BEGIN:VCALENDAR VERSION:2.0 PRODID:-//talks.cam.ac.uk//v3//EN BEGIN:VTIMEZONE TZID:Europe/London BEGIN:DAYLIGHT TZOFFSETFROM:+0000 TZOFFSETTO:+0100 TZNAME:BST DTSTART:19700329T010000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:+0100 TZOFFSETTO:+0000 TZNAME:GMT DTSTART:19701025T020000 RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT CATEGORIES:Isaac Newton Institute Seminar Series SUMMARY:Kirk Lecture: Machine-Learning Enabled Imaging: F rom Microscopy to Medical Imaging to Astronomy - R ebecca Willett (University of Chicago) DTSTART;TZID=Europe/London:20211027T160000 DTEND;TZID=Europe/London:20211027T170000 UID:TALK164641AThttp://talks.cam.ac.uk URL:http://talks.cam.ac.uk/talk/index/164641 DESCRIPTION:
In many scientific and medical settings\, we ca nnot directly observe images of interest\, such as a person&rsquo\;s internal organs\, the microscop ic structure of materials or cells\, or distant st ars and galaxies. Rather\, we use MRI scanners\, m icroscopes\, and satellites to collect indirect da ta that require sophisticated algorithms to form a n image. Historically\, these methods have relied on mathematical models of simple image structures to improve the quality and resolution of the resul ting images. In this talk\, I will describe recent efforts to harness vast collections of images to train computers to learn more complex models of im age structure\, yielding more accurate and higher- resolution images than ever. These new methods lea d to new insights into designing neural networks i n a principled manner to jointly leverage both tra ining data and physical models of how imaging data is collected. We will conclude with a discussion of some of the main open questions and exciting ne w directions in this emerging field.
LOCATION:Seminar Room 1\, Newton Institute CONTACT: END:VEVENT END:VCALENDAR