Classification of emotional stress and physical stress using a multispectral based deep feature extraction model

Real-time stress detection based on artificial intelligence for people with an intellectual disability

People with severe intellectual disabilities (ID) could have difficulty expressing their stress which may complicate timely responses from caregivers. The present study proposes an automatic stress detection system that can work in real-time. The system uses wearable sensors that record physiological signals in combination with machine learning to detect physiological changes related to stress. Four experiments were conducted to assess if the system could detect stress in people with and without ID. Three experiments were conducted with people without ID (n = 14, n = 18, and n = 48), and one observational study was done with people with ID (n = 12). To analyze if the system could detect stress, the performance of random, general, and personalized models was evaluated. The mixed ANOVA found a significant effect for model type, F(2, 134) = 116.50, p < .001. Additionally, the post-hoc t-tests found that the personalized model for the group with ID performed better than the random model, t(11) = 9.05, p < .001. The findings suggest that the personalized model can detect stress in people with and without ID. A larger-scale study is required to validate the system for people with ID.

IoT-Based Solution for Detecting and Monitoring Upper Crossed Syndrome

A sedentary lifestyle has caused adults to spend more than 6 h seated, which has led to inactivity and spinal issues. This context underscores the growing sedentary behavior, exemplified by extended sitting hours among adults and university students. Such inactivity triggers various health problems and spinal disorders, notably Upper Crossed Syndrome (UCS) and its association with thoracic kyphosis, which can cause severe spinal curvature and related complications. Traditional detection involves clinical assessments and corrective exercises; however, this work proposes a multi-layered system for a back brace to detect, monitor, and potentially prevent the main signs of UCS. Building and using a framework that detects and monitors signs of UCS has facilitated patient-doctor interaction, automated the detection process for improved patient-physician coordination, and helped improve patients' spines over time. The smart wearable brace includes inertial measurement unit (IMU) sensors targeting hunched-back postures. The IMU sensors capture postural readings, which are then used for classification. Multiple classifiers were used where the long short-term memory (LSTM) model had the highest accuracy of 99.3%. Using the classifier helped detect and monitor UCS over time. Integrating the wearable device with a mobile interface enables real-time data visualization and immediate feedback for users to correct and mitigate UCS-related issues.