AI-Driven Vital Parameter Analysis through CNN Architecture

C.Muthamilselvi; M.Vinodhini; M.Deepalakshmi; P.Sureka

doi:https://doi.org/10.47001/JAIET/2026.301003

Authors

C.Muthamilselvi

Department of Electronics and Communication Engg., Kumaraguru College of Technology, Coimbatore, TamilNadu, India

M.Vinodhini

Department of Electronics and Communication Engg., Kumaraguru College of Technology, Coimbatore, TamilNadu, India

M.Deepalakshmi

Department of Electronics and Communication Engg., Kumaraguru College of Technology, Coimbatore, TamilNadu, India

P.Sureka

Department of Electronics and Communication Engg., Kumaraguru College of Technology, Coimbatore, TamilNadu, India

Abstract

Continuous monitoring and accurate analysis of vital parameters are essential for early diagnosis, disease prevention, and effective patient management in modern healthcare systems. Traditional monitoring methods often rely on manual observation or threshold-based alert mechanisms, which may fail to capture complex physiological patterns. This research proposes an AI-Driven Vital Parameter Analysis through CNN Architecture, designed to enhance the accuracy and reliability of real-time health monitoring systems. The proposed framework integrates wearable or IoT-enabled biomedical sensors to collect physiological signals such as heart rate, blood pressure, respiratory rate, body temperature, and oxygen saturation. The collected time-series data undergo preprocessing steps including noise filtering, normalization, and segmentation before being fed into a Convolutional Neural Network (CNN). The CNN architecture automatically extracts hierarchical features from physiological signals, enabling precise classification of normal and abnormal health conditions. Unlike traditional machine learning models that require manual feature engineering, the CNN-based approach learns discriminative features directly from raw sensor inputs, improving diagnostic performance. Experimental evaluation demonstrates high classification accuracy, improved sensitivity, and reduced false alarm rates compared to conventional statistical and rule-based methods. The system supports real-time inference and can be deployed on edge devices or cloud platforms for remote patient monitoring. The proposed AI-driven framework offers scalability, reliability, and enhanced decision support for smart healthcare applications, contributing to proactive medical intervention and improved patient outcomes.

Keywords

Artificial Intelligence (AI) Convolutional Neural Network (CNN) Vital Sign Monitoring Physiological Signal Analysis Smart Healthcare Wearable Sensors Internet of Things (IoT) Real-Time Health Monitoring

Citation of this Article

C.Muthamilselvi, M.Vinodhini, M.Deepalakshmi, & P.Sureka. (2026). AI-Driven Vital Parameter Analysis through CNN Architecture. Journal of Artificial Intelligence and Emerging Technologies (JAIET). 3(1), 15-22. Article DOI: https://doi.org/10.47001/JAIET/2026.301003

References

Kiranyaz, S., Ince, T., & Gabbouj, M. (2015). Real-time patient-specific ECG classification by 1-D CNN. IEEE Transactions on Biomedical Engineering, 63(3), 664–675.
Rajpurkar, P., et al. (2017). Cardiologist-level arrhythmia detection with CNNs. Nature Medicine, 25, 65–69.
Acharya, U. R., et al. (2017). Deep CNN for arrhythmia detection. Information Sciences, 415–416, 190–198.
Hannun, A. Y., et al. (2019). Cardiologist-level arrhythmia detection and classification. Nature Medicine, 25(1), 65–69.
Yildirim, O. (2018). Arrhythmia detection using deep CNN model. Computers in Biology and Medicine, 102, 411–420.
Ribeiro, A. H., et al. (2020). Automatic ECG diagnosis using deep neural networks. Nature Communications, 11, 1760.
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521, 436–444.
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780.
Goodfellow, I., et al. (2016). Deep Learning. MIT Press.
WHO. (2021). Cardiovascular diseases (CVDs) fact sheet.
Clifford, G. D., et al. (2017). AF classification challenge. Physiological Measurement.
Zhang, Z., et al. (2017). Deep learning in biomedical signal analysis. IEEE Reviews in Biomedical Engineering.
Esteva, A., et al. (2017). Dermatologist-level classification with deep neural networks. Nature.
Faust, O., et al. (2018). Deep learning for healthcare applications. IEEE Reviews in Biomedical Engineering.
Al Rahhal, M. M., et al. (2016). Deep learning approach for arrhythmia detection. Information Sciences, 358–359, 262–275.
Osowski, S., Hoai, L. T., & Markiewicz, T. (2004). Support vector machine based expert system for reliable heartbeat recognition. IEEE Transactions on Biomedical Engineering, 51(4), 582–589.
Lagerholm, M., Peterson, C., Braccini, G., Edenbrandt, L., & Sornmo, L. (2000). Clustering ECG complexes using Hermite functions and self-organizing maps. IEEE Transactions on Biomedical Engineering, 47(7), 838–848.
Kiranyaz, S., Ince, T., & Gabbouj, M. (2015). Real-time patient-specific ECG classification by 1-D CNN. IEEE Transactions on Biomedical Engineering, 63(3), 664–675.
Rajpurkar, P., Hannun, A. Y., Haghpanahi, M., Bourn, C., & Ng, A. Y. (2017). Cardiologist-level arrhythmia detection with convolutional neural networks. Nature Medicine, 25, 65–69.
Yildirim, Ö. (2018). A novel wavelet sequence-based deep learning approach for ECG signal classification. Computer Methods and Programs in Biomedicine, 152, 55–64.
Zubair, M., et al. (2019). CNN-LSTM based framework for human activity recognition using wearable sensors. Journal of Healthcare Informatics Research, 3, 67–80.
Sharma, S., & Singh, B. (2020). IoT-based human health monitoring system using ECG data. International Journal of Advanced Computer Science and Applications, 11(1).
Al-Naffakh, M., et al. (2021). Deep learning for monitoring cardiovascular diseases using ECG and PPG data. IEEE Access, 9, 1032–1043.
Sannino, G., & De Pietro, G. (2020). A deep learning approach for ECG-based heartbeat classification for arrhythmia detection. Journal of Ambient Intelligence and Humanized Computing, 11, 1401–1411.
Faust, O., Hagiwara, Y., Hong, T. J., Lih, O. S., & Acharya, U. R. (2019). Deep learning for healthcare applications based on physiological signals: A review. Computer Methods and Programs in Biomedicine, 161, 1–13.

