No abstract

The early diagnosis of cardiovascular diseases (CVDs) can effectively prevent them from worsening. The source of the disease can be effectively detected through analysis with cardiac magnetic resonance imaging (CMRI). The segmentation of the left ventricle (LV) in CMRI images plays an indispensable role in the diagnosis of CVDs. However, the automated segmentation of LV is a challenging task, as it is confused with neighboring regions in the cardiac MRI. Deep learning models are effective in performing such complex segmentation because of the high performing convolutional neural networks (CNN). However, since segmentation using CNN involves the pixel-level classification of the image, it lacks the contextual information that is highly desirable in analyzing medical images. In this research, we propose a modified U-Net model to accurately segment the LV using context-enabled segmentation. The proposed model achieves the automatic segmentation and quantitative assessment of LV. The proposed model achieves the state-of-the-art accuracy by effectively utilizing various hyperparameters, such as batch size, batch normalization, activation function, loss function and dropout. Our method demonstrated a statistical significance in the endo- and epicardial walls with a dice score of 0.96 and 0.93, respectively, an average perpendicular distance of 1.73 and percentage of good contours of 96.22 were achieved. Furthermore, a high positive correlation of 0.98 between the clinical parameters, such as ejection fraction, end diastolic volume (EDV), end systolic volume (ESV) and gold standard was obtained.

No abstract

In order to diagnose cardiovascular disease (CVD) in its early stages, the position of the Left Ventricle (LV) and relevant parameters associated with it plays significant role in the medical field. The timely diagnosis of CVDs works as a lifesaver in many cases. In the earlier days, the position and functioning of the LV was assessed by the error-prone manual methods. Nowadays, the newer and smart technologies have allowed the medical practitioners to make use of auto-segmentation methods for diagnosis of heart problems in early stages. It is difficult to assess the functioning of the LV due to some listed reasons a) a bigger span and changing the size of LV in MRI scanning , b) Varied myocardial and blood-pool fragments, c) Similarity in shape between the LV and other body organs and d) Noise in images. Hence assessing the LV for the accurate identification of echocardiographic parameters still remains the challenge for diagnosing CVDs. Many researchers deploy methods based on Machine learning (ML) and deep learning (DL) to get accurate results for LV segmentation (LVS0. It helps in segmenting the LV and revealing the clearer parts of the image for better classification and diagnosis. In this research study, three methods are deployed for the segmentation of LV images namely CNN based U-Net Model , VGG 16 and ResNet 152. These methods have been implemented for the segmentation of the images obtained from MRI scan to explore the position of left ventricle and problems in LV which leads to other lethal heart ailments. These approaches help in the identification of cardiac parameters related to CVDs. The proposed algorithms are compared using standard performance metrics to assess the output and viability of the projected techniques as discussed in the result section of this article. The Blockchain database has been considered as the source of input images and this research is applicable universally due to its nature of considering latest technologies to identify CVDs. The results of DL algorithms (DLA) reveal that the CNN-based U-Net Model outperforms the other two methods (VGG 16 and ResNet 152) for accurate identification of CVDs from the LV segmentation techniques.