Working towards clinician-assisted diagnosis using advanced MRS imaging technology

Professor Carolyn Mountford, the CEO and Director of Research at the Translational Research Institute (TRI), leads a team that is developing an advanced imaging technique to detect biochemical changes in tissues - with the potential to detect early warning signs of inflammatory conditions, mental illness, brain trauma, pre-cancerous changes and cancer.

This technology, called Magnetic Resonance Spectroscopy (MRS), can provide an insight into biochemical changes in tissues that may be considered early warning signs of disease, without the need for invasive testing. 

MRS results can be obtained following a normal MRI scan, in as little as 20 minutes. Compared to the visual information obtained from an MRI about the anatomical structure of tissues, MRS can provide an insight into what biochemical changes may occur in certain areas of the body during the progression of disease.

Professor Mountford’s team specialises in a particular type of MRS called 2-dimensional localised correlated spectroscopy (2D L-COSY), which provides a more detailed picture of these biochemical changes than the traditional one-dimensional approach. 

Towards clinician-assisted diagnosis: Harnessing a ‘machine learning’ approach to data analysis

Using patient 2D MRS L-COSY data, Professor Mountford’s research team is working towards identifying data thresholds to enable 'clinician-assisted diagnosis' for various conditions.

The team employs machine-learning approach, whereby adaptive software incorporates patient data into an existing bank of information to generate thresholds or ‘diagnostic classifiers’ for certain conditions.

The aim is then for clinicians to use these diagnostic classifiers to detect early signs of disease, based on a patient's tissue biochemistry, before symptoms emerge.
 

Current projects

This research is part of work undertaken at the TRI Innovation and Translation Centre in Collaboration with Siemens Healthineers, including: