying GMH-IVH.

Abstract

Background: Prenatal germinal matrix-intraventricular hemorrhage (GMH-IVH) is a leading cause of infant mortality and neurodevelopmental impairment. Manual diagnosis and lesion segmentation are labor-intensive and error-prone. Deep learning models offer potential for automation but typically require large annotated datasets, which are challenging to obtain. Purpose: To develop and validate an annotation-free deep learning framework for automated detection and segmentation of GMH-IVH on brain MRI. Materials and Methods: This retrospective study analyzed 2D T2-weighted MRI data from pregnant women collected from October 2015 to October 2023 at one hospital (internal validation) and two hospitals (external validation). Eligible participants included healthy fetuses and those with GMH-IVH. FreeHemoSeg was developed and trained using pseudo GMH-IVH images synthesized from normal fetal data guided by medical priors. Primary outcomes included diagnostic accuracy (area under the ROC curve [AUROC], sensitivity, specificity) and segmentation accuracy (Dice similarity coefficient [DSC]). A reader study evaluated clinical utility. Results: A total of 1674 stacks from 558 pregnant women were analyzed. FreeHemoSeg achieved the highest performance in both internal (sensitivity: 0.914, 95% CI 0.869-0.945; specificity: 0.966, 95% CI 0.946-0.978; DSC: 0.559, 95% CI 0.546-0.571) and external validation (sensitivity: 0.824, 95% CI 0.739-0.885; specificity: 0.943, 95% CI 0.913-0.964; DSC: 0.512, 95% CI 0.497-0.526), outperforming supervised and unsupervised methods. FreeHemoSeg assistance improved radiologists' sensitivity (from 0.882 to 0.941-1.000) and diagnostic confidence while reducing interpretation time by 16.0-52.7%. Conclusion: FreeHemoSeg accurately detects and localizes fetal brain hemorrhages without annotated training data, enabling earlier diagnosis and supporting timely clinical management.

Problem

• germinal matrix-intraventricular hemorrhage
• annotation-free
• fetal brain MRI
• deep learning
• medical image segmentation

Innovation

• annotation-free
• deep learning
• fetal MRI
• pseudo-synthetic data
• germinal matrix hemorrhage