Lifelog agent for human activity pattern analysis on health avatar platform

Feasibility and efficacy of TouchCare system using application for older adults living alone: a pilot pre-experimental study

BACKGROUND

With the number of older people living alone continuously rising, health-monitoring systems using information and communication technology (ICT) have been developed to manage their health issues. Life logging and human body communication sensor, types of ICT, have been adapted to manage and monitor health status of the elderly. However, its feasibility and efficacy remain unclear. This study aimed to examine the feasibility of TouchCare system which combined life logging with human body communication technology and its effect on the physical and psychological status of older adults living alone.

METHODS

The TouchCare system, which consisted of a wearable watch, touchpad sensors, TouchCare application, and context-aware artificial intelligence, was developed by DNX Co. Ltd and used by the participants for 5 months. Out of the 111 selected participants, 91 replied to the satisfaction survey, and 22 participated in further investigation regarding their physical and psychological status. Finally, health assessment from 14 participants and sensor data from 13 participants (mean age = 77.4; SD = 3.8) were analyzed to compare their health status and health-related behaviors before and after use of the system.

RESULTS

Out of the 91 participants who took the survey, 51.6% were satisfied with the system. Nutritional status (pre-intervention (10.6 ± 2.0) vs. post-intervention (11.8 ± 1.9), P = 0.04) and fall efficacy (pre-intervention (89.2 ± 15.3) vs. post-intervention (99.9 ± 0.5), P = 0.001) significantly improved after use of the system. Chronic pain (pre-intervention (4.8 ± 2.5) vs. post-intervention (4.4 ± 3.7), P = 0.78) and depressive symptoms (pre-intervention (5.7 ± 3.9) vs. post-intervention (5.4 ± 3.1), P = 0.60) reduced, while cognitive function (pre-intervention (4.1 ± 1.4) vs. post-intervention (4.6 ± 1.1), P = 0.15) and physical performance related to walking improved (pre-intervention (3.9 ± 0.2) vs. post-intervention (4.0 ± 0), P = 0.35), but were not significant. Behaviors related to physical activity and gait improved after use of the system; touch counts of refrigerator and microwave also increased with a decrease in night touch counts.

CONCLUSIONS

The TouchCare system was acceptable to older people living alone, and it efficiently managed their daily living while promoting their health-related behaviors. Further experimental studies are required to verify the effectiveness of the system, and to develop the system which meet the individualized needs of older people living alone.

A Novel Energy-Efficient Approach for Human Activity Recognition

In this paper, we propose a novel energy-efficient approach for mobile activity recognition system (ARS) to detect human activities. The proposed energy-efficient ARS, using low sampling rates, can achieve high recognition accuracy and low energy consumption. A novel classifier that integrates hierarchical support vector machine and context-based classification (HSVMCC) is presented to achieve a high accuracy of activity recognition when the sampling rate is less than the activity frequency, i.e., the Nyquist sampling theorem is not satisfied. We tested the proposed energy-efficient approach with the data collected from 20 volunteers (14 males and six females) and the average recognition accuracy of around 96.0% was achieved. Results show that using a low sampling rate of 1Hz can save 17.3% and 59.6% of energy compared with the sampling rates of 5 Hz and 50 Hz. The proposed low sampling rate approach can greatly reduce the power consumption while maintaining high activity recognition accuracy. The composition of power consumption in online ARS is also investigated in this paper.

An Automatic User-Adapted Physical Activity Classification Method Using Smartphones

In recent years, an increasing number of people have become concerned about their health. Most chronic diseases are related to lifestyle, and daily activity records can be used as an important indicator of health. Specifically, using advanced technology to automatically monitor actual activities can effectively prevent and manage chronic diseases. The data used in this paper were obtained from acceleration sensors and gyroscopes integrated in smartphones. We designed an efficient Adaboost-Stump running on a smartphone to classify five common activities: cycling, running, sitting, standing, and walking and achieved a satisfactory classification accuracy of 98%. We designed an online learning method, and the classification model requires continuous training with actual data. The parameters in the model then become increasingly fitted to the specific user, which allows the classification accuracy to reach 95% under different use environments. In addition, this paper also utilized the OpenCL framework to design the program in parallel. This process can enhance the computing efficiency approximately ninefold.