2D/3D medical image segmentation by embedding EfficientNet in Convolutional neural network

Abstract

3D médical image Processing with deep learning greatly suffers from a lack of data. Thus, studies carried out in this field are limited compared to 2D image analysis related works, where very large datasets exist. As a result, powerful and efficient 2D convolutional neural networks have been developed and trained. In this work, we investigate the way to transfer the performance of a two-dimensional classification network for the purpose of three-dimensional semantic segmentation of brain tumors. We propose an asymmetric U-Net network by integrating the EfficientNet model as part of the encoding branch. As the input data is in 3D, the first layers of the encoder are devoted to the reduction of the third dimension in order to fit the input of the EfficientNet network. Experimental results on validation data from the BraTS 2020 challenge demonstrate that the proposed method achieve promising performance.