Modelling future bone mineral density: Simplicity or complexity?

• Published in: Bone (New York, N.Y.) Vol. 187; p. 117178
• Main Authors: Erjiang, E., Carey, John J., Wang, Tingyan, Ebrahimiarjestan, Mina, Yang, Lan, Dempsey, Mary, Yu, Ming, Chan, Wing P., Whelan, Bryan, Silke, Carmel, O'Sullivan, Miriam, Rooney, Bridie, McPartland, Aoife, O'Malley, Gráinne, Brennan, Attracta
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.10.2024
• Subjects: Bone mineral density; Decision making; Deep learning; Longitudinal monitoring; Osteoporosis; Z-score; Deep learning; Longitudinal monitoring; Z-score; Osteoporosis; Bone mineral density; Decision making
• ISSN: 8756-3282; 1873-2763; 1873-2763
• DOI: 10.1016/j.bone.2024.117178

Osteoporotic fractures are a major global public health issue, leading to patient suffering and death, and considerable healthcare costs. Bone mineral density (BMD) measurement is important to identify those with osteoporosis and assess their risk of fracture. Both the absolute BMD and the change in BMD over time contribute to fracture risk. Predicting future fracture in individual patients is challenging and impacts clinical decisions such as when to intervene or repeat BMD measurement. Although the importance of BMD change is recognised, an effective way to incorporate this marginal effect into clinical algorithms is lacking. We compared two methods using longitudinal DXA data generated from subjects with two or more hip DXA scans on the same machine between 2000 and 2018. A simpler statistical method (ZBM) was used to predict an individual's future BMD based on the mean BMD and the standard deviation of the reference group and their BMD measured in the latest scan. A more complex deep learning (DL)-based method was developed to cope with multidimensional longitudinal data, variables extracted from patients' historical DXA scan(s), as well as features drawn from the ZBM method. Sensitivity analyses of several subgroups was conducted to evaluate the performance of the derived models. 2948 white adults aged 40–90 years met our study inclusion: 2652 (90 %) females and 296 (10 %) males. Our DL-based models performed significantly better than the ZBM models in women, particularly our Hybrid-DL model. In contrast, the ZBM-based models performed as well or better than DL-based models in men. Deep learning-based and statistical models have potential to forecast future BMD using longitudinal clinical data. These methods have the potential to augment clinical decisions regarding when to repeat BMD testing in the assessment of osteoporosis. •Statistical techniques & DL-based techniques can be used to develop models to predict future BMD.•The DL-based models performed better in women than men.•The ZBM-based models performed as well/better than the DL-based models in men.•These models can augment clinical decisions regarding when to repeat BMD testing.