Comparative analysis of active contour and convolutional neural network in rapid left-ventricle volume quantification using echocardiographic imaging. (February 2021)

• Record Type: Journal Article
• Authors: Zhu, Xiliang; Wei, Yang; Lu, Yu; Zhao, Ming; Yang, Ke; Wu, Shiqian; Zhang, Hui; Wong, Kelvin K.L.
• Is Part Of: Computer methods and programs in biomedicine. Volume 199(2021)
• Publication Date: 2021-02
• Subjects: Cardiac ultrasonography -- Intra-operative ultrasound -- left ventricle -- Active contour -- Convolutional neural network; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2020.105914
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: Develop robust segmentation techniques to segment left ventricle (LV) from ultrasound images. Use machine learning (ML) techniques such as active contour (AC) and convolutional neural network (CNN) for segmentation. Perform comparative analysis in terms of the ventricular area (VA), the ventricular maximum diameter (VMXD), the ventricular minimum diameter (VMID) and the ventricular long axis angle (AVLA). Compare automatic and visual approaches using the Dice similarity coefficient, Jaccard indices and Hausdorff distance to estimate the agreement of LV segmented. Prove the underlying techniques of LV segmentation is useful and practical. Abstract: In cardiology, ultrasound is often used to diagnose heart disease associated with myocardial infarction. This study aims to develop robust segmentation techniques for segmenting the left ventricle (LV) in ultrasound images to check myocardium movement during heartbeat. The proposed technique utilizes machine learning (ML) techniques such as the active contour (AC) and convolutional neural networks (CNNs) for segmentation. Medical experts determine the consistency between the proposed ML approach, which is a state-of-the-art deep learning method, and the manual segmentation approach. These methods are compared in terms of performance indicators such as the ventricular area (VA), ventricular maximum diameter (VMXD), ventricular minimum diameter (VMID), and ventricular long axis angle (AVLA) measurements. Furthermore, the Dice similarity coefficient, Jaccard index, and Hausdorff distance are measured to estimate the agreement of the LV segmented results between the automatic and visual approaches. The obtained results indicate that the proposed techniques for LV segmentation are useful and practical. There is no significant difference between the use of AC and CNN in image segmentation; however, the AC method could obtain comparable accuracy as the CNN method using less training data and less run-time.