Machine learning for abdominal aortic calcification assessment from bone density machine-derived lateral spine images
Machine learning for abdominal aortic calcification assessment from bone density machine-derived lateral spine images
Background: Lateral spine images for vertebral fracture assessment can be easily obtained on modern bone density machines. Abdominal aortic calcification (AAC) can be scored on these images by trained imaging specialists to assess cardiovascular disease risk. However, this process is laborious and requires careful training. Methods: Training and testing of model performance of the convolutional neural network (CNN) algorithm for automated AAC-24 scoring utilised 5012 lateral spine images (2 manufacturers, 4 models of bone density machines), with trained imaging specialist AAC scores. Validation occurred in a registry-based cohort study of 8565 older men and women with images captured as part of routine clinical practice for fracture risk assessment. Cox proportional hazards models were used to estimate the association between machine-learning AAC (ML-AAC-24) scores with future incident Major Adverse Cardiovascular Events (MACE) that including death, hospitalised acute myocardial infarction or ischemic cerebrovascular disease ascertained from linked healthcare data. Findings: The average intraclass correlation coefficient between imaging specialist and ML-AAC-24 scores for 5012 images was 0.84 (95% CI 0.83, 0.84) with classification accuracy of 80% for established AAC groups. During a mean follow-up 4 years in the registry-based cohort, MACE outcomes were reported in 1177 people (13.7%). With increasing ML-AAC-24 scores there was an increasing proportion of people with MACE (low 7.9%, moderate 14.5%, high 21.2%), as well as individual MACE components (all p-trend <0.001). After multivariable adjustment, moderate and high ML-AAC-24 groups remained significantly associated with MACE (HR 1.54, 95% CI 1.31–1.80 & HR 2.06, 95% CI 1.75–2.42, respectively), compared to those with low ML-AAC-24. Interpretation: The ML-AAC-24 scores had substantial levels of agreement with trained imaging specialists, and was associated with a substantial gradient of risk for cardiovascular events in a real-world setting. This approach could be readily implemented into these clinical settings to improve identification of people at high CVD risk. Funding: The study was supported by a National Health and Medical Research Council of Australia Ideas grant and the Rady Innovation Fund, Rady Faculty of Health Sciences, University of Manitoba.
Aortovascular disease, Cardiovascular disease, Dual-energy X-ray absorptiometry, Machine learning, Vascular calcification
104676
Sharif, Naeha
5d0b4845-bf4a-411e-ad35-a86027eff8d2
Gilani, Syed Zulqarnain
52ef5b81-29c8-4f80-a149-b94d29da825a
Suter, David
71c4ff17-962d-4b6b-a78b-19da42028241
Reid, Siobhan
b95e53ad-8d89-4151-8f90-e97e960e0153
Szulc, Pawel
1d62018f-3c1b-4ada-8f29-2624524023b9
Kimelman, Douglas
1977a6ec-afb2-4055-8fc5-2c722b72a9f3
Monchka, Barret A.
9297aee3-83b9-4457-8764-1f98b474ae70
Jozani, Mohammad Jafari
6b992ea6-2dbe-4265-af06-b16de5fb77a7
Hodgson, Jonathan M.
33a854a9-5174-4efa-adaa-5b7a18398e59
Sim, Marc
9f4ad207-620f-4dd9-8241-70269379cccf
Zhu, Kun
499a27a4-dbe3-42b6-8f20-c4cf02ea4194
Harvey, Nicholas C.
ce487fb4-d360-4aac-9d17-9466d6cba145
Kiel, Douglas P.
ff4d761f-9e48-415b-aae2-0c45381fd05f
Prince, Richard L.
f4f86568-7756-4da3-80ea-745cc5e9c606
Schousboe, John T.
f2b87d0a-88cb-462f-bc70-df2d984c9d1e
Leslie, William D.
5b2dd5d6-4569-40a3-a9b1-95152d11e4f1
Lewis, Joshua R.
