MRI cases containing cerebral tumors retrieval using Bayesian networks
Hedi Yazid, Karim Kalti, Fatma Elouni, Najoua Essoukri, Kalthoum Tlili · 2010
We propose in this paper a Bayesian model for the retrieving of MRI (magnetic resonance imaging) exams that contain cerebral tumors. Bayesian network proved its efficiency and reliability in several AI (Artificial Intelligence) problems and especially in aid-decision applications. To diagnose a cerebral tumor in a MRI exam, we need to interpret diverse sequences and to refer to visual descriptors and, also, to the patient's clinical information's (age, sex, other diseases ...etc.). Our main idea is argued by the probabilistic aspect chosen in the decision making of diagnosis process. This aspect will be translated as a probabilistic decision model. Our work is tested in a several medical cases that were collected from Sahloul Hospital. Performance indices of experiments are promising.