FastSurfer-HypVINN: Automated sub-segmentation of the hypothalamus and adjacent structures on high-resolutional brain MRI

TL;DR

FastSurfer-HypVINN is a rapid, fully automated deep learning tool for accurate sub-segmentation of the hypothalamus and adjacent brain structures on high-resolution MRI, enhancing scalability and reproducibility in neuroimaging studies.

Contribution

The paper introduces HypVINN, a novel deep learning method capable of automated hypothalamic sub-segmentation on high-resolution MRI, robust to missing modalities and validated across multiple datasets.

Findings

High segmentation accuracy on T1w and T1w/T2w images

Generalizes well to 1.0 mm MR scans from large cohorts

Segmentation completed in less than a minute on GPU

Abstract

The hypothalamus plays a crucial role in the regulation of a broad range of physiological, behavioural, and cognitive functions. However, despite its importance, only a few small-scale neuroimaging studies have investigated its substructures, likely due to the lack of fully automated segmentation tools to address scalability and reproducibility issues of manual segmentation. While the only previous attempt to automatically sub-segment the hypothalamus with a neural network showed promise for 1.0 mm isotropic T1-weighted (T1w) MRI, there is a need for an automated tool to sub-segment also high-resolutional (HiRes) MR scans, as they are becoming widely available, and include structural detail also from multi-modal MRI. We, therefore, introduce a novel, fast, and fully automated deep learning method named HypVINN for sub-segmentation of the hypothalamus and adjacent structures on 0.8 mm isotropic T1w and T2w brain MR images that is robust to missing modalities. We extensively validate our model with respect to segmentation accuracy, generalizability, in-session test-retest reliability, and sensitivity to replicate hypothalamic volume effects (e.g. sex-differences). The proposed method exhibits high segmentation performance both for standalone T1w images as well as for T1w/T2w image pairs. Even with the additional capability to accept flexible inputs, our model matches or exceeds the performance of state-of-the-art methods with fixed inputs. We, further, demonstrate the generalizability of our method in experiments with 1.0 mm MR scans from both the Rhineland Study and the UK Biobank. Finally, HypVINN can perform the segmentation in less than a minute (GPU) and will be available in the open source FastSurfer neuroimaging software suite, offering a validated, efficient, and scalable solution for evaluating imaging-derived phenotypes of the hypothalamus.