A computer aided diagnosis ‘CAD’ for brain glioma exploration

Manual analysis of brain glioma tumor lacks accuracy and is time consuming. Thus, to avoid human error, this paper presents an automatic and accurate computer aided diagnosis (CAD) system for brain glioma exploration on magnetic resonance imaging. A preprocessing approach was proposed to eliminate extra-cerebral features. Tumor and even its edema was automatically extracted by a fast distribution matching algorithm. Then, we realize a 3D reconstruction for brain tumor glioma and its edema in order to classify glioma tumor based on their radiologic appearance. As discussed with clinicians, the experimental results showed that the proposed computer aided diagnosis for brain glioma tumor achieves a good agreement. Thus, the proposed tool facilitates and speeds up the analysis of data and supports decision making.