A Real-Time Wearable Physiological Monitoring System for Home-Based Healthcare Applications

The acquisition of physiological data are essential to efficiently predict and treat cardiac patients before a heart attack occurs and effectively expedite motor recovery after a stroke. This goal can be achieved by using wearable wireless sensor network platforms for real-time healthcare monitoring. In this paper, we present a wireless physiological signal acquisition device and a smartphone-based software platform for real-time data processing and monitor and cloud server access for everyday ECG/EMG signal monitoring. The device is implemented in a compact size (diameter: 30 mm, thickness: 4.5 mm) where the biopotential is measured and wirelessly transmitted to a smartphone or a laptop for real-time monitoring, data recording and analysis. Adaptive digital filtering is applied to eliminate any interference noise that can occur during a regular at-home environment, while minimizing the data process time. The accuracy of ECG and EMG signal coverage is assessed using Bland–Altman analysis by comparing with a reference physiological signal acquisition instrument (RHS2116 Stim/Recording System, Intan). Signal coverage of R-R peak intervals showed almost identical outcome between this proposed work and the RHS2116, showing a mean difference in heart rate of 0.15 4.65 bpm and a Wilcoxon’s p value of 0.133. A 24 h continuous recording session of ECG and EMG is conducted to demonstrate the robustness and stability of the device based on extended time wearability on a daily routine.

PulseEdit: Editing Physiological Signal in Facial Videos for Privacy Protection

Recent studies have shown that physiological signals can be remotely captured from human faces using a portable color camera under ambient light. This technology, namely remote photoplethysmography (rPPG), can be used to collect users' physiological status who are sitting in front of a camera, which may raise physiological privacy issues. To avoid the privacy abuse of the rPPG technology, this paper develops PulseEdit, a novel and efficient algorithm that can edit the physiological signals in facial videos without affecting visual appearance to protect the user's physiological signal from disclosure. PulseEdit can either remove the trace of the physiological signal in a video or transform the video to contain a target physiological signal chosen by a user. Experimental results show that PulseEdit can effectively edit physiological signals in facial videos and prevent heart rate measurement based on rPPG. It is possible to utilize PulseEdit in adversarial scenarios against some rPPG-based visual security algorithms. We present analyses on the performance of PulseEdit against rPPG-based liveness detection and rPPG-based deepfake detection, and demonstrate its ability to circumvent these visual security algorithms.

An approach of re-organizing input dataset to enhance the quality of emotion recognition using the bio-signals dataset of MIT

The purpose of this paper is to propose an approach of re-organizing input data to recognize emotion based on short signal segments and increase the quality of emotional recognition using physiological signals. MIT's long physiological signal set was divided into two new datasets, with shorter and overlapped segments. Three different classification methods (support vector machine, random forest, and multilayer perceptron) were implemented to identify eight emotional states based on statistical features of each segment in these two datasets. By re-organizing the input dataset, the quality of recognition results was enhanced. The random forest shows the best classification result among three implemented classification methods, with an accuracy of 97.72% for eight emotional states, on the overlapped dataset. This approach shows that, by re-organizing the input dataset, the high accuracy of recognition results can be achieved without the use of EEG and ECG signals.

Forecasting adverse surgical events using self-supervised transfer learning for physiological signals

Hundreds of millions of surgical procedures take place annually across the world, which generate a prevalent type of electronic health record (EHR) data comprising time series physiological signals. Here, we present a transferable embedding method (i.e., a method to transform time series signals into input features for predictive machine learning models) named PHASE (PHysiologicAl Signal Embeddings) that enables us to more accurately forecast adverse surgical outcomes based on physiological signals. We evaluate PHASE on minute-by-minute EHR data of more than 50,000 surgeries from two operating room (OR) datasets and patient stays in an intensive care unit (ICU) dataset. PHASE outperforms other state-of-the-art approaches, such as long-short term memory networks trained on raw data and gradient boosted trees trained on handcrafted features, in predicting six distinct outcomes: hypoxemia, hypocapnia, hypotension, hypertension, phenylephrine, and epinephrine. In a transfer learning setting where we train embedding models in one dataset then embed signals and predict adverse events in unseen data, PHASE achieves significantly higher prediction accuracy at lower computational cost compared to conventional approaches. Finally, given the importance of understanding models in clinical applications we demonstrate that PHASE is explainable and validate our predictive models using local feature attribution methods.

Mutual Authentication in Body Area Networks (BANs) Using Multi-Biometric and Physiological Signal-Based Key Agreement

The development of wireless technology has had a major impact on the wireless body area networks (WBANs) especially in the medical field where a small wireless sensor is installed in, on, or around the patient’s body for real-time health monitoring and personalized medical treatment. However, the data is collected by the sensors and transmitted via wireless channels. This could make the channel vulnerable to being accessed and falsified by an unauthorized user and may put the lives of the patient at risk and might give a false alarm. Therefore, a secure authentication and data encryption scheme in BANs is needed in a device to establish the interaction. The asymmetric cryptosystems that function in BANs can cause a Man-in-the-Middle attack because the initial requirement in BAN requires the user to configure a master key or password. The impersonation attack may also involve BAN where other individual pretends to be the owner of the devices and lastly Eavesdropping attack where the attack eavesdrops on transmission to unlock devices. With the existing schemes, mutual authentication using the biometric features (fingerprint) and the physiological signal from the electrocardiogram database is used to make sure the authentication is more secure, reliable, and accurate. In this paper, we proposed a new multifactor authentication scheme on biometric authentication which is the retina scan. We proposed the retina scan because the retina of the human eye is unique, remains the same, and cannot be obtained from anywhere which makes it difficult to forge. We also added a new device which is a smart watch to receive a key agreement message from the fingerprint to double confirm the same identification. This is to make sure high security is obtained and offered simplicity, efficiency, and precision scheme for the authentication.