Interface gráfica e hardware modulado multiuso portátil de aquisição de ECG e EMG

Abstract

The growth in the market for wearable systems for monitoring vital signs has led to a problem related to the lack of implementation of these devices in practical environments, given the difficulty of integrating the acquisition of physiological data into relevant data for biomedical analysis. The main objective of the work proposed here is to develop a system that covers all the stages of an electronic system, namely hardware, firmware and a graphical interface compatible with acquisition, enabling real-time monitoring of electrocardiogram (ECG) and electromyogram (EMG) signals, providing the user with analogue and digital processing options, detection of events of interest and calculation of intrinsic parameters of each signal. Given that this system can be used for various purposes, the user must be able to intuitively change the acquisition and processing parameters, creating greater application versatility. As a result, the acquisition module was developed with connectors to interface with the electrode cable that serves as the sensor, with analogue conditioning calculated individually for each signal based on its area of interest, obtaining an appropriate conditioning frequency response for each signal. The conditioning includes an instrumentation amplifier, a low-pass filter and a high-pass filter. Since power transmission at 50 Hz (Portugal) is the biggest source of noise in the signal, a band-reject filter was implemented, achieving a signal-to-noise ratio of 1776.23 V/V for the ECG and 3544.05 V/V for the EMG. The collected data is transmitted via Bluetooth® using an ESP32 v4 microcontroller to the graphical interface. On the interface, the user has a controller tab that allows control of the Bluetooth® connection, changes to the sampling frequency, the continuous signal display window size, deleting the collected data, and saving the data for future analysis. There are also four switchable tabs: 'Acquisition', 'Contraction', 'Filter' and 'Heart Rate'. The first tab displays the signal, analogue and digital filtering controls and activation of autonomous event detection. A methodology adapted from reconstructing a synthetic signal based on the frequency ranges of the original ECG signal obtained using the Wavelet transform was implemented for the autonomous detection of ECG peaks. To detect contractions, a methodology was implemented to compare the lowfrequency modulator curve of the EMG signal with a configurable threshold that defines the contraction and rest periods. The second tab is responsible for displaying contraction data. The user can analyse separate contractions, edit, create or delete contractions, and the interface provides intrinsic data for each signal. In the third tab, the filters can be edited, where the user can change the parameters of the implemented digital low-pass, high-pass, and Notch filters. Finally, on the fourth tab, the user can analyse each cardiac cycle separately, as well as the heart rate during the entire signal period, with the interface providing the size of portions of the cycle commonly used in the literature to detect cardiac pathologies.