Validation of Opportunistic Artificial Intelligence-based Bone Mineral Density Measurements in Coronary Artery Calcium Scans.

Researchers

Journal

Modalities

Models

Abstract

Previously we reported a manual method of measuring thoracic vertebral bone mineral density (BMD) using quantitative computed tomography (QCT) in non-contrast cardiac CT scans used for coronary artery calcium (CAC) scoring. In this report, we present validation studies of an artificial intelligence (AI) based automated BMD measurement (AutoBMD) that recently received FDA approval as an opportunistic add-on to CAC scans.A deep learning model was trained to detect vertebral bodies. Subsequently signal processing techniques were developed to detect intervertebral discs and the trabecular components of the vertebral body. The model was trained using 132 CAC scans comprising 7649 slices. To validate AutoBMD, we used all 6776 cases of manual BMD measurements previously reported from CAC scans in the Multi-Ethnic Study of Atherosclerosis (MESA).Mean±SD for AutoBMD and manual BMD were 166.3 ± 47.9 g/cm3 and 163.0 ± 46 g/cm3 respectively (p=0.23). MESA cases were 47% male and 53% female with age 61.9±10.2. Human experts vs. AutoBMD reported 27% vs. 27% for osteoporosis, and 43% vs. 41% for osteopenia (p=0.7). A strong correlation was found between AutoBMD and manual measurements (R = 0.84, p<0.0001). AutoBMD averaged 15 seconds per report vs. 5.5 minutes for manual measurements (p<0.0001).AutoBMD is an FDA approved AI-enabled opportunistic tool that reports BMD with Z-scores and T-scores, and accurately detects osteoporosis and osteopenia in CAC scans, demonstrating results comparable to manual measurements. No extra cost of scanning and no extra radiation to patients, plus the high prevalence of asymptomatic osteoporosis, makes AutoBMD a promising candidate to enhance patient care, if our findings are corroborated in other studies.Copyright © 2023. Published by Elsevier Inc.