The aim of this project is to develop an automatic pipeline to generate subject-specific musculoskeletal models from optical 3D surface scans. Optical 3D scans can be acquired within one minute and reproduce the body shape of the patient. By training a neural network, we can accurately estimate internal parameters like skeleton shape and muscle properties based on the external body shape captured by the scans. This innovative approach eliminates the reliance on medical imaging data or manual input, making our pipeline applicable for integration into clinical and sports practices.

Caretech symposium 2023

“We are collecting a dataset of full-body MRI scans and optical 3D body scans of healthy adults.”

— Judith Cueto Fernandez

ESB conference 2024

Currently, musculoskeletal (MSK) models can be personalized by processing medical imaging data, like magnetic resonance imaging (MRI) or computerized tomography (CT), which is time-consuming and costly. In the absence of medical imaging, it is common to linearly scale (no deformations) a generic MSK model with male bone geometries to resemble an individual [1,2]. The aim of our project is to develop a pipeline to enable the creation of personalized musculoskeletal models from optical 3D body surface scans. This abstract shows preliminary data emphasizing the differences in anatomical bony landmarks, joint centres and muscle moment arms between a MSK model with personalized geometries and two linearly scaled generic models.