Automated extraction of left atrial volumes from two-dimensional computer tomography images using a deep learning technique. (1st October 2020)

• Record Type: Journal Article
• Authors: Chen, Hung-Hsun; Liu, Chih-Min; Chang, Shih-Lin; Chang, Paul Yu-Chun; Chen, Wei-Shiang; Pan, Yo-Ming; Fang, Ssu-Ting; Zhan, Shan-Quan; Chuang, Chieh-Mao; Lin, Yenn-Jiang; Kuo, Ling; Wu, Mei-Han; Chen, Chun-Ku; Chang, Ying-Yueh; Shiu, Yang-Che; Chen, Shih-Ann; Lu, Henry Horng-Shing
• Is Part Of: International journal of cardiology. Volume 316(2020)
• Page Start: 272
• Page End: 278
• Publication Date: 2020-10-01
• Subjects: Atrial fibrillation -- Deep learning -- Artificial intelligence -- Left atrium -- Segmentation; AF atrial fibrillation -- AUC area under the curve -- BSA body surface area -- CCHIA Collaboration Center of Health Information Application -- CFAE complex fractionated atrial electrogram -- CI confidence interval -- CNN convolutional neural network -- CS coronary sinus -- CT computed tomography -- HR hazard ratio -- ICD International Classification of Disease -- ILSVRC ImageNet Large Scale Visual Recognition Challenge -- IoU intersection over union -- LA left atrium -- LV left ventricle -- NPV non-pulmonary vein -- PV pulmonary vein -- PVCT pulmonary vein computed tomography -- PVI pulmonary vein isolation -- ReLU rectified linear unit -- ROC receiver operating characteristics -- 2D two-dimensional -- 3D three-dimensional; Cardiology -- Periodicals; Electronic journals; 616.12
• DOI: 10.1016/j.ijcard.2020.03.075
• Languages: English
• ISSNs: 0167-5273

Abstract

Abstract: Background: Precise segmentation of the left atrium (LA) in computed tomography (CT) images constitutes a crucial preparatory step for catheter ablation in atrial fibrillation (AF). We aim to apply deep convolutional neural networks (DCNNs) to automate the LA detection/segmentation procedure and create three-dimensional (3D) geometries. Methods: Five hundred eighteen patients who underwent procedures for circumferential isolation of four pulmonary veins were enrolled. Cardiac CT images (from 97 patients) were used to construct the LA detection and segmentation models. These images were reviewed by the cardiologists such that images containing the LA were identified/segmented as the ground truth for model training. Two DCNNs which incorporated transfer learning with the architectures of ResNet50/U-Net were trained for image-based LA classification/segmentation. The LA geometry created by the deep learning model was correlated to the outcomes of AF ablation. Results: The LA detection model achieved an overall 99.0% prediction accuracy, as well as a sensitivity of 99.3% and a specificity of 98.7%. Moreover, the LA segmentation model achieved an intersection over union of 91.42%. The estimated mean LA volume of all the 518 patients studied herein with the deep learning model was 123.3 ± 40.4 ml. The greatest area under the curve with a LA volume of 139 ml yielded a positive predictive value of 85.5% without detectable AF episodes over a period of one year following ablation. Conclusions: The deep learning provides an efficient and accurate way for automatic contouring and LA volume calculation based on the construction of the 3D LA geometry. Highlights: Deep learning provides an accurate way for automatic contouring of the left atrium. Left atrial volume computed by deep learning predicts AF recurrence after ablation. Left atrial volume could predict the one-year AF recurrence after ablation.