Non-invasive monitoring of uterine contractions throughout the menstrual cycle

Project Abstract While traditionally associated with pregnancy and labor, uterine contractions (UC) are also constantly present throughout the menstrual cycle. Abnormalities in UC frequency and/or direction throughout the cycle are correlated with many uterine disorders, which can have downstream impacts on overall health, necessitating continuous monitoring throughout the cycle. Current UC monitoring technologies are invasive, expensive, and/or operator independent. The most promising approach uses non-invasive, low-cost surface abdominal electrodes in a method known as electrohysterography (EHG). However, the current EHG method requires a patient-specific anatomical MRI scan and 128 wired electrodes placed around the body, restricting it to the lab or clinic setting. There remains an unmet need for ambulatory, low-cost, and operator-independent UC measurement technology. This research training plan will demonstrate feasibility of a wearable EHG (wEHG) device for measuring UC frequency and direction patterns without prior knowledge of patient-specific anatomy by leveraging both computational (Aim 1) and experimental (Aim 2) approaches. In Aim 1, a computational wEHG model using human anatomy from MRI scans will be developed based on prior work and used to create a dataset of UC waveforms with normal and abnormal frequency and direction. Using this dataset, a convolutional neural network (CNN) will be trained to distinguish abnormal UC using wEHG. In Aim 2, a wEHG device will be developed, adapted from prior work, with validation using a gold standard and deployed in healthy participants and those with endometriosis, a common UC disorder. The UC frequency and direction will be compared between the two groups, as well as the wEHG device user experience. The applicant has prior training in wearable devices for continuous monitoring, with 21 publications including 5 as first author since 2020. To further support completion of these aims, the applicant will conduct mentored research and training in electrophysiology simulations, wearable sensor design, and spatiotemporal signal analysis. The training plan includes coursework in physiology and statistical learning, hands-on lab skill training, regular advising meetings, and feedback on written and oral scientific exposition. In summary, this project will combine computational and experimental methods to evaluate the capability of using a wEHG device for longitudinal monitoring of UC, enabling a new avenue of research in uterine physiology. Furthermore, through technical and professional skill development, the fellowship applicant will develop as an independent researcher advancing cardiovascular health through sensors and algorithms.