05-06-2013
A CITI team from Multimodal Systems led by PI João Magalhães and students Flávio Martins and André Mourão scored high in AMIA Medical task of the 2013 ImageCLEF forum competition.The goal of ImageCLEF is to support the advancement of the field of visual media analysis, indexing, classification, and retrieval, and promotes the participation of both academic and industry research groups worldwide.The CITI team attacked several problems in the Medical Task and ranked 3th (among 22 high level international teams) which involves automatic classification of medical exams. As an example of the task, the software system is required to identify whether a RX picture represents a broken leg, or a pneumonic lungs. The CITI team ranked TOP MOST (among 46 high level international teams) in the Case based retrieval test, which involves correlating several media. In this task, the software system is required to look for information of patients with a similar situation, based on the automatic analysis of multimodal clinical data - text, images, etc). João Magalhães and collegues have been developing new techniques for image retrieval, classification and analysis, with successful applications in many domains, from interaction analysis, to immersive TV.
