Student Washington University in St. Louis, United States
Introduction: Electrophysical signals are caused by changes in the membrane potential of individual cells and can be divided into 3 categories. EEG is a signal from the brain, from which epilepsy can be diagnosed and hyper connection can be studied: for example, controlling a computer only through use of the brain and with no motor function. ECG is the signal of activity of the heart. This is used to study terminal disease such as arrhythmias, myocardial infarction, and aortic rupture. EMG is measured from signals of the muscle, a common practice for sports science. However, the main goal is to use EMG as a tool to monitor maternal health. Many developing countries and low income communities lack basic infrastructure: pregnant females have complications due to not being able to go to the hospital, which can result in death. To address this problem there is a need for long term uterine monitoring to predict childbirth with two focuses on this research: using ionic gel to keep the signal longer, and using a dry and stick electrode to reduce contact impedance. In addition, these focuses can also be applied to a simple, long term epilepsy monitoring device, with the recording being inside of the ear.
The main goal of this work is to use create electrodes which can fabricated in a low cost and scalable manner which are is ideally suited for long term and motion artifact tolerant recording of high quality electrophysiology signals to monitor uterine contractions and epilepsy.
Materials and
Methods: The e-textile system was made by the conductive polymer PEDOT:PSS. The backbone chain of PEDOT has a conjugated structure, which causes it to have electric conductivity, and the PSS stabilizes PEDOT. The first way is PEO reacts with PSS via a crosslinking process for better durability due to the strain. In addition, the chemical structure is changed. The solvent system dissolves the PSS so it does not have conductivity. The PEDOT has a crystal structure while the PSS has a grain structure, so the conductivity only occurs in the PEDOT crystal structure. Second, is screening charge via surfactant or molecule. PSS has a negative charge which interacts with the PEDOT via electrostatic force. Surfactant or PEO can be located between the PSS and PEDOT and screen the negative charge.
First the textile is treated with oxygen plasma to make a hydrophilic surface. Then it is screen printed with a PEDOT:PSS textile ink. The printed textile is annealed in an 80C oven for 3 hours for the crosslinking process. The final electrode is connected with an Ag wire and an Ag epoxy. In addition, soft thread is able to treat the motion artifact by moving the thread following motion. To do so, the PEO ration is controlled to create the strongest crosslinking reaction. A ratio of 20% PEO displays the highest ECG signals and the lowest impedance. To make sponge electrodes dissolve sugar and then coat the sponge with desired ink and let it dry in the oven.
Results, Conclusions, and Discussions: The 19 channel electrode patch is effective. Examining the graph of Frequency vs Impedance, the Impedance measurements at 10 Hz are in the range of 6000 to 22000 Ohms. This range is somewhat large, however, the ideal range of impedance is 5000-10000 Ohms, and most of the data is in this range. The larger impedances are mostly due to the distance between the electrode to the recording. In addition, all impedances decrease as a result of increased frequency, which is expected and provides validity to these measurements. Examining the maternal ECG recordings, the voltage recordings are clean given the time of the recording lasted 20 minutes, and using the QRS complex the heart rate could be determined as well. Examining the uterine EMG recordings, the voltages again are very crisp, and the root mean square shows the peaks in voltage very clearly. Comparing the textile recordings to both Bio-Semi electrodes and tocodynamometer monitoring(standard contractional monitoring systems), the textile electrode clearly shows contractional activity. There was not much noise for the e-textile electrode and it is very easy to visualize contractions with the clear increase and decrease in RMS as opposed to spikes in the TOPO recordings and more noise in the Bio-Semi recordings. Future work will include making the patch smaller in order to place multiple patches on the uterus for maximum recordings.
To obtain ear EEG data, spectrograms were plotted over a recording period of 4 minutes, from which the strength of the alpha wave at 10 Hz was measured to obtain SNR. The reference electrode had the strongest EEG readings, while all ear electrode tests had weaker EEG. The results are inconclusive on what type of electrode works the best for measuring ear EEG as the majority of the SNR values were around 9-10 dB. This may be due to both noise of the signal from the ear and that the alpha wave is not as strong from this location compared to on the forehead. Future work will require additional testing of current EEG airpod electrodes and determine whether recordings will be strong enough for further testing.
