Comparative assessment of established and deep learning-based segmentation methods for hippocampal volume estimation in brain magnetic resonance imaging analysis

Hsi-Chun Wang; Chia-Sho Chen; Chung-Chin Kuo; Teng-Yi Huang; Kuei-Hong Kuo; Tzu-Chao Chuang; Yi-Ru Lin; Hsiao-Wen Chung; Alzheimer's Disease Neuroimaging Initiative

doi:10.1002/nbm.5169

Comparative assessment of established and deep learning-based segmentation methods for hippocampal volume estimation in brain magnetic resonance imaging analysis

NMR Biomed. 2024 May 7:e5169. doi: 10.1002/nbm.5169. Online ahead of print.

Authors

Hsi-Chun Wang¹, Chia-Sho Chen¹, Chung-Chin Kuo¹, Teng-Yi Huang¹, Kuei-Hong Kuo^{2

3}, Tzu-Chao Chuang⁴, Yi-Ru Lin⁵, Hsiao-Wen Chung⁶; Alzheimer's Disease Neuroimaging Initiative

Affiliations

¹ Department of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan.
² Division of Medical Image, Far Eastern Memorial Hospital, New Taipei City, Taiwan.
³ School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan.
⁴ Department of Electrical Engineering, National Sun Yat-Sen University, Kaohsiung, Taiwan.
⁵ Department of Electronic and Computer Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan.
⁶ Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan.

PMID: 38712667
DOI: 10.1002/nbm.5169

Abstract

In this study, our objective was to assess the performance of two deep learning-based hippocampal segmentation methods, SynthSeg and TigerBx, which are readily available to the public. We contrasted their performance with that of two established techniques, FreeSurfer-Aseg and FSL-FIRST, using three-dimensional T1-weighted MRI scans (n = 1447) procured from public databases. Our evaluation focused on the accuracy and reproducibility of these tools in estimating hippocampal volume. The findings suggest that both SynthSeg and TigerBx are on a par with Aseg and FIRST in terms of segmentation accuracy and reproducibility, but offer a significant advantage in processing speed, generating results in less than 1 min compared with several minutes to hours for the latter tools. In terms of Alzheimer's disease classification based on the hippocampal atrophy rate, SynthSeg and TigerBx exhibited superior performance. In conclusion, we evaluated the capabilities of two deep learning-based segmentation techniques. The results underscore their potential value in clinical and research environments, particularly when investigating neurological conditions associated with hippocampal structures.

Keywords: deep learning; hippocampal segmentation.

Abstract

Grants and funding