Segmentation of the left ventricle in cardiac MRI based on convolutional neural network and level set function
- PDF / 1,796,559 Bytes
- 8 Pages / 595.224 x 790.955 pts Page_size
- 65 Downloads / 178 Views
ORIGINAL PAPER
Segmentation of the left ventricle in cardiac MRI based on convolutional neural network and level set function Ali Rostami1 · Mehdi Chehel Amirani1 · Hossein Yousef-Banaem2 Received: 23 May 2020 / Accepted: 29 June 2020 © IUPESM and Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Left ventricular segmentation in cardiac magnetic resonance images is considered as the most critical method to evaluate the cardiac function. In this paper, a hybrid method has been proposed to segment the left ventricle(endocardium). In this study, first, a new method has been proposed based on deep Convolutional Neural Network (CNN) to localize the LV in cardiac MRI. Then, the segmentation is completed through the localized LV by level set function. After segmentation, for each patient end-systole volume, end-diastole volume and ejection fraction are calculated to evaluate the left ventricle function. The evaluation of segmentation is done by specificity, sensitivity, accuracy, Average Perpendicular Distance (APD), and Dice indexes. According to obtain results for the proposed method, the mean specificity, sensitivity, and accuracy were 99.35, 94.11, and 94. Based on the results, the presented method is very reliable for the segmentation of the left ventricles and evaluation of the cardiac function. Keywords Left ventricle · Segmentation · CNN · Level set
1 Introduction Since cardiovascular disease is one of the most common causes of death in the world, the study of left ventricular structure and performance is significant in cardiac disturbances management. So, controlling and treating disturbances are one of the most important challenges faced by healthcare professionals [1, 2].To evaluate Heart function, often, MRI images are used because they have the advantages of low-risk and high tissue contrast [3]. Manual segmentation is a common method of left ventricle segmentation; however, this method is a boring and time-consuming task, and it has potential errors [4]. Ali Rostami
Ali [email protected] Mehdi Chehel Amirani [email protected] Hossein Yousef-Banaem [email protected] 1
Department of Electrical and Computer Engineering, Urmia University, Urmia, Iran
2
Skull Base Research Center, Loghman Hakim Hospital, Tehran, Iran
With the help of the left ventricle segmentation, most of the severe cardiac disease, such as myocardial infarction (MI), left ventricular hypertrophy (LVH) and ischemia can be detected. However, there are some problems in the process of segmentation as follows. – – – – – –
Papillary muscle (The papillary muscles are muscles located in the ventricles of the heart.) ventricular appearance. ventricular motion and displacement. Changes in heart size that varies between different people. a tiny volume of the myocardium. the boundary of endocardium and epicardium that are strictly bound together [5, 6].
A few automatic or semi-automatic segmentation methods have been developed in recent decades, which are either based on low-level image data or high-level models
Data Loading...
