PhysioROM

  • First Place in the HIMSS AsiaPac18 Challenge: awarded to Coviu (our partners) for best healthcare innovation
  • Cooperative Research Centre Project (CRC-P): Transforming joint surgery rehabilitation with artificial intelligence in telehealth
Intro

Product development and integration.

Working with a telehealth solution provider, Coviu, we made a plugin that allows physio to measure the range of motion of a patient over webcam. We are now the lead developers working with CSIRO Data61 to implement a clinically validated version of the project funded by a CRC-P.

Providing opportunities for cheap, equitable, and accessible healthcare is one of the central goals of the telehealth revolution. Telehealth connects doctors to remote patients using video conferencing and other digital tools. We are developing and clinically validating a computer vision application that enables physiotherapists to rehabilitate patients after knee-joint reconstruction over telehealth.

This project is funded by a Cooperative Research Centre Project (CRC-P) Grant from the Australian Federal Government — Australia’s primary research translation grant. The project involves multi-institutional collaboration between telehealth service provider Coviu, Australia’s national science agency CSIRO, the Physiotherapy School at the University of Western Australia and the Hollywood Functional Rehabilitation Clinic. Our primary role in the project is the research and development of the application and its deployment. In addition to our technical expertise, our consultation was instrumental in winning the grant and managing the many different stakeholder requirements.

We developed 3D and 2D human pose-estimation algorithms optimized to measure joint range of motion during physiotherapy sessions. We also used CSIRO’s Motion Capture Facility to create a large bespoke 3D human pose dataset to train our machine learning models. We are currently working closely with the physiotherapists to refine the algorithm and advising on their clinical validation of it.

We are assisting Coviu to prepare our research for commercialization. We are creating a scalable cloud deployment that serves computer vision applications. This is designed to allow real-time processing of video streams from many concurrent connections. We have also developed an intuitive interface to the application that plugs into Coviu’s online telehealth platform. The user experience of the plugin was designed in consultation with both clinicians and patients, allowing clinician customization and ensuring a clear and intuitive experience for patients.