Fetal electrocardiography (ECG) monitoring during pregnancy can provide crucial information for assessing the fetus’s health status and making timely decisions. This paper proposes a portable ECG monitoring system to record the abdominal ECG (AECG) of the pregnant woman, comprising both maternal ECG (MECG) and fetal ECG (FECG), which could be applied to fetal heart rate (FHR) monitoring at the home setting. The ECG monitoring system is based on data acquisition circuits, data transmission module, and signal analysis platform, which consists of low input-referred noise, high input impedance, and high resolution. The combination of the adaptive dual threshold (ADT) and the independent component analysis (ICA) algorithm is employed to extract the FECG from the AECG signals. To validate the performance of the proposed system, AECG is recorded and analyzed of pregnant women in three different postures (supine, seated, and standing). The result shows that the proposed system can record the AECG in different postures with good signal quality and high accuracy in fetal ECG and heart rate information. Sensitivity (Se), positive predictive accuracy (PPV), accuracy (ACC), and their harmonic mean (F1) are utilized as the metrics to evaluate the performance of the fetal QRS (fQRS) complexes extraction. The average Se, PPV, ACC, and F1 score are 99.62%, 97.90%, 97.40%, and 98.66% for the fQRS complexes extraction,, respectively. This paper shows the proposed system has a promising application in fetal health monitoring.

Human eye activity has been widely studied in many fields such as psychology, neuroscience, medicine, and human-computer interaction engineering. In previous studies, monitoring of human eye activity mainly depends on electrooculogram (EOG) that requires a contact sensor. This paper proposes a novel eye movement monitoring method called continuous wave doppler oculogram (cDOG). Unlike the conventional EOG-based eye movement monitoring methods, cDOG based on continuous wave doppler radar sensor (cDRS) can remotely measure human eye activity without placing electrodes on the head. To verify the feasibility of using cDOG for eye movement monitoring, we first theoretically analyzed the association between the radar signal and the corresponding eye movements measured with EOG. Afterward, we conducted an experiment to compare EOG and cDOG measurements under the conditions of eyes closure and opening. In addition, different eye movement states were considered, including right-left saccade, up-down saccade, eye-blink, and fixation. Several representative time domain and frequency domain features obtained from cDOG and from EOG were compared in these states, allowing us to demonstrate the feasibility of using cDOG for monitoring eye movements. The experimental results show that there is a correlation between cDOG and EOG in the time and frequency domain features, the average time error of single eye movement is less than 280.5 ms, and the accuracy of cDOG in eye movement detection is higher than 92.35%, when the distance between the cDRS and the face is 10 cm and eyes is facing the radar directly.

Atrial fibrillation (AF) is an insidious cardiac arrhythmia, with its incidence increasing annually. Timely screening and home-based interventions play a vital role in the effective management of atrial fibrillation (AF). Within the Internet of Medical Things (IoMT) landscape, wearable electrocardiogram (ECG) monitoring devices have been seamlessly integrated to monitor AF. Nonetheless, the substantial influx of ECG signals awaiting annotation and the exorbitant costs associated with employing specialists for manual annotation pose significant hurdles in the development of AF detection systems. Despite the potential utilization of AF detectors trained on open databases, their efficacy in analyzing continuous wearable ECGs remains inadequate. Self-supervised representation learning proficiently characterizes unlabeled data and enhances data utilization without labels. This study proposes an AF detection strategy based on self-supervised pre-training, aiming for optimal AF detection performance with minimal annotation costs. The proposed method employed self-supervised representation learning and pre-training strategy and was validated on four datasets from the 4^th China Physiological Signal Challenge (CPSC2021) database, achieving accuracies of 98.13%, 90.11%, 92.63%, and 91.29%. Additionally, validation on 20 wearable ECG recordings yielded mean accuracies of 99.38% and 99.48% for AF and Non-AF on unlabeled data from recordings used for fine-tuning. We achieved mean accuracies of 97.19% and 94.15% for AF and Non-AF on independent recordings. The results demonstrate the proposed method’s reliability for AF detection in wearable ECG monitoring.

Noncontact capacitive electrocardiogram (cECG) is gaining recognition in cardiovascular disease monitoring for its comfort and non-invasiveness. Compared to the conventional ECG, cECG signal quality is prone to degradation in practical applications due to motion artifacts and power line interference (PLI). This study proposed an optimized signal quality assessment method to identify and remove low-quality cECG signals. Firstly, the human body-electrode interface is modeled to analyze the generation mechanism and influence of cECG motion artifacts and PLI. Then, distinct signal quality indices (SQIs) are proposed to target the characteristics of these interferences. Moreover, optimized cECG features and previously proposed ECG features were combined as multi-features and presented to XGBoost for binary classification training. Finally, Shapley Additive Explanations (SHAP) were utilized for feature optimization to reduce redundant information. Validation on a labelled noncontact cECG database yields an impressive binary classification accuracy of 98.786 %, F1-score of 98.845 %, and kappa of 97.567 %. Moreover, its performance on a subject-independent validation set is also excellent, with an accuracy of 99.130 %, an F1-score of 96.937 %, and a kappa of 96.430 %. The optimized multi-features also demonstrate favorable performance in a triple classification model. The experimental results show that our method offers a precise and convenient solution for cECG signal quality assessment.