9ea0ba64-4328-46d6-b23e-7e2b7e489e07
August 2023
Sharif, Naeha
5d0b4845-bf4a-411e-ad35-a86027eff8d2
Gilani, Syed Zulqarnain
52ef5b81-29c8-4f80-a149-b94d29da825a
Suter, David
71c4ff17-962d-4b6b-a78b-19da42028241
Reid, Siobhan
b95e53ad-8d89-4151-8f90-e97e960e0153
Szulc, Pawel
1d62018f-3c1b-4ada-8f29-2624524023b9
Kimelman, Douglas
1977a6ec-afb2-4055-8fc5-2c722b72a9f3
Monchka, Barret A.
9297aee3-83b9-4457-8764-1f98b474ae70
Jozani, Mohammad Jafari
6b992ea6-2dbe-4265-af06-b16de5fb77a7
Hodgson, Jonathan M.
33a854a9-5174-4efa-adaa-5b7a18398e59
Sim, Marc
9f4ad207-620f-4dd9-8241-70269379cccf
Zhu, Kun
499a27a4-dbe3-42b6-8f20-c4cf02ea4194
Harvey, Nicholas C.
ce487fb4-d360-4aac-9d17-9466d6cba145
Kiel, Douglas P.
ff4d761f-9e48-415b-aae2-0c45381fd05f
Prince, Richard L.
f4f86568-7756-4da3-80ea-745cc5e9c606
Schousboe, John T.
f2b87d0a-88cb-462f-bc70-df2d984c9d1e
Leslie, William D.
5b2dd5d6-4569-40a3-a9b1-95152d11e4f1
Lewis, Joshua R.
9ea0ba64-4328-46d6-b23e-7e2b7e489e07
Sharif, Naeha, Gilani, Syed Zulqarnain, Suter, David, Reid, Siobhan, Szulc, Pawel, Kimelman, Douglas, Monchka, Barret A., Jozani, Mohammad Jafari, Hodgson, Jonathan M., Sim, Marc, Zhu, Kun, Harvey, Nicholas C., Kiel, Douglas P., Prince, Richard L., Schousboe, John T., Leslie, William D. and Lewis, Joshua R.
(2023)
Machine learning for abdominal aortic calcification assessment from bone density machine-derived lateral spine images.
EBioMedicine, 94, , [104676].
(doi:10.1016/j.ebiom.2023.104676).
Abstract
Background: Lateral spine images for vertebral fracture assessment can be easily obtained on modern bone density machines. Abdominal aortic calcification (AAC) can be scored on these images by trained imaging specialists to assess cardiovascular disease risk. However, this process is laborious and requires careful training. Methods: Training and testing of model performance of the convolutional neural network (CNN) algorithm for automated AAC-24 scoring utilised 5012 lateral spine images (2 manufacturers, 4 models of bone density machines), with trained imaging specialist AAC scores. Validation occurred in a registry-based cohort study of 8565 older men and women with images captured as part of routine clinical practice for fracture risk assessment. Cox proportional hazards models were used to estimate the association between machine-learning AAC (ML-AAC-24) scores with future incident Major Adverse Cardiovascular Events (MACE) that including death, hospitalised acute myocardial infarction or ischemic cerebrovascular disease ascertained from linked healthcare data. Findings: The average intraclass correlation coefficient between imaging specialist and ML-AAC-24 scores for 5012 images was 0.84 (95% CI 0.83, 0.84) with classification accuracy of 80% for established AAC groups. During a mean follow-up 4 years in the registry-based cohort, MACE outcomes were reported in 1177 people (13.7%). With increasing ML-AAC-24 scores there was an increasing proportion of people with MACE (low 7.9%, moderate 14.5%, high 21.2%), as well as individual MACE components (all p-trend <0.001). After multivariable adjustment, moderate and high ML-AAC-24 groups remained significantly associated with MACE (HR 1.54, 95% CI 1.31–1.80 & HR 2.06, 95% CI 1.75–2.42, respectively), compared to those with low ML-AAC-24. Interpretation: The ML-AAC-24 scores had substantial levels of agreement with trained imaging specialists, and was associated with a substantial gradient of risk for cardiovascular events in a real-world setting. This approach could be readily implemented into these clinical settings to improve identification of people at high CVD risk. Funding: The study was supported by a National Health and Medical Research Council of Australia Ideas grant and the Rady Innovation Fund, Rady Faculty of Health Sciences, University of Manitoba.
