Creating digital twins of the shoulder complex to improve pre-operative planning of joint replacement surgery

Computational models of the musculoskeletal system, coupled with wearable sensors for real-time feedback (i.e. the Digital Twin) have tremendous clinical application in the field of orthopaedics and joint replacement, which currently relies upon the skill and experience of a surgeon to achieve a good outcome. This is particularly true for shoulder arthroplasty (glenohumeral joint replacement), as the shoulder has a large range of motion that is stabilised and controlled by coordinated activation of many muscles crossing the joint. A computational model that represents the anatomy of the patient, can simulate the motion and forces of the joint and then predict clinical outcomes for varying implant designs and placements would provide orthopaedic surgeons with the necessary tools to plan surgeries and achieve better outcomes. However, several challenges need to be addressed before bringing the Digital Twin into clinical practice:

1. Creating models of the complex anatomy (bones, muscles, joints, ligaments etc) is time-consuming and required 3D medical imaging
2. Measuring and characterising movement of the shoulder is challenging, as the shoulder moves across the rib cage and lies underneath soft tissue
3. Bone is a living tissue, and responds to its surrounding mechanical environment, so predicting how an implant might perform in a patient requires some understanding and prediction of bone adaptation and remodelling processes.

We propose a paradigm shift towards personalised medicine for shoulder arthroplasty by developing methods to rapidly generate a Digital Twin of the shoulder, updating this model with wearable sensor technology to measure and characterise motion and computational approaches to bone remodelling. This talk will present some of the approaches and validation studies that have been completed thus far to address the challenges above. The methods and models developed within this work aim to provide the clinician with quantitative information in the shoulder arthroplasty surgery planning phase and ultimately improve long-term outcomes for patients.

The audio in this video begins at approximately 02:15.