Prediction of heart sound using the xLSTM method for hypertrophic cardiomyopathy
-
Shatiswaran Vigian
Faculty Computing Informatics, Multimedia University, Cyberjaya 63100, Malaysia
-
R Kanesaraj Ramasamy
Faculty Computing Informatics, Multimedia University, Cyberjaya 63100, Malaysia
-
Junaidi Abdullah
Faculty Computing Informatics, Multimedia University, Cyberjaya 63100, Malaysia
Abstract
Heart disease has emerged as a major global public health concern, driven by poor dietary habits, unhealthy lifestyle choices, and limited health awareness. Accurate diagnosis of cardiac conditions remains challenging for hospitals and clinical institutions. Hypertrophic cardiomyopathy is an autosomal dominant disorder caused by mutations in sarcomere protein genes that affect the contractile function of cardiac muscle. With the growing adoption of digital health systems, large volumes of patient data are now collected and stored, providing opportunities for computational approaches to support clinical decision making. Machine learning methods have become increasingly important for analysing complex and nonlinear patterns in medical data. This study presents an ensemble-based approach for heart sound classification and introduces an Extended Long Short-Term Memory (xLSTM) model for the detection of cardiac abnormalities. The method was evaluated using acoustic features extracted from phonocardiogram recordings. The proposed model achieved 96.93% accuracy, 93.50% sensitivity, and 99.63% specificity, indicating strong performance in distinguishing normal and abnormal heart sounds. In comparison with previously reported techniques, the ensemble strategy and the xLSTM architecture provided improved accuracy. Model performance was assessed using accuracy, precision, recall, and F1 score, confirming the effectiveness of the proposed approach for automated heart sound analysis.
Copyright (c) 2026 Shatiswaran Vigian, R Kanesaraj Ramasamy, Junaidi Abdullah
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
[1]King J, Lowery DR. Physiology, Cardiac Output. StatPearls Publishing; 2023. Available online: https://pubmed.ncbi.nlm.nih.gov/29262215/
[2]Powell-Wiley TM, Poirier P, Burke LE, et al. Obesity and Cardiovascular Disease: A Scientific Statement From the American Heart Association. Circulation. 2021; 143(21). doi: 10.1161/CIR.0000000000000973
[3]Reddy RK, Charles WN, Sklavounos A, et al. The effect of smoking on COVID‐19 severity: A systematic review and meta‐analysis. Journal of Medical Virology. 2021; 93(2): 1045–1056. doi: 10.1002/jmv.26389
[4]Iglesias MJ, Kruse LD, Sanchez-Rivera L, et al. Identification of Endothelial Proteins in Plasma Associated With Cardiovascular Risk Factors. Arteriosclerosis, Thrombosis, and Vascular Biology. 2021; 41(12): 2990–3004. doi: 10.1161/ATVBAHA.121.316779
[5]Nishiga M, Wang DW, Han Y, et al. COVID-19 and cardiovascular disease: from basic mechanisms to clinical perspectives. Nature Reviews Cardiology. 2020; 17(9): 543–558. doi: 10.1038/s41569-020-0413-9
[6]McPherson R, Tybjaerg-Hansen A. Genetics of Coronary Artery Disease. Circulation Research. 2016; 118(4): 564–578. doi: 10.1161/CIRCRESAHA.115.306566
[7]Nakao T, Natarajan P. Clonal hematopoiesis, multi-omics and coronary artery disease. Nature Cardiovascular Research. 2022; 1(11): 965–967. doi: 10.1038/s44161-022-00154-1
