Real-Time Cardiovascular Event Detection & Prediction using Sensor Fusion and Machine Learning

This project successfully designed, implemented, and validated a Cardiovascular Event Detection and Prediction System that combines physiological sensor-based measurement with supervised machine learning algorithms in a simple, practical, and cost-effective manner. Four supervised machine learning algorithms — Random Forest, SVM, KNN, and Decision Tree — were implemented, trained on the Cleveland Heart Disease dataset, and comparatively evaluated. The Random Forest classifier demonstrated the best performance with an accuracy of 93.18%, precision of 92.5%, recall of 94.1%, and F1-score of 93.3%. The trained model is deployed on a Flask-based web dashboard running on the computer, which accepts manually entered sensor values and clinical parameters and immediately displays the cardiovascular risk classification result on the screen. The system follows a clean and straightforward two-stage workflow — hardware measurement at the sensor level and manual data entry at the software level — making it simple to operate, easy to understand, and practically deployable in clinical settings without requiring any network infrastructure, IoT connectivity, or automated data pipelines. The project successfully provides a low-cost, non-invasive, and accessible alternative to traditional hospital-based cardiovascular diagnostics, demonstrating that an effective heart disease prediction system can be built using affordable hardware components and open-source machine learning tools.

[1]              L. Breiman, "Random Forests," Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. [2]              Mannat M., Amit K., Vivek K. Singh, and Yuddhvir S., "Heart Disease Prediction Analysis with Machine Learning," Proceedings of the IEEE International Conference on Devices and Technology (ICDT), 2025. [3]              N.L.V. Venugopal, G. Swetha, A. Sneha, S.K. Banerjee, V.D. Babu, and A. Lakshmanarao, "A Hybrid Model for Heart Disease Prediction using K-Means Clustering and Semi-Supervised Label Propagation," Proceedings of the 3rd International Conference for Advancement in Technology (ICONAT 2024), Goa, India, IEEE, 2024. [4]               Narendra M., Vinod J., and Gauranshi A., "Heart Disease Prediction Using Supervised Machine Learning Algorithms," Proceedings of the IEEE International Conference on Smart Computing and Data Science (ISCON), 2021. [5]               Senthilkumar M., Chandrasegar T., and Gautam S., "Effective Heart Disease Prediction Using Hybrid Machine Learning Techniques," IEEE Access, 2023. [6]              Suhitha K., Thanguturu L., Muthineni P.S.S., and Venubabu R., "Heart Disease Prediction Using Hybrid ML Algorithms," Proceedings of the IEEE International Conference on Smart Computing and Data Science (ICSCDS), 2023. [7]               T. Cover and P. Hart, "Nearest Neighbor Pattern Classification," IEEE Transactions on Information Theory, vol. 13, no. 1, pp. 21–27, 1967. [10]  Tarun K. Agrawal and Dharmendra Gupta, "Heart Disease Diagnosis Using Hybrid Machine Learning Model in Layered Classification Approach," International Journal of Scientific and Engineering Research (IJSER), Vol. 12, Issue 12, December 2024.