Machine learning in medical imaging

Relatore:  Marleen de Bruijne - University Medical Center Rotterdam
  martedì 12 maggio 2015 alle ore 17.00 16:45 rinfresco, 17:00 inizio seminario
Quantitative analysis of medical imaging data plays an increasingly important role both in large-scale clinical studies and in diagnosis, monitoring, and prognosis of disease in individual patients. Traditionally, medical image analysis techniques measure factors that are already known to be related to disease, such as for instance the density of lung tissue or the size of certain brain structures. The current availability of large databases that combine medical images with other patient data allows for a more data-driven approach in which the important characteristics related to disease outcome are learned directly from a training data set. This talk will cover different approaches to learn models of shape, appearance, and motion from imaging data. I will present techniques to address common issues in large medical imaging studies: differences between images acquired with varying scan protocols, weakly annotated data, and reconstruction of missing data. Applications are in lung, cardiovascular, and brain imaging

Ca' Vignal 1, Piano terra, Aula E

Manuele Bicego

Referente esterno
Data pubblicazione
7 maggio 2015

Offerta formativa