Real-Time Heart Rate Monitoring and Analysis for Different Age Groups Using Integration of Cloud and ML

Most IoT-based heart rate monitoring systems suffer from high latency, low scalability, and excessive dependence on the cloud, which restricts real-time performance. This article addresses these challenges by proposing an efficient, real-time heart rate monitoring system based on edge-enabled Internet of Things (IoT) architecture. The system monitors heart rate variations across different age groups, from children to the elderly, with the assistance of IoT technology. ESP32’s built-in Wi-Fi facilitates easy data transfer to Thing Speak for real-time storage and analysis. However, using Thing Speak in isolation and Wi-Fi in isolation introduces latency, scalability issues, and poor performance in low-connectivity scenarios. In response to these scalability and latency issues, future systems will look beyond Thing Speak and Wi-Fi, such as AWS IoT Core and Azure IoT Hub, which provide real-time analysis and better support for large-scale healthcare monitoring. Furthermore, the system’s dependency on Wi-Fi can be inconvenient in low-connectivity scenarios. Future implementations will include communication protocols, such as Bluetooth low energy (BLE) and LoRaWAN, providing higher coverage and lower power consumption heart rate was measured four times: morning (5:00–7:00 AM), afternoon (1:00–3:00 PM), evening (5:00–7:00 PM), and night (8:00–9:00 PM). Heart rate signals in real-time are noisy because of sensor drift or motion artifacts. Preprocessing techniques, such as smoothing, removal of outliers, and normalization were employed to counter these effects. Wavelet transformation and adaptive filters will be explored in future research to further clean the signal. Future research will explore on-device forecasting and anomaly detection with 97.8% accuracy, better than state-of-the-art IoT-based heart rate monitoring systems. Precision, recall, F1-score, and confusion matrix will be explored in future research to establish better false positive and negative rates.