Efficient multi-class fetal brain segmentation in high resolution MRI reconstructions with noisy labels

TL;DR

This paper introduces a transfer learning approach with noisy labels for automatic multi-class fetal brain segmentation in high-resolution MRI, demonstrating generalization across different reconstruction methods without manual annotations.

Contribution

It presents a novel transfer learning method that effectively segments fetal brain MRI into multiple tissue types using noisy labels, reducing the need for manual segmentation and improving generalization.

Findings

Network successfully segments into 7 tissue types across reconstruction methods.

Transfer learning improves segmentation accuracy compared to training from scratch.

No manual segmentation needed for training, enabling unbiased neurodevelopment analysis.

Abstract

Segmentation of the developing fetal brain is an important step in quantitative analyses. However, manual segmentation is a very time-consuming task which is prone to error and must be completed by highly specialized indi-viduals. Super-resolution reconstruction of fetal MRI has become standard for processing such data as it improves image quality and resolution. However, dif-ferent pipelines result in slightly different outputs, further complicating the gen-eralization of segmentation methods aiming to segment super-resolution data. Therefore, we propose using transfer learning with noisy multi-class labels to automatically segment high resolution fetal brain MRIs using a single set of seg-mentations created with one reconstruction method and tested for generalizability across other reconstruction methods. Our results show that the network can auto-matically segment fetal brain reconstructions into 7 different tissue types, regard-less of reconstruction method used. Transfer learning offers some advantages when compared to training without pre-initialized weights, but the network trained on clean labels had more accurate segmentations overall. No additional manual segmentations were required. Therefore, the proposed network has the potential to eliminate the need for manual segmentations needed in quantitative analyses of the fetal brain independent of reconstruction method used, offering an unbiased way to quantify normal and pathological neurodevelopment.