TY - GEN
T1 - Geometric Deep Learning for Post-Menstrual Age Prediction Based on the Neonatal White Matter Cortical Surface
AU - Vosylius, Vitalis
AU - Wang, Andy
AU - Waters, Cemlyn
AU - Zakharov, Alexey
AU - Ward, Francis
AU - Le Folgoc, Loic
AU - Cupitt, John
AU - Makropoulos, Antonios
AU - Schuh, Andreas
AU - Rueckert, Daniel
AU - Alansary, Amir
N1 - Publisher Copyright:
© 2020, Springer Nature Switzerland AG.
PY - 2020
Y1 - 2020
N2 - Accurate estimation of the age in neonates is useful for measuring neurodevelopmental, medical, and growth outcomes. In this paper, we propose a novel approach to predict the post-menstrual age (PA) at scan, using techniques from geometric deep learning, based on the neonatal white matter cortical surface. We utilize and compare multiple specialized neural network architectures that predict the age using different geometric representations of the cortical surface; we compare MeshCNN, Pointnet++, GraphCNN, and a volumetric benchmark. The dataset is part of the Developing Human Connectome Project (dHCP), and is a cohort of healthy and premature neonates. We evaluate our approach on 650 subjects (727 scans) with PA ranging from 27 to 45 weeks. Our results show accurate prediction of the estimated PA, with mean error less than one week.
AB - Accurate estimation of the age in neonates is useful for measuring neurodevelopmental, medical, and growth outcomes. In this paper, we propose a novel approach to predict the post-menstrual age (PA) at scan, using techniques from geometric deep learning, based on the neonatal white matter cortical surface. We utilize and compare multiple specialized neural network architectures that predict the age using different geometric representations of the cortical surface; we compare MeshCNN, Pointnet++, GraphCNN, and a volumetric benchmark. The dataset is part of the Developing Human Connectome Project (dHCP), and is a cohort of healthy and premature neonates. We evaluate our approach on 650 subjects (727 scans) with PA ranging from 27 to 45 weeks. Our results show accurate prediction of the estimated PA, with mean error less than one week.
KW - Brain age
KW - Cortical surface
KW - Developing brain
KW - Geometric deep learning
KW - Graph neural networks
KW - MeshCNN
KW - PointNet
UR - http://www.scopus.com/inward/record.url?scp=85093079195&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-60365-6_17
DO - 10.1007/978-3-030-60365-6_17
M3 - Conference contribution
AN - SCOPUS:85093079195
SN - 9783030603649
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 174
EP - 186
BT - Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, and Graphs in Biomedical Image Analysis - 2nd International Workshop, UNSURE 2020, and 3rd International Workshop, GRAIL 2020, Held in Conjunction with MICCAI 2020, Proceedings
A2 - Sudre, Carole H.
A2 - Fehri, Hamid
A2 - Arbel, Tal
A2 - Baumgartner, Christian F.
A2 - Dalca, Adrian
A2 - Tanno, Ryutaro
A2 - Van Leemput, Koen
A2 - Wells, William M.
A2 - Sotiras, Aristeidis
A2 - Papiez, Bartlomiej
A2 - Ferrante, Enzo
A2 - Parisot, Sarah
PB - Springer Science and Business Media Deutschland GmbH
T2 - 2nd International Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, UNSURE 2020, and the 3rd International Workshop on Graphs in Biomedical Image Analysis, GRAIL 2020, held in conjunction with the 23rd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2020
Y2 - 8 October 2020 through 8 October 2020
ER -
