PhD Candidate
Max-Planck Institute for Intelligent Systems, Germany
Areas of Interest
- Artificial Intelligence
- Graphics
Poster
A population-based implant shapes set for fracture surgery
Abstract
To improve the treatment of bone fractures, implant manufacturers are producing anatomically contoured plates that match the surface of the target bone. But so far, those plates only fit part of the population and often need to be bent during surgery for a better fit. Existing methods to adapt implant shapes to a population involve a human in the loop who evaluates the fit and adjusts the shape consequently. In contrast, we propose a method that better leverages the statistic variability of the bone dataset. Making one plate per bone is impractical on a manufacturing level, so we aim to get a certain number of plate shapes, such that one plate fits several bones and most of the population is accommodated. We proceed by clustering, using a plate to bone distance metric, and a surgical fit criterion. Our preliminary results show that a set of 5 plate shapes accommodate more than 6% of our population sample, without the need to further bend the plate during surgery.
Bio
I am a Ph.D. student in the Department of Perceiving Systems at the Max Planck Institute for Intelligent Systems, advised by Professors Sergi Pujades and Michael J. Black. My work focuses on building a model of the long bones for the human body in order to design better fitting plates for fracture surgery. This work is part of the larger research project CAMed aiming to create 3D printed custom implants.
