Cloud-Based Healthcare Monitoring System using CNN-LSTM for Early Detection of Patient Health Risks

Continuous monitoring of physiological parameters is increasingly important for patients who require observation beyond conventional hospital settings. Traditional monitoring approaches often depend on periodic measurements and manual review, which can delay recognition of abnormal patterns. This paper proposes a cloud-based healthcare monitoring system that combines Internet of Things (IoT) data acquisition, cloud storage, automated preprocessing, and a convolutional neural network–long short-term memory (CNN-LSTM) model for early health-risk assessment. The proposed framework accepts time-dependent physiological observations such as heart rate, oxygen saturation, body temperature, blood pressure, respiratory rate, and electrocardiographic signals when available. A cloud layer provides centralized ingestion, storage, access control, and dashboard connectivity, while the deep-learning layer learns local signal representations and temporal dependencies. The system is designed as a research framework rather than a clinically validated diagnostic device; therefore, no fabricated patient results or accuracy values are reported. A methodology for dataset preparation, windowing, normalization, model training, evaluation, alert generation, and deployment is defined. The paper also discusses data quality, privacy, latency, interoperability, false alarms, explainability, and the limitations of cloud-dependent healthcare applications. Public physiological repositories such as PhysioNet can be used in future experimental validation, while real clinical deployment would require appropriate governance, validation, and regulatory review. The proposed architecture provides a practical foundation for scalable remote patient monitoring and deep-learning-assisted clinical decision support.

S.Brindha Devi. (2026).

Cloud-Based Healthcare Monitoring System using CNN-LSTM for Early Detection of Patient Health Risks

International Journal of Current Science Research (IJCSR)

e-ISSN: 2454-5422

12(7), 10-25.

1. World Health Organization Regional Office for Europe, “Scaling up telemedicine in the WHO European Region,” Policy Brief, 2025.
2. R. S. H. Istepanian, S. Hu, N. Philip, and M. Sungoor, “The potential of Internet of Things (IoT) for healthcare,” in Proc. 2011 5th Int. Conf. on Pervasive Computing Technologies for Healthcare, 2011.
3. A. Islam, M. M. Rahman, M. A. Rahman, M. A. Hossain, and M. M. Hossain, “Internet of Things for Health Care: A Comprehensive Survey,” IEEE Access, vol. 3, pp. 678–708, 2015.
4. S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.
5. Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, pp. 436–444, 2015.
6. G. B. Moody and R. G. Mark, “The impact of the MIT-BIH Arrhythmia Database,” IEEE Engineering in Medicine and Biology Magazine, vol. 20, no. 3, pp. 45–50, 2001.
7. PhysioNet, “MIT-BIH Long-Term ECG Database, version 1.0.0,” PhysioNet, DOI: 10.13026/C2KS3F.
8. PhysioNet, “PhysioNet Databases,” MIT Laboratory for Computational Physiology, accessed 2026.
9. J. Gubbi, R. Buyya, S. Marusic, and M. Palaniswami, “Internet of Things (IoT): A vision, architectural elements, and future directions,” Future Generation Computer Systems, vol. 29, no. 7, pp. 1645–1660, 2013.
10. V. M. R. Penmetsa and S. S. K. Venkatesh, “Cloud computing in healthcare: A review,” International Journal of Computer Applications, vol. 116, no. 1, 2015.
11. V. Rajkomar, J. Dean, and I. Kohane, “Machine learning in medicine,” New England Journal of Medicine, vol. 380, pp. 1347–1358, 2019.