Automated deep learning segmentation of high-resolution 7 T postmortem MRI for quantitative analysis of structure-pathology correlations in neurodegenerative diseases
Abstract
Postmortem MRI allows brain anatomy to be examined at high resolution and to link pathology measures with morphometric measurements. However, automated segmentation methods for brain mapping in postmortem MRI are not well developed, primarily due to limited availability of labeled datasets, and heterogeneity in scanner hardware and acquisition protocols. In this work, we present a high resolution of 135 postmortem human brain tissue specimens imaged at 0.3 mm isotropic using a T2w sequence on a 7T whole-body MRI scanner. We developed a deep learning pipeline to segment the cortical mantle by benchmarking the performance of nine deep neural architectures, followed by post-hoc topological correction. We then segment four subcortical structures (caudate, putamen, globus pallidus, and thalamus), white matter hyperintensities, and the normal appearing white matter. We show generalizing capabilities across whole brain hemispheres in different specimens, and also on unseen images acquired at 0.28 mm^3 and 0.16 mm^3 isotropic T2*w FLASH sequence at 7T. We then compute localized cortical thickness and volumetric measurements across key regions, and link them with semi-quantitative neuropathological ratings. Our code, Jupyter notebooks, and the containerized executables are publicly available at: https://pulkit-khandelwal.github.io/exvivo-brain-upenn
Keywords
Cite
@article{arxiv.2303.12237,
title = {Automated deep learning segmentation of high-resolution 7 T postmortem MRI for quantitative analysis of structure-pathology correlations in neurodegenerative diseases},
author = {Pulkit Khandelwal and Michael Tran Duong and Shokufeh Sadaghiani and Sydney Lim and Amanda Denning and Eunice Chung and Sadhana Ravikumar and Sanaz Arezoumandan and Claire Peterson and Madigan Bedard and Noah Capp and Ranjit Ittyerah and Elyse Migdal and Grace Choi and Emily Kopp and Bridget Loja and Eusha Hasan and Jiacheng Li and Alejandra Bahena and Karthik Prabhakaran and Gabor Mizsei and Marianna Gabrielyan and Theresa Schuck and Winifred Trotman and John Robinson and Daniel Ohm and Edward B. Lee and John Q. Trojanowski and Corey McMillan and Murray Grossman and David J. Irwin and John Detre and M. Dylan Tisdall and Sandhitsu R. Das and Laura E. M. Wisse and David A. Wolk and Paul A. Yushkevich},
journal= {arXiv preprint arXiv:2303.12237},
year = {2023}
}
Comments
Preprint submitted to NeuroImage Project website: https://pulkit-khandelwal.github.io/exvivo-brain-upenn