Unification of heterogeneous data towards the prediction of oral cancer reoccurrence

摘要: Oral cancer is the predominant neoplasm of head and neck. Annually, more than 500.000 new cases oral are reported, worldwide. After initial treatment its complete disappearance, a state called remission, reoccurrence rates still remain quite high early identification such relapses matter great importance. Up to now, several approaches have been proposed for this purpose yielding however, unsatisfactory results. This mainly attributed fragmented nature these studies which took into account only limited subset factors involved in development cancer. In work we propose unified orchestrated approach based on Dynamic Bayesian Networks (DBNs) prediction after disease has reached remission. Several heterogeneous data sources featuring clinical, imaging genomic information assembled analyzed over time, order procure informative biomarkers correlate with progression identify potential (local or metastatic) disease.