Machine Learning Aided Detection of Brain Aneurysms

Intracranial aneurysms are relatively common, occurring in 2-5% of the general population. Rupture of the aneurysm can result in a stroke with a devastating 30-day mortality of 45%. Further, severe medical conditions are possible in which up to one third of patients may die before reaching the hospital and one third will become severely debilitated. Aneurysms are, however, difficult to find on magnetic resonance and computed tomography angiography scans (MRA and CTA scans) â especially when small or located close to the bone. Through the use of the existing knowledge and expertise of the neurovascular team at Toronto Western Hospital, as well as their database of CTA and MRA scans, the aim of this project is to improve on efforts in this field through the development of an artificial intelligence and machine learning method for the detection of intracranial aneurysms with a diagnostic accuracy approaching that of trained neuroradiologists. As unspecialized general radiologists read most neuroimaging exams due to a lack of specialized physicians in remote and rural areas, a long-term goal is to use results from this work to build a tool for the support of imaging review by non-specialists in emergency situations.

Faculty Supervisor:

Marzyeh Ghassemi

Student:

Daniel Dastoor

Partner:

Education and Scientific Services of Toronto Neuroradiology

Discipline:

Computer science

Sector:

Health care and social assistance

University:

Program: