3261

View Abstract / View Presentation

A comparative study between multi-view 2D CNN and multi-view 3D anisotropic CNN for brain tumor segmentation

Ritu Lahoti¹, Lakshay Agarwal¹, Neelam Sinha¹, and Vinod Reddy¹
¹International Institute of Information Technology Bangalore (IIITB), Bengaluru, India

On comparing the performance of multi-view 2D CNN and 3D anisotropic CNN on BraTS dataset for brain tumor segmentation, we report using four metrics that anisotropic CNN outperforms 2D CNN with higher sensitivity as it takes into account both global and local features efficiently.

Fig. 1. Flow diagram for first approach: Multi-view 2D CNN

Fig. 2. Flow diagram for second approach: Multi-view 3D anisotropic CNN