Date: 1 December 2015
Location: Wolfson Lecture Theatre, Computing, Queen Mother Building
Host: Prof. Stephen McKenna
Title: Phenotyping and Computer Vision – a match made in heaven
Abstract: An increasing population and climate change put pressure on several societally important factors. Health costs are increasing and at the same time feeding the world becomes a challenge. Imaging is central to furthering our understanding of biology via phenotyping and can also help us deliver affordable and sustainable health care and agricultural output. In this talk I will show progress in developing solutions for phenotyping with MRI (cardiac and brain MRI), but also in coupling computer vision with sensing and compression to present affordable solutions to plant phenotyping in agriculture. Specifically, I will present recent work towards non-invasive, needle free approaches to ischemia diagnosis with cardiac BOLD MRI. I will show how subspace decompositions with structured shift-invariant dictionary learning can be used to extract phenotypes and build detection mechanisms. On the second part, I will highlight interesting computer vision problems that arise in high throughput plant phenotyping and our affordable platform (http://phenotiki.com).
BIO: Dr. Sotirios A. Tsaftaris received the M.Sc. and Ph.D. degrees in Electrical and Computer Engineering from Northwestern University, Evanston, IL, in 2003 and 2006, respectively, and the Diploma degree in Electrical and Computer Engineering from the Aristotle University of Thessaloniki, Thessaloniki, Greece, in 2000. He is currently a Chancellor's Fellow (at a Senior Lecturer level) at the University of Edinburgh. Previously he was an Assistant Professor with IMT Institute for Advanced Studies, Lucca, Italy. He remains as the Director of the Pattern Recognition and Image Analysis Unit at IMT. Prior to that, he held a joint Research Assistant Professor appointment at Northwestern University with the Departments of Electrical Engineering and Computer Science (EECS) and Radiology Feinberg School of Medicine. He maintains an adjunct appointment with EECS. He has published extensively, particularly in interdisciplinary fields, with more than 100 journal and conference papers in his active record. His research interests are image analysis, image processing, data mining and machine learning, and distributed computing.
http://prian.imtlucca.it (lab site)
(Please note change in time to start at 13:00 hrs - was originally scheduled to start at 12:00 hrs)