Enhanced classification of left ventricular hypertrophy in cardiac patients using extended Siamese CNN

• Published in: Physics in medicine & biology Vol. 69; no. 14; pp. 145001 - 145017
• Main Authors: Farhad, Moomal, Masud, Mohammad Mehedy, Beg, Azam, Ahmad, Amir, Memon, Sehar
• Format: Journal Article
• Language: English
• Published: England IOP Publishing 02.07.2024
• Subjects: Echocardiography; Euclidean distance; few-shot learning; Humans; Hypertrophy, Left Ventricular - diagnostic imaging; Hypertrophy, Left Ventricular - physiopathology; Image Processing, Computer-Assisted - methods; left ventricular hypertrophy; Neural Networks, Computer; zero-shot learning; left ventricular hypertrophy; Euclidean distance; echocardiography; zero-shot learning; few-shot learning
• ISSN: 0031-9155; 1361-6560; 1361-6560
• DOI: 10.1088/1361-6560/ad548a

Left ventricular hypertrophy (LVH) is the thickening of the left ventricle wall of the heart. The objective of this study is to develop a novel approach for the accurate assessment of LVH) severity, addressing the limitations of traditional manual grading systems. We propose the Multi-purpose Siamese Weighted Euclidean Distance Model (MSWED), which utilizes convolutional Siamese neural networks and zero-shot/few-shot learning techniques. Unlike traditional methods, our model introduces a cutoff distance-based approach for zero-shot learning, enhancing accuracy. We also incorporate a weighted Euclidean distance targeting informative regions within echocardiograms. We collected comprehensive datasets labeled by experienced echocardiographers, including Normal heart and various levels of LVH severity. Our model outperforms existing techniques, demonstrating significant precision enhancement, with improvements of up to 13% for zero-shot and few-shot learning approaches. Accurate assessment of LVH severity is crucial for clinical prognosis and treatment decisions. Our proposed MSWED model offers a more reliable and efficient solution compared to traditional grading systems, reducing subjectivity and errors while providing enhanced precision in severity classification.