AutoOCT: Development of a fully automated system for unbiased coronary artery disease evaluation for drug efficacy studies and risk prediction

Atherosclerosis is the build up of fatty deposits in the artery walls, which causes them to narrow. As plaques develop they become unstable and can rupture, leading to heart attack or stroke. It’s therefore important for doctors to have reliable ways of assessing the risk of a plaque rupturing.

Optical coherence tomography (OCT) is the best method for studying these atherosclerotic plaques in our blood vessels, but it has limitations: currently it cannot distinguish between the boundaries of different tissues, it sometimes shows features that aren’t really there (artefacts) and it generates large amounts of data which is difficult to process and interpret.

Professor Bennett’s has and his team have developed a new method called ‘Auto-OCT’, a fully automated system that uses artificial intelligence to correct artefacts, identifies disease areas and measures the features of high-risk plaques as effectively as current methods used by doctors. In this project, Auto-OCT will be further developed so that it’s feasibility in a clinical setting can be assessed.

To do this, the team will need to prove that Auto-OCT can reliably identify plaque progression and changes in plaque composition.

If approved by healthcare regulators, it could then be used to help doctors predict which plaques are most likely to rupture and to monitor whether drugs are effective at preventing plaque rupture.