Text
1-s2.0-S2352396423002414-main
- Proof
More information
Accepted/In Press date: 8 June 2023
e-pub ahead of print date: 11 July 2023
Published date: August 2023
Additional Information:
Funding Information:
The study was supported by a National Health and Medical Research Council of Australia Ideas grant and the Rady Innovation Fund, Rady Faculty of Health Sciences, University of Manitoba.The study was supported by a National Health and Medical Research Council of Australia Ideas grant (APP1183570) and the Rady Innovation Fund, Rady Faculty of Health Sciences, University of Manitoba. Hologic Inc. provided the software for JTS for image review. The salary of Dr. Lewis is supported by a National Heart Foundation Future Leader Fellowship (ID: 102817). The salary of Dr. Gilani's was partly supported through the Raine Priming Grant awarded by the Raine Medical Research Foundation. The salary of Dr. Sim is supported by a Royal Perth Hospital Research Foundation Fellowship (RPHRF CAF 00/21) and an Emerging Leader Fellowship from the Western Australian Future Health Research and Innovation Fund. Dr. Kiel's time was supported by a grant from the National Institute of Arthritis, Musculoskeletal and Skin Diseases (R01 AR 41398). Dr Harvey is supported by the UK Medical Research Council (MC_PC_21003; MC_PC_21001) and NIHR Southampton Biomedical Research Centre. The authors wish to thank the Australian Co-ordinating Registry, the Registries of Births, Deaths and Marriages, the Coroners, the National Coronial Information System and the Victorian Department of Justice and Community Safety for enabling Cause of Death Unit Record File (COD URF) data to be used for this publication. The authors acknowledge the Manitoba Centre for Health Policy for use of data contained in the Population Health Research Data Repository (HIPC 2016/2017–29). The results and conclusions are those of the authors and no official endorsement by the Manitoba Centre for Health Policy, Manitoba Health, Seniors and Active Living, or other data providers is intended or should be inferred. This article has been reviewed and approved by the members of the Manitoba Bone Density Program Committee.
Funding Information:
The study was supported by a National Health and Medical Research Council of Australia Ideas grant ( APP1183570 ) and the Rady Innovation Fund , Rady Faculty of Health Sciences , University of Manitoba . Hologic Inc. provided the software for JTS for image review. The salary of Dr. Lewis is supported by a National Heart Foundation Future Leader Fellowship (ID: 102817). The salary of Dr. Gilani's was partly supported through the Raine Priming Grant awarded by the Raine Medical Research Foundation . The salary of Dr. Sim is supported by a Royal Perth Hospital Research Foundation Fellowship (RPHRF CAF 00/21) and an Emerging Leader Fellowship from the Western Australian Future Health Research and Innovation Fund. Dr. Kiel's time was supported by a grant from the National Institute of Arthritis , Musculoskeletal and Skin Diseases (R01 AR 41398). Dr Harvey is supported by the UK Medical Research Council (MC_PC_21003; MC_PC_21001) and NIHR Southampton Biomedical Research Centre .
Publisher Copyright:
© 2023 The Authors
Keywords:
Aortovascular disease, Cardiovascular disease, Dual-energy X-ray absorptiometry, Machine learning, Vascular calcification
Identifiers
Local EPrints ID: 479888
URI: http://eprints.soton.ac.uk/id/eprint/479888
ISSN: 2352-3964
PURE UUID: ee7dc80e-98b4-41e7-a3c9-9f38cab120b0
Catalogue record
Date deposited: 28 Jul 2023 16:41
Last modified: 06 Jun 2024 01:42
Export record
Altmetrics
Contributors
Author:
Naeha Sharif
Author:
Syed Zulqarnain Gilani
Author:
David Suter
Author:
Siobhan Reid
Author:
Pawel Szulc
Author:
Douglas Kimelman
Author:
Barret A. Monchka
Author:
Mohammad Jafari Jozani
Author:
Jonathan M. Hodgson
Author:
Marc Sim
Author:
Kun Zhu
Author:
Douglas P. Kiel
Author:
Richard L. Prince
Author:
John T. Schousboe
Author:
William D. Leslie
Author:
Joshua R. Lewis
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
