Fully Automatic 3D Bi-Atria Segmentation from Late Gadolinium-Enhanced MRIs Using Double Convolutional Neural Networks
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
- Auckland Bioengineering Institute, University of Auckland
- The University of Auckland
Segmentation of the 3D human atria from late gadolinium-enhanced (LGE)-MRIs is crucial for understanding and analyzing the underlying atrial structures that sustain atrial fibrillation (AF), the most common cardiac arrhythmia. However, due to the lack of a large labeled dataset, current automated methods have only been developed for left atrium (LA) segmentation. Since AF is sustained across both the LA and right atrium (RA), an automatic bi-atria segmentation method is of high interest. We have therefore created a 3D LGE-MRI database from AF patients with both LA and RA labels to train a double, sequentially used convolutional neural network (CNN) for automatic LA and RA epicardium and endocardium segmentation. To mitigate issues regarding the severe class imbalance and the complex geometry of the atria, the first CNN accurately detects the region of interest (ROI) containing the atria and the second CNN performs targeted regional segmentation of the ROI. The CNN comprises of a U-Net backbone enhanced with residual blocks, pre-activation normalization, and a Dice loss to improve accuracy and convergence. The receptive field of the CNN was increased by using 5 × 5 kernels to capture large variations in the atrial geometry. Our algorithm segments and reconstructs the LA and RA within 2 s, achieving a Dice accuracy of 94% and a surface-to-surface distance error of approximately 1 pixel. To our knowledge, the proposed approach is the first of its kind, and is currently the most robust automatic bi-atria segmentation method, creating a solid benchmark for future studies.
- Atrial segmentation, Convolutional neural network, MRI