MIDL Vision 2021

Automated triaging of head MRI examinations using CNNs David Wood is a research associate at King’s College London . His paper about prioritizing abnormal scans has been accepted for a coveted long oral presentation at this year’s MIDL. He speaks to us ahead of his discussion session today. In recent years, there has been a growing demand for head MRI scans, but a shortage of radiologists to report them, so the time from having a scan to receiving a radiology report is increasing year on year. In the UK, for example, the reporting time for outpatient head MRI scans has increased every year since 2012, with only 2% of radiology departments able to fulfil their contractual requirements. With an increasingly aging population, and the pandemic creating additional backlogs, this situation is only going to get worse. We know that early detection of neurological conditions leads to better outcomes for the patient and the health system, so it is important to see abnormal scans as quickly as possible. The aim of this work is to take a queue of reports and automatically identify those with a high chance of being abnormal so that they can be prioritized by the radiologist. 6 Presentation VISION MIDL