BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Talks.cam//talks.cam.ac.uk// X-WR-CALNAME:Talks.cam BEGIN:VEVENT SUMMARY:Deep supervised level set method: an approach to fully automated s egmentation of cardiac MR images in patients with pulmonary hypertension - Jinming Duan\, Imperial College DTSTART:20180209T140000Z DTEND:20180209T150000Z UID:TALK93838@talks.cam.ac.uk CONTACT:Rachel Furner DESCRIPTION:Pulmonary hypertension (PH) is a heterogeneous multifactorial syndrome that overlaps multiple clinical classifications. It normally foll ows a rapidly progressive clinical course and thereby has high potential o f causing cardiac failure. In the UK\, it has a prevalence of 97 cases per million with standardised death rates between 4.5 and 12.3 per 100\,000. Cardiac magnetic resonance (CMR) is recognised as the gold standard for im aging of the heart. In particular\, it is a promising modality for automat ed survival prediction in PH. Accurate segmentation of CMR images is a fun damental step in predicting PH survival.\n\n\nIn this talk\, we introduce a novel and accurate method for segmentation of PH CMR images. The method explicitly takes into account the image features learned from a deep neura l network. To this end\, we estimate joint probability maps over both regi on and edge locations in CMR images using a fully convolutional network. D ue to the distinct morphology of PH hearts\, these probability maps can th en be incorporated in a single nested level-set optimisation framework to achieve multi-region segmentation with high efficiency. The proposed metho d obviates the need for level set initialisation\, a common drawback assoc iated with the local minimum nature of level set optimisation. This method is therefore fully automated. We show results on CMR cine images and demo nstrate that the proposed level set method supervised by deep convolutiona l networks leads to substantial improvements for CMR image segmentation in PH patients.\n LOCATION:MR11\, Centre for Mathematical Sciences END:VEVENT END:VCALENDAR