Skeletal ageing phenotype digital biomarker for personalized bone care
Description: Skeletal ageing phenotype digital biomarker for personalized bone careThis Joint Research Project addresses the clinical need for more personalized and improved management of bone health, through the development of a digital health biomarker to assess skeletal (biological) ageing. We call this digital technology the skeletal ageing phenotype (SAP) score, as it supports clinical decision by improving the identification of patients that are at risk of developing fractures and predicting patient-specific response to pharmacological (drug) treatment. This novel technology builds on recent innovations in the cross-cutting research capabilities on artificial intelligence (AI) to predict health outcomes, and computer simulations to understand mechanisms of bone frailty.Fragility fractures are a major socio-economic burden, especially with the ageing of population in Asia. Over 50% of hip fractures are expected to occur in Asia by 2050, hence advancement in the diagnosis and prognosis of bone frailty is required to address this global public health issue. Recent advances in AI suggest that the assessment of biological age, and comparing it to the chronological age, can predict musculoskeletal outcome and help identify patients who are in need of early interventions. Moreover, advances in computational modelling showed that biofidelic simulations of falls from standing height improves the prediction of hip fracture risk and enable in silico trials of potential interventions. The synergistic combination of the two approaches in a hypermodel overcomes limitations associated with AI and mechanistic models alone, to leverage on high-dimensional data to predict outcomes, while improving understanding of underlying mechanisms of ageing. In this project, we will build AI-based (inductive) models to process clinical and three-dimensional imaging data that are collected as part of routine clinical work-flow, to extract features that will inform mechanistic (deductive), image-driven models of ageing. The mechanistic ageing models will be used in subject-specific biofidelic simulations of falls to characterise the mechanical response of bone under impact loading. Finally, the extracted clinical, imaging and structural features will serve as inputs to a hypermodel to predict skeletal age. All the models will be tested in smaller datasets, before being validated in large, combined cohorts from the Korean population against clinical outcomes. The developed SAP score will be applied to understand patient variance in terms of actual fracture status, mortality, and response to treatment, to personalize clinical decision making and treatment plans. By first building a gold standard based on quantitative computed tomography (QCT) images, the results can be extended to 2D planar medical images to maximize its utilization in clinical settings.
Amount: CHF 249,300
Funding type: research
Funder scheme: Korean-Swiss Science and Technology Programme
Start date: 2024-01-01
End date: 2026-12-31
Investigator: Helgason, Benedikt
Institution: ETH Zurich – ETHZ
Subfield: Orthopedics and Sports Medicine
Field: Medicine
Domain: Health Sciences
Funded outputs count: 0
Funder award ID: 218553
OpenAlex ID: G7645579365