Remote monitoring and recognition of physical activities of elderly people within smart homes and detection of the deviations in their daily activities from previous behavior is one of the fundamental research challenges for the development of ambient assisted living system. This system is also very helpful in monitoring the health of a swiftly aging population in the developed countries. In this chapter, a framework is proposed for remote monitoring and recognition of physical activities of elderly people using smart phone accelerometer sensor data using deep learning models. The main objective of the proposed framework is to provide preventive measures for the emergency health issues such as cardiac arrest, sudden falls, dementia, or arthritis. For the performance evaluation of the proposed framework, two different benchmark accelerometer sensor datasets, UCI and WISDM, are used. Results analysis confirms the performance of the proposed scheme in terms of accuracy, F1-score, root-mean square error (RMSE).
TopIntroduction
Recent advancement in sensing, communication and intelligent computing technologies have provided motivation to the researchers for the rapid development and design of smart environments for different applications such as smart cities, smart home, smart healthcare, smart office and classroom etc(Satpathy,2016). Concept of the smart home is first realized in (Lutolf,1992) with the objective to integrates different services of the home environments with help of Internet of Things concept. The main motivation for the development of Smart home research is to provide a smart space for the aging population for getting helps from the system in real-time. According to the report published by the WHO in (Report, 2019), elderly population of age above 60 has been increasing at fast rate in the present decade and will get in touch about 2 billion in 2050.
Since elderly people have specific issues related to their health and specially old-age related disease such as Alzheimer, cardiovascular, diabetes, Parkinson’s disease, and also suffers a lot due to limitations in physical functions of their body. In Smart Home applications, remote monitoring and recognition of physical activities of elderly people who live independently, is a very demanding research areas. Although there are many solutions are proposed in the literature for human activity detection, fast and efficient scheme are required for monitoring and recognition of physical activities of elderly people for early detection of their imminent critical conditions. In (Rashidi,2013), a survey of wearable sensors and intelligent tools designed for monitoring the health of older adults is discussed. Similarly, a review of healthcare schemes using wireless sensor network technologies and use of ambient intelligence technique and data mining schemes are discussed in(Salih,2013; Peetoom,2014) for smart monitoring of old age patients suffering from chronic diseases.
In (Pulkkinen, 2013), monitoring of the daily activities using different types of sensor for old age people suffering from diabetes is presented. This work provides a platform for gathering the sensor data and detects the patient’s daily routines and recommends physical exercise. Like (Pulkkinen,2013), in (Chatterjee,2013), a in-home monitoring system is proposed for monitoring and detecting the daily activities of elderly people suffering from diabetic using wearable body sensors and predicts blood glucose level to determines the status of health. In (Chaminda,2012), a smart reminder platform for informing the old age people suffering from memory impairment disease, about their forgotten daily activities. In this research work, Patient’s current location, present behavior data and past activities data are used to infer the prediction for informing reminders to the patient. In (Chernbumroong,2014), a real-time multi-sensor activity monitoring and recognition system is proposed for monitoring the daily activities of the old age people using seven different kinds of body sensors. Similar work is also proposed in (Hussain, 2015) to acquire the daily activities sensor data using wearable sensors and then predict the behavior of the patients. In (Fleury,2010), a self-care system is proposed to assist the old age people living independently and used a SVM-based machine learning algorithm for classification of different daily activities in smart health care applications.