Meet The Researcher

My early interest in astronomy and astrophotography is probably what led me down the path of medical imaging.

We have been developing machine learning techniques for analysing medical imaging and assisting radiologists with diagnostic imaging interpretation. As part of this work, we identified key application areas in MS neuroimaging. This was exciting for us since both the machine learning and clinical application presented unique challenges (and steep learning curves).

The recent development of deep learning artificial intelligence techniques has already impacted all aspects of medical research and clinical practice, especially imaging. It is very exciting to see how ongoing progress in deep learning will improve the diagnosis and monitoring of MS, ultimately leading to better outcomes for patients with MS.

Our project will develop deep learning techniques for analysing longitudinal MS neuroimaging. In the first part of the project, we will develop algorithms for detecting and localising lesion changes on MRI scans. Using these algorithms, we will develop visualization tools to highlight areas in the brain where there are lesion changes. Such tools may help radiologists detect subtle lesion changes and reduce false negative reports. In the second part of the project, we aim to develop models for predicting stable/progressive lesions on neuroimaging. These may improve prognostication as well as influence choice of therapy.

Magnetic resonance imaging (MRI) plays a central role in the clinical management of MS. MRI is key to diagnosing MS, as well as monitoring disease activity and response to therapy. The artificial intelligence tools we will develop for analysing MRI scans will assist clinicians and radiologists to monitor MS disease activity, and predict progression and response to therapy. Such tools will enhance decision making and enable the tailoring of therapies to individuals based on imaging findings.

As a clinician scientist, I most enjoy working on projects that I believe will improve patient outcomes, since this gives extra meaning to the research. To that end, one of the challenges we constantly face is identifying relevant questions and promising avenues to take our research, since there are always ideas that pop up during the course of routine clinical work!