Among electrophysiological indices, cardiac signal in the form of electrocardiogram (ECG) has been the most popular and widely used signal, not only for patients but also for healthy populations for vital sign or well-being monitoring. However, during the COVID-19 pandemic with practice of social distancing, hospital visits are not recommended and convenient. The recent advances in electronics, Internet of Things (IoT) and data sciences have paved the way for continuous electrophysiological monitoring for humans in the home setting. However, for a special group of population like pregnant women, weekly visits are still required during the third trimester of pregnancy. This calls for a home-based portable and comfortable device which could be used to monitor both the mother and the fetus. In this thesis, we develop a flexible sensor patch that can fit the anatomical surface of people’s abdomen which can measure ECG signals continuously and wirelessly. Instead of using traditional contact electrodes, we investigate the potential of non-contact electrodes which could be unobtrusive and more comfortable. The abdominal ECG signals are transferred to an Android smart phone via Bluetooth Low Energy (BLE) communication. To monitor the mother and the fetus, in-house schemes have been developed to extract fetal ECG and maternal ECG. We characterized the non-contact electrodes using impedance analysis and tested with different insulation materials and thicknesses. We then validated the patch in different cases like sitting, standing and walking, mimicking practice scenarios. We also further validated the entire system with pregnant women.

Deep learning-based models have revolutionized biomedical image and video segmentation, enabling precise and automated analysis of complex structures. This advancement is particularly critical in the study of zebrafish cardiovascular videos, where accurate segmentation of the heart is essential for understanding cardiac function and development. In this work, we focus on the Zebrafish Automated Cardiac Analysis Framework (ZACAF), utilizing the U-net architecture to achieve high-accuracy segmentation of the zebrafish heart from video frames to calculate important cardiovascular parameters.After multiple collaborations with researchers studying different phenotypes in zf using ZACAF, to enhance the generalizability of our model across varying datasets and conditions, we employed transfer learning techniques, leveraging pre-trained models to adapt to new data efficiently. Additionally, we incorporated Test-Time Augmentation (TTA) to further improve model robustness and accuracy by applying various transformations to the input data during inference. This approach proposed a systematic solution for adopting ZACAF in broader genetic studies using deep learning algorithms. Recognizing the importance of temporal dynamics in video data, we extended our work to integrate temporal features into the segmentation model. By analyzing changes between consecutive frames, we aim to capture the heartbeat dynamics more effectively, providing a comprehensive tool for cardiac analysis in zebrafish embryos. Our approach not only advances the field of biomedical video segmentation but also contributes to the broader understanding of cardiac function in developmental biology.