A performance comparison of the bayesian graphical model and the Possibilistic graphical model applied in a brain MRI cases retrieval contribution
Hedi Yazid, Karim Kalti, Najoua Essoukri BenAmara · 2013
This paper proposes a comparison between the Bayesian networks and the Possibilistic networks facing the treatment of a similarity measurement problem. The proposed similarity measure is incorporated in brain tumors MRI cases retrieval contribution. Both methods represent an interesting way in the treatment of the computer aided decision problems. Our main idea is argued by the uncertain aspect embodied in the decision making of the diagnosis process. This aspect is translated into a graphical modelling of the treated study framework that is concretized by the two models mentioned above. Our work is tested on several medical cases collected from Sahloul Hospital. Experiments are oriented to analyse the performance of both models while testing experimentations with missing data.