Binary segmentation of medical images using implicit spline representations and deep learning
This work addresses binary segmentation in medical imaging, specifically for congenital heart disease CT scans, representing an incremental improvement over existing methods.
The authors tackled medical image segmentation by combining implicit spline representations with deep learning, achieving a state-of-the-art average volumetric test Dice score of almost 92% on a congenital heart disease CT dataset.
We propose a novel approach to image segmentation based on combining implicit spline representations with deep convolutional neural networks. This is done by predicting the control points of a bivariate spline function whose zero-set represents the segmentation boundary. We adapt several existing neural network architectures and design novel loss functions that are tailored towards providing implicit spline curve approximations. The method is evaluated on a congenital heart disease computed tomography medical imaging dataset. Experiments are carried out by measuring performance in various standard metrics for different networks and loss functions. We determine that splines of bidegree $(1,1)$ with $128\times128$ coefficient resolution performed optimally for $512\times 512$ resolution CT images. For our best network, we achieve an average volumetric test Dice score of almost 92%, which reaches the state of the art for this congenital heart disease dataset.