Visualization-Driven Time-Series Extraction from Wearable Systems Can Facilitate Differentiation of Passive ADL Characteristics among Stroke and Healthy Older Adults

Wearable technologies allow the measurement of unhindered activities of daily living (ADL) among patients who had a stroke in their natural settings. However, methods to extract meaningful information from large multi-day datasets are limited. This study investigated new visualization-driven time-series extraction methods for distinguishing activities from stroke and healthy adults. Fourteen stroke and fourteen healthy adults wore a wearable sensor at the L5/S1 position for three consecutive days and collected accelerometer data passively in the participant’s naturalistic environment. Data from visualization facilitated selecting information-rich time series, which resulted in classification accuracy of 97.3% using recurrent neural networks (RNNs). Individuals with stroke showed a negative correlation between their body mass index (BMI) and higher-acceleration fraction produced during ADL. We also found individuals with stroke made lower activity amplitudes than healthy counterparts in all three activity bands (low, medium, and high). Our findings show that visualization-driven time series can accurately classify movements among stroke and healthy groups using a deep recurrent neural network. This novel visualization-based time-series extraction from naturalistic data provides a physical basis for analyzing passive ADL monitoring data from real-world environments. This time-series extraction method using unit sphere projections of acceleration can be used by a slew of analysis algorithms to remotely track progress among stroke survivors in their rehabilitation program and their ADL abilities.

Efficient and Explainable Deep Neural Networks for Airway Symptom Detection in Support of Wearable Health Technology

Mobile health wearables are often embedded with small processors for signal acquisition and analysis. These embedded wearable systems are, however, limited with low available memory and computational power. Advances in machine learning, especially deep neural networks (DNNs), have been adopted for efficient and intelligent applications to overcome constrained computational environments. In this study, evolutionary optimized DNNs were analyzed to classify three common airway-related symptoms, namely coughs, throat clears and dry swallows. As opposed to typical microphone-acoustic signals, mechano-acoustic data signals, which did not contain identifiable speech information for better privacy protection, were acquired from laboratory-generated and publicly available datasets. The optimized DNNs had a low footprint of less than 150 kB and predicted airway symptoms of interests with 83.7% accuracy on unseen data. By performing explainable AI techniques, namely occlusion experiments and class activation maps, mel-frequency bands up to 8,000 Hz were found as the most important feature for the classification. We further found that DNN decisions were consistently relying on these specific features, fostering trust and transparency of proposed DNNs. Our proposed efficient and explainable DNN is expected to support edge computing on mechano-acoustic sensing wearables for remote, long-term monitoring of airway symptoms.

Validity of Trunk Acceleration Measurement With a Chest-worn Monitor for Assessment of Exercise Intensity

Background: Recent advancements in wearable technology has enabled easy measurement of daily activities, which can be applied in rehabilitation practice for the purposes such as maintaining and increasing the activity levels of the patients. A smart clothing system is one of the newly developed wearable systems that enables the measurement of physical activity such as heart rate and/or acceleration. In this study, we aimed to examine the validity of trunk acceleration measurement using a smart clothing system (‘hitoe’ system) in assessing the physical activity, which was measured using the expiratory gas analysis. Methods: Twelve healthy individuals participated in the study. The trunk acceleration was simultaneously measured using a triaxial accelerometer embedded in a smart clothing activity monitoring system (‘hitoe’ system), and the percent VO2 reserve (%VO2R) was determined by performing expiratory gas analysis during treadmill testing. Three parameters, that is, moving average (MA), moving standard deviation (MSD), and moving root mean square (RMS), were calculated using the norm of the trunk acceleration. The relationships between these accelerometer-based parameters and %VO2R from expiratory gas analysis for each individual were examined. Results: The values of MA, MSD, RMS, and %VO2R were significantly different between levels 1, 2, 3, and 4 in the Bruce protocol (P<0.01). The average coefficients of determination for individual regression for %VO2R vs. MA, %VO2R vs. MSD, and %VO2R vs. RMS were 0.89±0.05, 0.96±0.03 and 0.91±0.05, respectively. The parameters based on the trunk acceleration measurements were significantly correlated with %VO2R and activity levels. Among the parameters examined, MSD showed the best correlation with %VO2R, indicating high validity of the parameter for assessing physical activity. Conclusions: The present results support the validity of the MSD calculated from the trunk acceleration measured with a smart clothing system in assessing the exercise intensity.Trial registration: UMIN000034967Registered 21 November 2018 (retrospectively registered).

Individual Use of Smartwatches with Self-Monitoring EKG Functionality for Diagnosing Arrhythmias: A Scoping Review (Preprint)

