Developing AI to Identify Normal and Pathological Periprosthetic Osteoblastic Activity for Total Knee Arthroplasty

Summary

This project focuses on using AI and biomechanics to improve diagnosis and rehabilitation after knee replacement surgery.

How are we doing it?

The study will involve creating and training AI models using patient SPECT/CT scans, integrating FEA-derived stress maps to simulate mechanical load on implants. The combined data will help classify normal versus pathological bone responses, supporting more accurate and personalised rehabilitation strategies.

What happens next?

The next steps include developing segmentation models for SPECT/CT images to isolate regions of interest, followed by generating FEA-based stress maps to capture mechanical strain around the implant. These outputs will then be fused into the AI model for classification and validation through a pilot study with clinical partners. The findings will guide future NHS collaborations and larger-scale studies in post-surgical care.

Funding

NIHR Exeter BRC.

People Involved

Prof Karen Knapp, Dr. Junning Chen, Prof Xujiong Ye, Dr. Ben Waterson.