A study of Heart Stroke Sensing Using Machine Learning Techniques

In conclusion, predictions of stroke based on machine learning represent a promising application within healthcare systems. With diverse preprocessing techniques and algorithms, exact models may be derived to help identify individuals at a higher risk for a stroke. Thus, features, algorithms, and metrics for evaluation do make a difference in the performance of models. Overall, results indicate the possibility of timely interventions and probably lives saved through preventive healthcare employing machine learning techniques. Future studies might center on several improvements of stroke prediction models. Among the areas would be to generalize the types of datasets used for more varieties of datasets with higher variations of demographic and clinical variables. Advanced techniques such as deep learning or ensemble methods will lead to better performance in predictive results. Real-time data from wearable devices and mobile health applications are also another area, which can be thoroughly interesting in increasingly dynamically assessing risks regarding stroke or any chronic diseases. An extension of the study may be an operational model that will predict stroke risk and determine the risk factors. Such models may open gates for personalized preventive measures to the individual patients. Besides, the co-operation of the health workers while working on the models can be guaranteed to be clinically relevant and practical to apply in the field outside a research setting. More work can expand research beyond stroke to include other cardiovascular diseases to give an overall view of the stroke and heart problems issues.

[1]      Sivapalan, G., Nundy, K., Dev, S., Cardiff, B., & Deepu, J. ANNet- A lightweight neural network for ECG anomaly detection in IoT edge sensors. ‖IEEE Transactions on Biomedical Circuits and Systems, 2022. [2]      Koh, H.C., Tan, G., et al. Data mining applications in healthcare. ‖Journal of Healthcare Information Management ,  19(2), 65, 2011 [3]      Yoo, I., Alafaireet, P., Marinov, M., Pena-Hernandez, K., Gopidi, R., Chang, J.-F., & Hua, L. Data mining in healthcare and biomedicine: A survey of the literature. Journal of Medical Systems, 36(4), 2431-2448, 2012. [4]      Meschia, J.F., Bushnell, C., Boden-Albala, B., Braun, L.T., Bravata, D.M., Chaturvedi, S., Creager, M.A., Eckel, R.H., Elkind, M.S., Fornage, M., et al. Guidelines for the primary prevention of stroke: A statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke, 45(12), 3754-3832, 2014. [5]      Harmsen, P., Lappas, G., Rosengren, A., & Wilhelmsen, L. Long-term risk factors for stroke: Twenty-eight years of follow-up of 7457 middle-aged men in Goteborg, Sweden. Stroke, 37(7), 1663-1667, 2006 [6]      Nwosu, C.S., Dev, S., Bhardwaj, P., Veeravalli, B., & John, D. Predicting stroke from electronic health records. In 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (pp. 5704-5707). IEEE. 2019. [7]      Pathan, M.S., Jianbiao, Z., John, D., Nag, A., & Dev, S. Identifying stroke indicators using rough sets. IEEE Access, 8, 210318-210327, 2020 [8]      Jeena, R.S., & Kumar, S. Stroke prediction using SVM. In Proceedings of the International Conference on Control, Instrumentation, Communication and Computational Technologies (ICCICCT) (pp. 600-602). IEEE.2016. [9]      Hanifa S.-M., & Raja-S K. Stroke risk prediction through non-linear support vector classification models. International Journal of Advanced Research in Computer Science, 1(3), 2010.  [10]   Luk, J.K., Cheung, R.T., Ho, S., & Li, L. (2006). Does age predict outcome in stroke rehabilitation? A study of 878 Chinese subjects. Cerebrovascular Diseases, 21(4), 229-234.