BACKGROUND cardiovascular diseases (CVDs) have become prevalent in the world. They cause millions of deaths globally with the World Health Organization putting the figure at 17.9 million people every year. These statistics indicate the need for healthcare systems to leverage contemporary advanced technology to detect and diagnose CVDs and provide appropriate and timely care to reduce mortality rates. OBJECTIVE To conduct a scoping review exploring individual use of smartwatches with self-monitoring ECG functionality for diagnosing arrhythmias. METHODS Source were selected from six credible bibliographic databases: PubMed, Medline, EMBASE, PsycInfo, CINAHL, and Google Scholar. Intervention-related terms were used to identify relevant sources. Additionally, a forward search strategy was used to search the databases and identify appropriate peer-reviewed journals. RESULTS The research returned 230 sources, out of which 40 met the inclusion criterion. The studies revealed that increased research, development, and adoption of smartwatches and other wearable devices have intensified in the past two decades. The studies showed that using smartwatches can detect cardiac arrhythmias although this depends on the algorithms and biometric sensors utilized in the smartwatches. Watches with advanced algorithms, PPG, and EKG functionalities exhibit high accuracy, sensitivity, and specificity, detecting AFib and other arrhythmias with high efficacy. Therefore, the best way for technology companies to improve their watches’ accuracy is to design and use advanced algorithms and combine PPG, EKG, activity, and biochemical sensors. Conclusion: The contemporary healthcare space is replete with wearable and non-wearable ¬systems and devices central to detecting health conditions and informing the relevant stakeholders to take corrective actions. Smartwatches are wearable devices used chiefly by patients, health, and fitness enthusiasts to detect and monitor a series of conditions, such as heart rate. Their use has fostered timely detection of cardiac arrhythmias, and therefore, caregivers and policy-makers should emphasize their use. CONCLUSIONS Technological systems have proliferated many human spaces in the last three decades, including education, healthcare, and entertainment. Their use has improved operational efficiency, reduced costs, saved lives, and increased organizations’ bottom lines. Healthcare systems use technological devices and appliances to diagnose patients, perform surgeries, improve pharmacy operations, and reduce medical errors. That way, most healthcare facilities provide quality care, attaining positive clinical outcomes. The contemporary healthcare space is replete with wearable and non-wearable ¬systems and devices central to detecting health conditions and informing the relevant stakeholders – caregivers, patients, and family members – to take corrective actions. Smartwatches are wearable devices used chiefly by patients, health, and fitness enthusiasts to detect and monitor a series of conditions, such as heart rate. They are highly effective in detecting cardiac arrhythmias, and therefore, caregivers and policy-makers should emphasize their use.

A Power-Efficient Sensing Approach for Pulse Wave Palpation-Based Heart Rate Measurement

Heart rate (HR) is an essential indicator of health in the human body. It measures the number of times per minute that the heart contracts or beats. An irregular heartbeat can signify a severe health condition, so monitoring heart rate periodically can help prevent heart complications. This paper presents a novel wearable sensing approach for remote HR measurement by a compact resistance-to-microcontroller interface circuit. A heartbeat’s signal can be detected by a Force Sensing Resistor (FSR) attached to the body near large arteries (such as the carotid or radial), which expand their area each time the heart expels blood to the body. Depending on how the sensor interfaces with the subject, the FSR changes its electrical resistance every time a pulse is detected. By placing the FSR in a direct interface circuit, those resistance variations can be measured directly by a microcontroller without using either analog processing stages or an analog-to-digital converter. In this kind of interface, the self-heating of the sensor is avoided, since the FSR does not require any voltage or bias current. The proposed system has a sampling rate of 50 Sa/s, and an effective resolution of 10 bits (200 mΩ), enough for obtaining well-shaped cardiac signals and heart rate estimations in real time by the microcontroller. With this approach, the implementation of wearable systems in health monitoring applications is more feasible.

Emergence of flexible technology in developing advanced systems for post-stroke rehabilitation: a comprehensive review

Objective. Stroke is one of the most common neural disorders, which causes physical disabilities and motor impairments among its survivors. Several technologies have been developed for providing stroke rehabilitation and to assist the survivors in performing their daily life activities. Currently, the use of flexible technology (FT) for stroke rehabilitation systems is on a rise that allows the development of more compact and lightweight wearable systems, which stroke survivors can easily use for long-term activities. Approach. For stroke applications, FT mainly includes the “flexible/stretchable electronics”, “e-textile (electronic textile)” and “soft robotics”. Thus, a thorough literature review has been performed to report the practical implementation of FT for post-stroke application. Main results. In this review, the highlights of the advancement of FT in stroke rehabilitation systems are dealt with. Such systems mainly involve the “biosignal acquisition unit”, “rehabilitation devices” and “assistive systems”. In terms of biosignals acquisition, electroencephalography (EEG) and electromyography (EMG) are comprehensively described. For rehabilitation/assistive systems, the application of functional electrical stimulation (FES) and robotics units (exoskeleton, orthosis, etc.) have been explained. Significance. This is the first review article that compiles the different studies regarding flexible technology based post-stroke systems. Furthermore, the technological advantages, limitations, and possible future implications are also discussed to help improve and advance the flexible systems for the betterment of the stroke community.

Monitoring Lower Back Activity in Daily Life Using Small Unintrusive Sensors and Wearable Electronics in the Context of Rheumatic and Musculoskeletal Diseases

Due to a sedentary lifestyle, the amount of people suffering from musculoskeletal back diseases has increased over the last few decades. To monitor and cure these disabilities, sensors able to monitor the patient for long-term measurement during daily life and able to provide real-time feedback are required. There are only a few wearable systems that are capable to acquire muscle activity (sEMG) and posture at the same time. Moreover, previously reported systems do not target back sensor and typically comprise bulky uncomfortable solutions. In this paper, we present a new wearable sensor network that is designed to measure muscle activity and posture specialized for back measurement. Special care was taken to propose a discrete and comfortable solution. The prototype only measures 3.1 mm in thickness on the spine, making this sensor system the thinnest and lightest one in the literature to our best knowledge. After testing, it was shown that the sensor system is able to acquire two surface electromyography signals concurrently, to gather acceleration and rotation speed from the patient’s lower back, and to transmit data to a computer or a smartphone via serial communication or Bluetooth low energy for a few hours for later processing and analysis.