MRI-SPECT DATA FUSION FOR TEMPORAL LOBE EPILEPSY SURGERY CANDIDATE SELECTION

摘要

This paper presents a data fusion algorithm in a decision-support system to identify potential candidates for surgery in temporal lobe epilepsy. To this end, multimodality images including magnetic resonance imaging (MRI) and single photon emission computed tomography (SPECT) are used to predict surgery outcome. Effective features such as hippocampus structure and texture are extracted and combined to make reliable decisions. The experimental results using a support vector machine classifier show that the proposed approach may reliably predict the surgery outcome.