Enhancing Generalized Fetal Brain MRI Segmentation using A Cascade Network with Depth-wise Separable Convolution and Attention Mechanism

TL;DR

This work tackles the challenge of robust fetal brain MRI segmentation across multi-site data and abnormal anatomies by introducing CasUNext, a cascaded two-stage network with Loc-Net for localization and Seg-Net for fine segmentation. The architecture leverages depthwise separable convolutions and inverted bottlenecks for efficiency, plus an Attention Gate to fuse features across skip connections, enabling effective integration of low- and high-level information. Evaluated on 150 scans (20–36 weeks) from two scanners and 50 abnormal brains, CasUNext achieves state-of-the-art Dice coefficients (e.g., $96.1\%$ Dice and $95.9\%$ MIOU on coronal data) and demonstrates robustness to motion artifacts and anatomical variability. Ablation studies confirm the critical contribution of the cascade, depthwise separable convolutions, and attention mechanisms to overall performance. These results suggest CasUNext can facilitate reliable fetal brain quantification across multi-site datasets and abnormal cases, with potential for downstream volumetric analyses and parcellation.

Abstract

Automatic segmentation of the fetal brain is still challenging due to the health state of fetal development, motion artifacts, and variability across gestational ages, since existing methods rely on high-quality datasets of healthy fetuses. In this work, we propose a novel cascade network called CasUNext to enhance the accuracy and generalization of fetal brain MRI segmentation. CasUNext incorporates depth-wise separable convolution, attention mechanisms, and a two-step cascade architecture for efficient high-precision segmentation. The first network localizes the fetal brain region, while the second network focuses on detailed segmentation. We evaluate CasUNext on 150 fetal MRI scans between 20 to 36 weeks from two scanners made by Philips and Siemens including axial, coronal, and sagittal views, and also validated on a dataset of 50 abnormal fetuses. Results demonstrate that CasUNext achieves improved segmentation performance compared to U-Nets and other state-of-the-art approaches. It obtains an average Dice coefficient of 96.1% and mean intersection over union of 95.9% across diverse scenarios. CasUNext shows promising capabilities for handling the challenges of multi-view fetal MRI and abnormal cases, which could facilitate various quantitative analyses and apply to multi-site data.

Figures (7)

• Figure 1: The algorithm flow of CasUnext
• Figure 2: The cascade network structure of CasUnext
• Figure 3: Loc-net/Seg-net structure
• Figure 4: The structure of the Attention Gate
• Figure 5: Comparison of results on the axial, coronal, and sagittal imaging dataset