###Lim(b)etrics: An AI-enabled toolkit for capturing better preclinical musculoskeletal metrics and efficient translation of clinically relevant therapeutics
Project ID: 2228bd1192 (You will need this ID for your application)
Research Theme: Healthcare Technologies
UCL Lead department: Division of Surgery and Interventional Sciences
Lead Supervisor: Tom Carlson
Project Summary:
Osteoarthritis (OA) is a leading cause of chronic pain and disability, so is a major focus for pharmaceutical development. However, currently available devices for measuring limb function in preclinical rodent models are crude and often involve placing the animal in an unnatural environment or extensive handling. This leads to data that is biased due to panic exhibited by the mouse. A better solution would be to take these measurements in an environment without humans present, where the mouse is calm.
We are therefore developing AI-enabled diagnostics to use in our preclinical settings which will translate to better therapeutics in the clinical setting. In particular, we are developing a low-cost modular AI system to be integrated into the preclinical setting that will use state-of-the-art machine learning techniques to automatically classify behaviours and generate clinically relevant metrics for the duration of the trial.
This project builds upon two of our previous MSc projects, which have already shown good progress using computer vision techniques to segment and track rodents. Therefore, this PhD will (1) further develop the computer-vision based tracking system with IR cameras for monitoring rodents in the dark (mice are nocturnal), and (2) develop an instrumented modular stair that can be placed within the animal housing to measure reaction forces from each paw when climbing. This will involve some mechatronic prototyping using electronic components such as strain gauges and microcontrollers (e.g. Arduino) in combination with CAD 3D printing / laser cutting etc. Machine techniques and advance signal processing will be required to calibrate and verify the prototype. The PhD will culminate with a validation of the integrated system using mice at the Royal Veterinary College (RVC).
The project will be co-supervised by Prof Tom Carlson (UCL) and Dr Scott Roberts (RVC).