[8]Bouril A, Aleinikava D, Guillem MS, et al. Automated Classification of Normal and Abnormal Heart Sounds using Support Vector Machines. In: Proceedings of the 2016 Computing in Cardiology Conference; 11–14 September 2016; Vancouver, BC, Canada. doi: 10.22489/CinC.2016.158-329
[9]Dao AT. Wireless laptop-based phonocardiograph and diagnosis. PeerJ. 2015; 3: e1178. doi: 10.7717/peerj.1178
[10]Cester L, Starshynov I, Jones Y, et al. Remote laser-speckle sensing of heart sounds for health assessment and biometric identification. Biomedical Optics Express. 2022; 13(7): 3743. doi: 10.1364/BOE.451416
[11]Rahmawati D. Accuracy of the NADAS Criteria in Helping Establish the Diagnosis in Children With Suspected to Have Congenital Heart Disease. Paediatrica Indonesiana. 2023; 63(4): 267–73. doi: 10.14238/pi63.4.2023.267-73
[12]Hochreiter S, Schmidhuber J. Long Short-Term Memory. Neural Computation. 1997; 9(8): 1735–1780. doi: 10.1162/neco.1997.9.8.1735
[13]Hochreiter S, Heusel M, Obermayer K. Fast model-based protein homology detection without alignment. Bioinformatics. 2007; 23(14): 1728–1736. doi: 10.1093/bioinformatics/btm247
[14]Schmidhuber J. Deep learning in neural networks: An overview. Neural Networks. 2015; 61: 85–117. doi: 10.1016/j.neunet.2014.09.003
[15]Vaswani A, Shazeer N, Parmar N, et al. Attention Is All You Need. arXiv preprint. 2023. doi: 10.48550/arXiv.1706.03762
[16]Graves A, Jaitly N, Mohamed A. Hybrid speech recognition with Deep Bidirectional LSTM. In: Proceedings of the 2013 IEEE Workshop on Automatic Speech Recognition and Understanding; 8–12 December 2013; Olomouc, Czech Republic. pp. 273–278. doi: 10.1109/ASRU.2013.6707742
[17]Karpathy A, Johnson J, Fei-Fei L. Visualizing and Understanding Recurrent Networks. arXiv preprint. 2015. doi: 10.48550/arXiv.1506.02078
[18]Sutskever I, Vinyals O, Le QV. Sequence to Sequence Learning with Neural Networks. arXiv preprint. 2014. doi: 10.48550/arXiv.1409.3215
[19]Zaremba W, Sutskever I. Learning to Execute. arXiv preprint. 2015. doi: 10.48550/arXiv.1410.4615
[20]Karpathy A, Fei-Fei L. Deep visual-semantic alignments for generating image descriptions. In: Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 7–12 June 2015; Boston, MA, USA. pp. 3128–3137. doi: 10.1109/CVPR.2015.7298932
[21]Hossain MZ, Sohel F, Shiratuddin MF, et al. A Comprehensive Survey of Deep Learning for Image Captioning. ACM Computing Surveys. 2019; 51(6): 1–36. doi: 10.1145/3295748
[22]Kratzert F, Klotz D, Brenner C, et al. Rainfall–runoff modelling using Long Short-Term Memory (LSTM) networks. Hydrology and Earth System Sciences. 2018; 22(11): 6005–6022. doi: 10.5194/hess-22-6005-2018
[23]Kratzert F, Klotz D, Herrnegger M, et al. Toward Improved Predictions in Ungauged Basins: Exploiting the Power of Machine Learning. Water Resources Research. 2019; 55(12): 11344–11354. doi: 10.1029/2019WR026065
[24]Nearing G, Cohen D, Dube V, et al. Global prediction of extreme floods in ungauged watersheds. Nature. 2024; 627(8004): 559–563. doi: 10.1038/s41586-024-07145-1
[25]Lokman A, Ramasamy RK. Smart Toilet: Threats and Challenges Identifying Human Presence Using IoT Sensors. In: Proceedings of the 3rd International Conference on Big Data and Internet of Things; 22–24 August 2019; Melbourne VIC Australia. pp. 56–60. doi: 10.1145/3361758.3361765
[26]Shinde R, Lavanya V. Heart Disease Prediction using Jellyfish Search Optimization and Deep Residual Networks. International Journal of Electrical and Electronics Engineering. 2023; 10(5): 165–173. doi: 10.14445/23488379/IJEEE-V10I5P115