[1] Kiranyaz, S., Ince, T., & Gabbouj, M. (2015). Real-time patient-specific ECG classification by 1-D CNN. IEEE Transactions on Biomedical Engineering, 63(3), 664–675.

[2] Rajpurkar, P., et al. (2017). Cardiologist-level arrhythmia detection with CNNs. Nature Medicine, 25, 65–69.

[3] Acharya, U. R., et al. (2017). Deep CNN for arrhythmia detection. Information Sciences, 415–416, 190–198.

[4] Hannun, A. Y., et al. (2019). Cardiologist-level arrhythmia detection and classification. Nature Medicine, 25(1), 65–69.

[5] Yildirim, O. (2018). Arrhythmia detection using deep CNN model. Computers in Biology and Medicine, 102, 411–420.

[6] Ribeiro, A. H., et al. (2020). Automatic ECG diagnosis using deep neural networks. Nature Communications, 11, 1760.

[7] LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521, 436–444.

[8] Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780.

[9] Goodfellow, I., et al. (2016). Deep Learning. MIT Press.

[10] WHO. (2021). Cardiovascular diseases (CVDs) fact sheet.

[11] Clifford, G. D., et al. (2017). AF classification challenge. Physiological Measurement.

[12] Zhang, Z., et al. (2017). Deep learning in biomedical signal analysis. IEEE Reviews in Biomedical Engineering.

[13] Esteva, A., et al. (2017). Dermatologist-level classification with deep neural networks. Nature.

[14] Faust, O., et al. (2018). Deep learning for healthcare applications. IEEE Reviews in Biomedical Engineering.

[15] Al Rahhal, M. M., et al. (2016). Deep learning approach for arrhythmia detection. Information Sciences, 358–359, 262–275.

[16] Osowski, S., Hoai, L. T., & Markiewicz, T. (2004). Support vector machine based expert system for reliable heartbeat recognition. IEEE Transactions on Biomedical Engineering, 51(4), 582–589.

[17] Lagerholm, M., Peterson, C., Braccini, G., Edenbrandt, L., & Sornmo, L. (2000). Clustering ECG complexes using Hermite functions and self-organizing maps. IEEE Transactions on Biomedical Engineering, 47(7), 838–848.

[18] Kiranyaz, S., Ince, T., & Gabbouj, M. (2015). Real-time patient-specific ECG classification by 1-D CNN. IEEE Transactions on Biomedical Engineering, 63(3), 664–675.

[19] Rajpurkar, P., Hannun, A. Y., Haghpanahi, M., Bourn, C., & Ng, A. Y. (2017). Cardiologist-level arrhythmia detection with convolutional neural networks. Nature Medicine, 25, 65–69.

[20] Yildirim, Ö. (2018). A novel wavelet sequence-based deep learning approach for ECG signal classification. Computer Methods and Programs in Biomedicine, 152, 55–64.

[21] Zubair, M., et al. (2019). CNN-LSTM based framework for human activity recognition using wearable sensors. Journal of Healthcare Informatics Research, 3, 67–80.

[22] Sharma, S., & Singh, B. (2020). IoT-based human health monitoring system using ECG data. International Journal of Advanced Computer Science and Applications, 11(1).

[23] Al-Naffakh, M., et al. (2021). Deep learning for monitoring cardiovascular diseases using ECG and PPG data. IEEE Access, 9, 1032–1043.

[24] Sannino, G., & De Pietro, G. (2020). A deep learning approach for ECG-based heartbeat classification for arrhythmia detection. Journal of Ambient Intelligence and Humanized Computing, 11, 1401–1411.

[25] Faust, O., Hagiwara, Y., Hong, T. J., Lih, O. S., & Acharya, U. R. (2019). Deep learning for healthcare applications based on physiological signals: A review. Computer Methods and Programs in Biomedicine, 161, 1–13.