Improved Segmentation of the Intracranial and Ventricular Volumes in Populations with Cerebrovascular Lesions and Atrophy Using 3D CNNs

Authors

Emmanuel E. Ntiri, University of Toronto
Melissa F. Holmes, University of Toronto
Parisa M. Forooshani, University of Toronto
Joel Ramirez, University of Toronto
Fuqiang Gao, University of Toronto
Miracle Ozzoude, University of Toronto
Sabrina Adamo, University of Toronto
Christopher J.M. Scott, University of Toronto
Dar Dowlatshahi, L'Hôpital d'Ottawa
Jane M. Lawrence-Dewar, Thunder Bay Regional Health Research Institute
Donna Kwan, York University
Anthony E. Lang, Toronto Western Hospital University of Toronto
Sean Symons, University of Toronto
Robert Bartha, Robarts Research InstituteFollow
Stephen Strother, University of Toronto
Jean Claude Tardif, Institut de Cardiologie de Montreal
Mario Masellis, University of Toronto
Richard H. Swartz, University of Toronto
Alan Moody, University of Toronto
Sandra E. Black, University of Toronto
Maged Goubran, University of Toronto

Document Type

Article

Publication Date

10-1-2021

Journal

Neuroinformatics

Volume

19

Issue

4

First Page

597

Last Page

618

URL with Digital Object Identifier

10.1007/s12021-021-09510-1

Abstract

Successful segmentation of the total intracranial vault (ICV) and ventricles is of critical importance when studying neurodegeneration through neuroimaging. We present iCVMapper and VentMapper, robust algorithms that use a convolutional neural network (CNN) to segment the ICV and ventricles from both single and multi-contrast MRI data. Our models were trained on a large dataset from two multi-site studies (N = 528 subjects for ICV, N = 501 for ventricular segmentation) consisting of older adults with varying degrees of cerebrovascular lesions and atrophy, which pose significant challenges for most segmentation approaches. The models were tested on 238 participants, including subjects with vascular cognitive impairment and high white matter hyperintensity burden. Two of the three test sets came from studies not used in the training dataset. We assessed our algorithms relative to four state-of-the-art ICV extraction methods (MONSTR, BET, Deep Extraction, FreeSurfer, DeepMedic), as well as two ventricular segmentation tools (FreeSurfer, DeepMedic). Our multi-contrast models outperformed other methods across many of the evaluation metrics, with average Dice coefficients of 0.98 and 0.96 for ICV and ventricular segmentation respectively. Both models were also the most time efficient, segmenting the structures in orders of magnitude faster than some of the other available methods. Our networks showed an increased accuracy with the use of a conditional random field (CRF) as a post-processing step. We further validated both segmentation models, highlighting their robustness to images with lower resolution and signal-to-noise ratio, compared to tested techniques. The pipeline and models are available at: https://icvmapp3r.readthedocs.io and https://ventmapp3r.readthedocs.io to enable further investigation of the roles of ICV and ventricles in relation to normal aging and neurodegeneration in large multi-site studies.