The proposed approach: The computational approach is composed of four steps. First, the data is pre-processed. In each view feature with low variance are filtered out. Furthermore, the features are clustered in order to reduce the input dimension. From each cluster prototype are extracted. These prototypes are the only features used in following steps (a). Second, the prototypes are ranked by the patient class separability and the most significant ones are selected (b). Third, the patients are clustered and the membership matrices are obtained (c). Fourth, a late integration approach is utilized to integrate clustering results (d).