Xu Shengzhou, Lu Haoran, Cheng Shiyu, Pei Chengdan
College of Computer Science, South-Central Minzu University, Wuhan 430074, China.
Network Information Center, Wuhan Institute of Technology, Wuhan 430205, China.
Int J Biomed Imaging. 2022 Jul 8;2022:8669305. doi: 10.1155/2022/8669305. eCollection 2022.
Cardiovascular diseases are reported as the leading cause of death around the world. Automatic segmentation of the left ventricle (LV) from magnetic resonance (MR) images is essential for an early diagnosis. An enhanced ResUnet is proposed in this paper to improve the performance of extracting LV endocardium and epicardium from MR images, improving the accuracy of the model by introducing a medium skip connection for the contracting path and a short skip connection for the residual unit. Also, a depth-wise separable convolution replaces the typical convolution operation to improve training efficiency. In the MICCAI 2009 LV segmentation challenge test dataset, the percentages of "good" contours, dice metric, and average perpendicular distance of endocardium (epicardium) are 99.12% ± 2.29%(100% ± 0%), 0.93 ± 0.02 (0.96 ± 0.01), and 1.60 ± 0.42 mm (1.37 ± 0.23 mm), respectively. Experimental results demonstrate that the proposed model obtains promising performance and outperforms state-of-the-art methods. By incorporating these various skip connections, the segmentation accuracy of the model is significantly improved, while the depth-wise separable convolution also improves the model efficiency.
据报道，心血管疾病是全球主要的死亡原因。从磁共振（MR）图像中自动分割左心室（LV）对于早期诊断至关重要。本文提出了一种增强型ResUnet，以提高从MR图像中提取LV心内膜和心肌外膜的性能，通过为收缩路径引入中间跳跃连接和为残差单元引入短跳跃连接来提高模型的准确性。此外，深度可分离卷积取代了典型的卷积操作以提高训练效率。在MICCAI 2009 LV分割挑战测试数据集中，“良好”轮廓的百分比、骰子系数以及心内膜（心肌外膜）的平均垂直距离分别为99.12%±2.29%（100%±0%）、0.93±0.02（0.96±0.01）和1.60±0.42毫米（1.37±0.23毫米）。实验结果表明，所提出的模型取得了良好的性能，优于现有方法。通过结合这些不同的跳跃连接，模型的分割精度得到了显著提高，而深度可分离卷积也提高了模型效率。