[27]Ainiwaer A, Hou WQ, Qi Q, et al. Deep learning of heart-sound signals for efficient prediction of obstructive coronary artery disease. Heliyon. 2024; 10(1): e23354. doi: 10.1016/j.heliyon.2023.e23354
[28]Ren J, Mao X, Fu S. Image-based flow decomposition using empirical wavelet transform. Journal of Fluid Mechanics. 2021; 906: A22. doi: 10.1017/jfm.2020.817
[29]Tolic A, Boshkoska BM, Skansi S. Chrono Initialized LSTM Networks With Layer Normalization. IEEE Access. 2024; 12: 115219–115236. doi: 10.1109/ACCESS.2024.3445329
[30]Ghosh K, Bellinger C, Corizzo R, et al. The class imbalance problem in deep learning. Machine Learning. 2024; 113(7): 4845–4901. doi: 10.1007/s10994-022-06268-8
[31]Fernando T, Gammulle H, Denman S, et al. Deep Learning for Medical Anomaly Detection—A Survey. ACM Computing Surveys. 2022; 54(7): 1–37. doi: 10.1145/3464423
[32]Amiriparian S, Schmitt M, Ottl S, et al. Deep Unsupervised Representation Learning for Audio-Based Medical Applications. In: Deep Learners and Deep Learner Descriptors for Medical Applications, Intelligent Systems Reference Library. Springer International Publishing; 2020. pp. 137–164. doi: 10.1007/978-3-030-42750-4_5
[33]Vinay NA, Vidyasagar KN, Rohith S, et al. An RNN-Bi LSTM Based Multi Decision GAN Approach for the Recognition of Cardiovascular Disease (CVD) From Heart Beat Sound: A Feature Optimization Process. IEEE Access. 2024; 12: 65482–65502. doi: 10.1109/ACCESS.2024.3397574
[34]Aji NB, Kurnianingsih K, Masuyama N, et al. CNN-LSTM for Heartbeat Sound Classification. JOIV : International Journal on Informatics Visualization. 2024; 8(2): 735. doi: 10.62527/joiv.8.2.2115
[35]Rizal A, Fauzi H, Wulandari DR, et al. Classification of Normal and Abnormal Heart Sounds Using Hjorth Features and LSTM – RNN Algorithm. Journal of Physics: Conference Series. 2023; 2673(1): 012003. doi: 10.1088/1742-6596/2673/1/012003
[36]Zeinali Y, Niaki STA. Heart sound classification using signal processing and machine learning algorithms. Machine Learning with Applications. 2022; 7: 100206. doi: 10.1016/j.mlwa.2021.100206
[37]Latifi SA, Ghassemian H, Imani M. Multi-branch convolutional network and LSTM-CNN for heart sound classification. Physical and Engineering Sciences in Medicine. 2025; 49(1): 263–287. doi: 10.1007/s13246-025-01664-5
[38]Jing H, Jie R, Siqi L, et al. A cardiac audio classification method based on image expression of multidimensional features. Scientific Reports. 2024; 14(1): 23196. doi: 10.1038/s41598-024-73237-7
[39]Deng M, Meng T, Cao J, et al. Heart sound classification based on improved MFCC features and convolutional recurrent neural networks. Neural Networks. 2020; 130: 22–32. doi: 10.1016/j.neunet.2020.06.015
[40]Nogueira DM, Ferreira CA, Gomes EF, et al. Classifying Heart Sounds Using Images of Motifs, MFCC and Temporal Features. Journal of Medical Systems. 2019; 43(6): 168. doi: 10.1007/s10916-019-1286-5
[41]Khan FA, Abid A, Khan MS. Automatic heart sound classification from segmented/unsegmented phonocardiogram signals using time and frequency features. Physiological Measurement. 2020; 41(5): 055006. doi: 10.1088/1361-6579/ab8770
[42]Chen W, Sun Q, Chen X, et al. Deep Learning Methods for Heart Sounds Classification: A Systematic Review. Entropy. 2021; 23(6): 667. doi: 10.3390/e23060667
