Document Type

Journal Article

Publication Title

eBioMedicine

Publisher

Elsevier

School

Nutrition and Health Innovation Research Institute / School of Science

RAS ID

58127

Funders

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.

Grant Number

NHMRC Number: APP1183570

Comments

Sharif, N., Gilani, S. Z., Suter, D., Reid, S., Szulc, P., Kimelman, D., ... & Lewis, J. R. (2023). Machine learning for abdominal aortic calcification assessment from bone density machine-derived lateral spine images. EBioMedicine, 94, article 104676. https://doi.org/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.

DOI

10.1016/j.ebiom.2023.104676

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Download

Share

 
COinS