A new similarity measure based on Bayesian Network signature correspondence for brain tumors cases retrieval

Case retrieval constitutes an interesting area of research which contributes to the evolution of several domains. The similarity measure module is a fundamental step in the retrieval process which affects remarkably on a retrieval system. In this context, we suggest in this paper a similarity measure applied to brain tumor cases retrieval. The rationale behind the proposed measure consists in quantifying the diagnosis correspondence followed by a clinician while comparing two medical cases. Our idea is characterized by the use of the Bayesian inference in the formulation of the proposed measure. The Bayesian network is applied in the classification task and it describes the decision-making process of a radiologist facing a tumor. The proposed similarity algorithm is based essentially on graph correspondence based on signature nodes comparison from the Bayesian classifiers. experiments were directed to compare the performance of the proposed similarity measure method with classical methods of similarity quantification. The performance indices of our proposition are promising.