PulsECG - A Cuffless Non-Invasive Blood Pressure Monitoring Device through Neural Network Analysis of ECG and PPG signals

The monitoring of electrocardiogram (ECG), photoplethysmogram (PPG) and arterial blood pressure is crucial for preserving and enhancing individual health and well-being. These vital parameters offer profound insights into cardiac and pulmonary functions and are indispensable for the diagnosis and management of a plethora of health conditions. This paper presents the design and development of PulsECG, a portable medical device engineered to estimate arterial blood pressure using a cuffless approach. It acquires ECG signals according to the Einthoven's Triangle, monitors blood oxygen levels, and derives blood pressure non-invasively through the use of a neural network. The neural network at the heart of PulsECG leverages a combination of convolutional and bidirectional LSTM layers to process time-series input from dual-channel PPG and ECG signals. A custom database of 20 subjects is collected to train the network on real-life scenario. To this purpose, a custom data acquisition process has been designed, which alternates blood pressure measurements with ECG & PPG recordings, providing a dataset that underpins the network learning. The results show the neural network is able to correctly predict systolic and diastolic blood pressures, proving a high correlation with the ground truth (sphygmomanometer), despite a slight trend towards overestimation. This research advances the integration of neural network models into portable medical devices like PulsECG, fostering telemedicine and continuous health tracking. It opens novel ways for improved patient care, offering a solution for real-time health monitoring, and represents a step forward to combine artificial intelligence with medical technology.