Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children

Purpose: To develop and evaluate an automated segmentation method for accurate quantification of abdominal adipose tissue (AAT) depots (superficial subcutaneous adipose tissue [SSAT], deep subcutaneous adipose tissue [DSAT], and visceral adipose tissue [VAT]) in neonates and young children. Materia...

Full description

Saved in:
Bibliographic Details
Main Authors: Kway, Yeshe Manuel, Thirumurugan, Kashthuri, Tint, Mya Thway, Michael, Navin, Shek, Lynette Pei-Chi, Yap, Fabian Kok Peng, Tan, Kok Hian, Godfrey, Keith M., Chong, Yap Seng, Fortier, Marielle Valerie, Marx, Ute C., Eriksson, Johan G., Lee, Yung Seng, Velan, S. Sendhil, Feng, Mengling, Sadananthan, Suresh Anand
Other Authors: Lee Kong Chian School of Medicine (LKCMedicine)
Format: Article
Language:English
Published: 2023
Subjects:
Online Access:https://hdl.handle.net/10356/164171
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-164171
record_format dspace
spelling sg-ntu-dr.10356-1641712023-01-09T01:08:37Z Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children Kway, Yeshe Manuel Thirumurugan, Kashthuri Tint, Mya Thway Michael, Navin Shek, Lynette Pei-Chi Yap, Fabian Kok Peng Tan, Kok Hian Godfrey, Keith M. Chong, Yap Seng Fortier, Marielle Valerie Marx, Ute C. Eriksson, Johan G. Lee, Yung Seng Velan, S. Sendhil Feng, Mengling Sadananthan, Suresh Anand Lee Kong Chian School of Medicine (LKCMedicine) Duke–National University of Singapore Medical School KK Women’s and Children’s Hospital Science::Medicine Deep Learning Convolutional Neural Networks Purpose: To develop and evaluate an automated segmentation method for accurate quantification of abdominal adipose tissue (AAT) depots (superficial subcutaneous adipose tissue [SSAT], deep subcutaneous adipose tissue [DSAT], and visceral adipose tissue [VAT]) in neonates and young children. Materials and Methods: This was a secondary analysis of prospectively collected data, which used abdominal MRI data from Growing Up in Singapore Towards healthy Outcomes, or GUSTO, a longitudinal mother–offspring cohort, to train and evaluate a convolutional neural network for volumetric AAT segmentation. The data comprised imaging volumes of 333 neonates obtained at early infancy (age 2 weeks, 180 male neonates) and 755 children aged either 4.5 years (n = 316, 150 male children) or 6 years (n = 439, 219 male children). The network was trained on images of 761 randomly selected volumes (neonates and children combined) and evaluated on 100 neonatal volumes and 227 child volumes by using 10-fold validation. Automated segmentations were compared with expert-generated manual segmentation. Segmentation performance was assessed using Dice scores. Results: When the model was tested on the test datasets across the 10 folds, the model had strong agreement with the ground truth for all testing sets, with mean Dice similarity scores for SSAT, DSAT, and VAT, respectively, of 0.960, 0.909, and 0.872 in neonates and 0.944, 0.851, and 0.960 in children. The model generalized well to different body sizes and ages and to all abdominal levels. Conclusion: The proposed segmentation approach provided accurate automated volumetric assessment of AAT compartments on MR images of neonates and children. Agency for Science, Technology and Research (A*STAR) Ministry of Health (MOH) National Medical Research Council (NMRC) Supported in part by the Singapore National Research Foundation under its Translational and Clinical Research (TCR) Flagship Programme and administered by the Singapore Ministry of Health National Medical Research Council, Singapore (NMRC/TCR/004-NUS/2008; NMRC/TCR/012-NUHS/2014). Additional funding is provided by the Singapore Institute for Clinical Sciences, Agency for Science Technology and Research, Singapore. 2023-01-09T01:08:37Z 2023-01-09T01:08:37Z 2021 Journal Article Kway, Y. M., Thirumurugan, K., Tint, M. T., Michael, N., Shek, L. P., Yap, F. K. P., Tan, K. H., Godfrey, K. M., Chong, Y. S., Fortier, M. V., Marx, U. C., Eriksson, J. G., Lee, Y. S., Velan, S. S., Feng, M. & Sadananthan, S. A. (2021). Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children. Radiology: Artificial Intelligence, 3(5), e200304-. https://dx.doi.org/10.1148/ryai.2021200304 2638-6100 https://hdl.handle.net/10356/164171 10.1148/ryai.2021200304 34617030 2-s2.0-85119720032 5 3 e200304 en NMRC/TCR/004-NUS/2008 NMRC/TCR/012-NUHS/2014 Radiology: Artificial intelligence © 2021 RSNA. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Science::Medicine
