Cloud-AI4AIM

The primary aim of this project is to develop new technologies to improve the reliability and timeliness of diagnosis of neurological disorders. By integrating state-of-the-art computer vision techniques and expert clinical knowledge, a system that is minimally invasive, maximally scalable, and highly cost-effective could be developed to assess infant movements. Specifically, our proposed project will access and annotate/label videos collected from parents/carers, and extract movement data from videos of infants to detect abnormal movement patterns using machine learning.

This IAA project will enable us to access the aforementioned video database through:

1. Setting up a Trusted Research Environment (TRE) – a TRE will need to be set up on a secured cloud platform which will be sitting on Amazon Web Service (AWS). The TRE will ensure all the patient data will be used in a secure and compliant computing environment.
2. Developing a software prototype to facilitate and support clinicians in annotating the video data in the new TRE.
3. Organising co-creation workshops with stakeholders, including end-users (parents/carers), clinicians, and industrial partners, to inform us of the future translation research.