On the Compactness, Efficiency, and Representation of 3D Convolutional Networks: Brain Parcellation as a Pretext Task
This work addresses the need for efficient and flexible networks in medical imaging, particularly for brain MR image analysis, but it is incremental as it builds on existing building blocks like dilated convolution and residual connections.
The authors tackled the challenge of designing efficient deep architectures for volumetric medical image segmentation by proposing a high-resolution, compact 3D convolutional network using dilated convolution and residual connections, achieving results comparable to state-of-the-art methods while being an order of magnitude more compact in brain parcellation tasks.
Deep convolutional neural networks are powerful tools for learning visual representations from images. However, designing efficient deep architectures to analyse volumetric medical images remains challenging. This work investigates efficient and flexible elements of modern convolutional networks such as dilated convolution and residual connection. With these essential building blocks, we propose a high-resolution, compact convolutional network for volumetric image segmentation. To illustrate its efficiency of learning 3D representation from large-scale image data, the proposed network is validated with the challenging task of parcellating 155 neuroanatomical structures from brain MR images. Our experiments show that the proposed network architecture compares favourably with state-of-the-art volumetric segmentation networks while being an order of magnitude more compact. We consider the brain parcellation task as a pretext task for volumetric image segmentation; our trained network potentially provides a good starting point for transfer learning. Additionally, we show the feasibility of voxel-level uncertainty estimation using a sampling approximation through dropout.