AI augmented Digital Twin for knee surgery planning – AIKNEE

Digital Twins (DT) brings a new perspective for personalized medicine, especially in orthopedics for musculoskeletal assessment, injury prevention, or surgical planning. The KNEE-AI project focuses on the development of tools and methodologies to enhance state of the art digital twins, with the focus on systematic use of state-of-the-art Artificial Intelligence algorithms to predict with a high level of confidence the mechanical response and risk of tear of knee ligaments, especially the collateral and cruciate ligaments. Such augmented Digital Twin will be applied on several populations to be able to act as a decision aid model: (i) People who undergo a total knee arthroplasty, where lateral ligaments, and (if preserved) posterior cruciate ligament can be severely impacted by the presence of the implant. (ii) People who are either at risk of ligament injury (top level athletes) or those already injured (iii), to assess the impact of the loss of cruciate or lateral ligaments. Several scientific challenges are hindering the systematic utilization of Digital Twins: (i) the development of personalized digital twins must rely on efficient medical imaging analysis, segmentation and on reconstruction/meshing algorithms; (ii) simulation time, considering risk of ligament tear analysis, is not compatible with clinical decision time, especially if a wide range of simulation are required to transform the digital twin onto an efficient decision aid tool. (iii) There is no clear consensus on the model definition, boundary conditions and validation of such knee joint digital twin. The project objective is therefore the enhancement, using latest AI developments, of personalized digital twins of the knee articulation with a particular emphasis on the analysis of cruciate ligaments and collateral ligaments to address those 3 challenges.