A Study on Activity and Mood Recognition in Healthcare Settings Using Smartphones

Abstract

In the recent years with the continuous development of wireless communication and mobile devices is becoming increasingly popular Wireless health is very important and to assist development in multidisciplinary research Mobile healthcare technology will be the key to change user's behavior and increase for self-management of wellness via mobile devices The platform provides convenient tools to record subjects’ physical activity mental health status and deliver customized recommendation Clinicians can observe information on how many physical activities and mental health status to make better recommendations for them The purpose of this study is to develop an activity pattern recognition system to assist clinicians via machine learning technology to automatically identify and record physical activity and emotional status for assisting improvement subjects' healthy Theories in feature extraction feature selection machine learning technology selection classifier and Cognitive Behavioral Therapy combination to provide physical activity services of the physical activity recognition system Specifically the study was aimed to: 1) develop activity and mood recognition system Using machine learning technology to extract the features and developing the recognize model for physical activity recognition and mood recognition 2) develop a communication platform for patients and clinicians 3) apply the physical activity recognition system and mood recognition to type 2 diabetes patients and patients with mild depression and 4) after the system intervention we focus on data analysis and discussion For the assessment of the method we used the precision rate of the 10-fold cross-validation to select classification and recruit subjects Experimental results show that the proposed approaches give an encouraging improvement in its tasks Case study also shows that the literacy aptitude test and the performance of reading comprehension were significantly improved The outcomes are expected to provide useful information for wireless health researchers and computer scientists to develop the related assistive technology and also contribute to useful application in the future The outcomes are expected to provide useful information for wireless health researchers and computer scientists to develop the related assistive technology and also contribute to useful application in the future

Date of Award	2015 Aug 20
Original language	English
Supervisor	Jung-Hsien Chiang (Supervisor)