EORS-ISTA Symposium: AI, Machine Learning & Data Science as Basis for Future Developments in Medical Device Related Clinical Applications - An Industrial Point of View

Description

The presentation from the ISTA 2025 conference held in Rome, begins with an overview of machine learning and data science applications to advancing medical devices, specifically in knee arthroplasty over the past 20-25 years. While knee arthroplasty has seen high success rates, patient satisfaction hovers around 80-85%, with 1 in 5 patients reporting dissatisfaction. A key focus is on the complexities of achieving proper alignment in knee procedures, as highlighted by classifications like MacDessi’s CPAK and Bushman’s categorization of knee alignment issues.



Professor Thomas Grupp discusses various classification systems, biomechanical challenges, and the impacts of alignment correction on outcomes. They emphasize the importance of a functional alignment strategy and navigating the orthopaedic operating room with real-time biomechanical simulations to enhance procedural outcomes. The evolving integration of advanced finite element modeling, personalized to patient-specific parameters, reveals significant computational demands but offers insights into optimal knee prosthetic designs.



Machine learning is introduced as a potential future innovation, capable of enhancing personalized approaches by predicting knee function based on extensive data. Solutions include digital twins and enhanced musculoskeletal modeling to accommodate diverse patient anatomies by simulating individual responses to implant designs and orientations.



In conclusion, the ultimate aim is to identify which implant design aligns best with each individual’s anatomical and kinematic needs, ensuring improved satisfaction and functional outcomes for knee arthroplasty patients.

Specialties

Conferences