Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.699374
Title: Ambient assisted living systems for older people with Alzheimer's
Author: Al-Shaqi, Riyad
ISNI:       0000 0004 5989 3544
Awarding Body: Cardiff University
Current Institution: Cardiff University
Date of Award: 2016
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
The older people population in the world is increasing as a result of advances in technology, public health, nutrition and medicine. People aged sixty or over were more than 11.5% of the global population in 2012. By 2050, this percentage is expected to be doubled to two billion and around thirty-three countries will have more than ten million people aged sixty or more each. With increasing population age around the word, medical and everyday support for the older people, especially those who live with Alzheimer’s who can't be trusted for consistence interaction with their environment, attract the attention of scientists and health care providers. Existing provisions are often deemed inadequate; e.g.; current UK housing services for the older people are inadequate for an aging population both in terms of quality and quantity. Many older people prefer to spend their remaining life in their home environment; over 40% of the older people have concerns about having to move into a care home when they become old and nearly 70% of them worry about losing their independence or becoming dependent on others. There is, therefore, a growing interest in the design and implementation of smart and intelligent Ambient Assisted Living (AAL) systems that can provide everyday support to enable the older people to live independently in their homes. Moreover, such systems will reduce the cost of health care that governments have to tackle in providing assistance for this category of citizens. It also relieves relatives from continuous and often tedious supervision of these people around the clock, so that their life and commitments are not severely affected. Hence, recognition, categorization, and decision-making for such peoples’ everyday life activities is very important to the design of proper and effective intelligent support systems that are able to provide the necessary help for them in the right manner and time. Consequently, the collection of monitoring data for such people around the clock to record their vital signs, environmental conditions, health condition, and activities is the entry level to design such systems. This study aims to capture everyday activities using ambient sensory II information and proposes an intelligent decision support system for older people living with Alzheimer’s through conducting field study research in the Kingdom of Saudi Arabia within their homes and health care centres. The study considers the older people, who live with Alzheimer’s in Kingdom of Saudi Arabia. Since Alzheimer’s is a special form of dementia that can be supported in early stages with the ambient assistive systems. Further, the results of the field study can also be generalized to societies, which are interested in the mental and cognitive behaviour of older people. This generalization is related to the existence of common similarities in their daily life. Moreover, the approach is a generalized approach. Hence it can also be utilized on a new society which is conducting the same field study. This study initially presents a real-life observation process to identify the most common activities for these patients’ group. Then, a survey analysis is carried out to identify the daily life activities based on the observation. The survey analysis is accomplished using a U-test (Mann-Whitney). According to the analysis, it has been found that these people have fourteen common activities. However, three of these activities such as sleeping, walking (standing) and sitting cover about 72% of overall activities. Therefore, this study focuses on the recognition of these three common activities to demonstrate the effectiveness of the research. The activity recognition is carried out using a common image processing technique, called Phase-Correlation and Log-Polar (PCLP) transformation. According to results, the techniques predicted human activities of about 43.7%. However, this ratio is low to utilise for further analysis. Therefore, an Artificial Neural Network (ANN)- based PCLP model is developed to increase the accuracy of activity recognition. The enhanced PCLP transformation method can predict nearly 80% of the evaluated activities. Moreover, this study also presents a decision support system for Alzheimer’s people, which will provide these people with a safe environment. The decision support system utilises an extended sensory-based system, including a vision sensor, vital signs sensor and environmental sensor with expert rules. The proposed system was implemented on an older people patient with 87.2% accuracy.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.699374  DOI: Not available
Keywords: R Medicine (General) ; TA Engineering (General). Civil engineering (General)
Share: