This blog post explores 2 new proposed Internet of Things (IoT) health monitoring devices for remote patient monitoring (RPM) and mental fatigue detection. The first study introduces a smart belt with pressure sensors for estimating heart rate. The researchers evaluate the feasibility of two heart rate estimation methods, highlighting the system’s practicality and low-cost deployment.
The second study focuses on developing a low-power, low-cost, programmable neural network processor for human motion sensor detection using long- and short-term memory networks. It proposes a design to improve latency, power consumption, and programmability. Additionally, the article discusses the importance of detecting mental fatigue in athletes and emphasizes the need for innovative solutions to enhance their physical and psychological well-being.
Study 1: New Proposed IoT Health Monitoring Devices
A new field study in IoT health monitoring devices tested a smart belt with pressure sensors to measure Ballistocardiogram (BCG) signals. The study involved 24 volunteers who wore the smart belt. They also agreed to wear a finger pulse oximeter, capable of producing a photoplethysmography (PPG) signal, considered the “gold standard” for comparison.
The main goal of the field study was to determine the feasibility of estimating heart rate values using the collected BCG signals. The researchers utilized two methods for HR estimation from BCG signals. The first was an algorithmic approach and the second was based on a Convolutional Neural Network (CNN).
The study’s results showed that the performance in the telemedicine environment was slightly worse than in the controlled environment, with a 29% higher MAE for the algorithmic approach and a 52% higher MAE for the CNN-based process.
However, despite the slightly reduced performance, several positive aspects of the proposed telemedicine solution were found. These included a low packet loss ratio (indicating sound data transmission), efficient processing of the collected biomedical data, low-cost deployment, and positive user feedback. The researched concluded that the system showed robustness, reliability, and practicality. Based on these findings, the authors will further their research to achieve new targets to improve and expand the system’s capabilities and applications.
Study 2: New Proposed IoT Health Monitoring Devices
The research described in this following IoT health monitoring devices study focuses on developing a low-power, low-cost, programmable neural network processor. Specifically, it proposes a human motion sensor detection method based on long- and short-term memory (LSTM) networks. The aim is to create a practical system for studying human motion detection with mobile sensors and portable devices.
The article presents current research on sports psychological fatigue detection. It analyzes the production mechanism, external performance, evidence, and entertainment measures related to sports psychological fatigue to find a better solution to eliminate mental fatigue in athletes. They propose using deep neural networks and mobile sensor technology to research and explore mental recognition of athletes in sports.
The use of human motion information, particularly from inertial motion sensors, is widely applicable in areas such as human-computer interaction, virtual reality, sports, and medical research. Mobile sensor devices and wearable devices are also a rapidly developing research area in human motion recognition methods. The study emphasizes that athletes engaged in high-pressure, high-volume, and high-load training are prone to experiencing mental fatigue.
Psychological testing in sports is considered a complex problem, and researchers in sports medicine, sports physiology, and sports psychology worldwide are studying the nature, characteristics, and mechanisms of sports fatigue. The ultimate goal of the research is to find faster ways to eliminate mental fatigue in athletes.
Study Key Points
These two studies focused on IoT health monitoring devices touch upon promising advancements in remote patient monitoring and mental fatigue detection. The first study demonstrates the feasibility and practicality of using a smart belt with pressure sensors to estimate heart rate. The second study proposes a low-power, low-cost neural network processor design to improve human motion sensor detection, addressing latency, power consumption, and programmability challenges. Both studies emphasize the importance of innovative solutions to enhance overall well-being in healthcare and sports.
To stay updated on the latest research in remote patient monitoring, visit the Tenovi blog. Or subscribe to Tenovi’s weekly Monday newsletter for a summary of new research and the prior week’s articles.
