Regular SE(3) Group Convolutions for Volumetric Medical Image Analysis
This addresses the need for better model performance and equivariance in medical imaging tasks, though it is incremental based on prior work on separable group convolutions.
The paper tackled the problem of achieving SE(3) (roto-translation) equivariance in volumetric medical image analysis by developing a separable group convolution kernel, resulting in up to a 16.5% accuracy gain over regular CNNs and improved generalization.
Regular group convolutional neural networks (G-CNNs) have been shown to increase model performance and improve equivariance to different geometrical symmetries. This work addresses the problem of SE(3), i.e., roto-translation equivariance, on volumetric data. Volumetric image data is prevalent in many medical settings. Motivated by the recent work on separable group convolutions, we devise a SE(3) group convolution kernel separated into a continuous SO(3) (rotation) kernel and a spatial kernel. We approximate equivariance to the continuous setting by sampling uniform SO(3) grids. Our continuous SO(3) kernel is parameterized via RBF interpolation on similarly uniform grids. We demonstrate the advantages of our approach in volumetric medical image analysis. Our SE(3) equivariant models consistently outperform CNNs and regular discrete G-CNNs on challenging medical classification tasks and show significantly improved generalization capabilities. Our approach achieves up to a 16.5% gain in accuracy over regular CNNs.