Deep Learning
Convolutional Neural Networks
spellingShingle Science::Medicine
Deep Learning
Convolutional Neural Networks
Kway, Yeshe Manuel
Thirumurugan, Kashthuri
Tint, Mya Thway
Michael, Navin
Shek, Lynette Pei-Chi
Yap, Fabian Kok Peng
Tan, Kok Hian
Godfrey, Keith M.
Chong, Yap Seng
Fortier, Marielle Valerie
Marx, Ute C.
Eriksson, Johan G.
Lee, Yung Seng
Velan, S. Sendhil
Feng, Mengling
Sadananthan, Suresh Anand
Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children
description Purpose: To develop and evaluate an automated segmentation method for accurate quantification of abdominal adipose tissue (AAT) depots (superficial subcutaneous adipose tissue [SSAT], deep subcutaneous adipose tissue [DSAT], and visceral adipose tissue [VAT]) in neonates and young children. Materials and Methods: This was a secondary analysis of prospectively collected data, which used abdominal MRI data from Growing Up in Singapore Towards healthy Outcomes, or GUSTO, a longitudinal mother–offspring cohort, to train and evaluate a convolutional neural network for volumetric AAT segmentation. The data comprised imaging volumes of 333 neonates obtained at early infancy (age 2 weeks, 180 male neonates) and 755 children aged either 4.5 years (n = 316, 150 male children) or 6 years (n = 439, 219 male children). The network was trained on images of 761 randomly selected volumes (neonates and children combined) and evaluated on 100 neonatal volumes and 227 child volumes by using 10-fold validation. Automated segmentations were compared with expert-generated manual segmentation. Segmentation performance was assessed using Dice scores. Results: When the model was tested on the test datasets across the 10 folds, the model had strong agreement with the ground truth for all testing sets, with mean Dice similarity scores for SSAT, DSAT, and VAT, respectively, of 0.960, 0.909, and 0.872 in neonates and 0.944, 0.851, and 0.960 in children. The model generalized well to different body sizes and ages and to all abdominal levels. Conclusion: The proposed segmentation approach provided accurate automated volumetric assessment of AAT compartments on MR images of neonates and children.
author2 Lee Kong Chian School of Medicine (LKCMedicine)
author_facet Lee Kong Chian School of Medicine (LKCMedicine)
Kway, Yeshe Manuel
Thirumurugan, Kashthuri
Tint, Mya Thway
Michael, Navin
Shek, Lynette Pei-Chi
Yap, Fabian Kok Peng
Tan, Kok Hian
Godfrey, Keith M.
Chong, Yap Seng
Fortier, Marielle Valerie
Marx, Ute C.
Eriksson, Johan G.
Lee, Yung Seng
Velan, S. Sendhil
Feng, Mengling
Sadananthan, Suresh Anand
format Article
author Kway, Yeshe Manuel
Thirumurugan, Kashthuri
Tint, Mya Thway
Michael, Navin
Shek, Lynette Pei-Chi
Yap, Fabian Kok Peng
Tan, Kok Hian
Godfrey, Keith M.
Chong, Yap Seng
Fortier, Marielle Valerie
Marx, Ute C.
Eriksson, Johan G.
Lee, Yung Seng
Velan, S. Sendhil
Feng, Mengling
Sadananthan, Suresh Anand
author_sort Kway, Yeshe Manuel
title Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children
title_short Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children
title_full Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children
title_fullStr Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children
title_full_unstemmed Automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in MRI of neonates and young children
title_sort automated segmentation of visceral, deep subcutaneous, and superficial subcutaneous adipose tissue volumes in mri of neonates and young children
publishDate 2023
url https://hdl.handle.net/10356/164171
_version_ 1754611300615323648