A Neuromorphic Patient Monitoring Framework Integrating Biologically Inspired Spiking Neuron Models and IoT-Based Real-Time Visualization

Abstract

Continuous patient monitoring plays a vital role in modern healthcare systems  by enabling early detection of abnormalities and offer remote supervision. Traditional patient monitoring systems based on conventional digital signal processing often suffer from high power consumption and weak biological interpretability. To address these challenges, this paper presents a neuromorphic patient monitoring framework that integrates biologically inspired spiking neuron models with Internet-of-Things (IoT). The proposed system continuously acquires key physiological parameters using biomedical sensors. The sensory network is built around NodeMCU interfaced with pulse-oximeter units, temperature sensors such as LM35 and DS18B20, pressure transducers, and auxiliary devices including an LM393 microphone, APDS-9960 optical sensor, and MPU-6050 inertial unit for capturing motion-related and contextual information.

The analog front-end provides signal conditioning through amplification (gain ≈ 600–800), ECG band-pass filtering between 0.5 and 40 Hz, and digitization using 10-bit ADCs before wireless transmission. All acquired signals are pre-processed, normalized, and encoded into spike-based representations suitable for neuromorphic computing. The resulting neural activity with real-time numerical readings and realistic waveforms, is transmitted over Wi-Fi and visualized through a web-based dashboard. The interface concurrently presents raw biosignals, extracted vital signs, and neuromorphic spike trains, offering a comprehensive and intuitive picture of the patient’s physiological condition. The proposed system ignores transmitting continuous data’s when the patient is in normal state. Also, it triggers alert when the patient condition is abnormal. By combining low-power embedded hardware, biologically plausible computation, and cloud connectivity, the proposed approach demonstrates a promising pathway toward next-generation remote healthcare monitoring systems. This study illustrates the potential of neuromorphic IoT platforms for future smart healthcare applications such as remote continuous monitoring, early disease diagnosis, and potential integration into FPGA/ASIC-based neuromorphic medical devices.