The Ifmbe Proceedings Series

Published Date: 02 Nov 2017

Abstract— This paper presents an architecture allowing simple access to Ambient Intelligent Systems deployed in domestic environments, particularly focused on people with balance disorders. Our aim is to develop a systemic life-space solution, for indoor environments, by integrating heterogeneous physiological and environmental real-time data that will result in a continuous monitoring and early warning system of people with mobility problems due to loss of balance. The designed control system is adaptive, and it can accommodate to changing conditions of inhabitants. The system recognizes the state of the environment by integrating different contextual data from different devices and sensors. It processes both video and audio signals to detect dangerous events and trigger automatic warnings.

Keywords— Enter up to five keywords and separate them by commas.

INTRODUCTION

Loss of balance is the inability to maintain an upright posture during standing and walking. The human postural control is a complex skill which requires the contribution of vision, vestibular sense, proprioception, muscle strength and reaction time. With increased age, there is a progressive loss of functioning of these systems which can contribute to balance perturbations[6os]. On the other hand, loss of balance should not be dismissed as an unavoidable consequence of aging, as most of these pathologies can be handled properly as long as there is early detection.

Due to the demographic and social changes the percentage of population aged 65 and above in the developed countries will rise from 7.5% in 2009 to 16% in 2050[26].In addition, about one-third of the elderly population reports difficulty with balance or walking. In adults over the age of 65, balance problems are related to falls. Falls that take place indoors are expected to result in hip fracture [9]. The 85% of fractures occur at home and 25% of them are due to environmental risks in the home. In parallel, the majority of older people lives in private households and, along with increasing age and declining health, tend to spend more and more of their time inside the home or in its immediate surroundings. Furthermore, fear of falling results to social isolation, inactivity and sometimes depression. In addition, more recent studies provide information on the predictive value of balance and most of these conclude that elderly people with decreased balance abilities have a higher risk of developing Activities of Daily Living (ADL) disability [3, 4]. Without receiving sufficient care, elderly are at risk of losing their independence. Thus, a system permitting elderly to live safely at home is more than needed.

In recent years special attention has given to applications of ambient intelligent aimed at increasing the security of elderly people in their living environment, prolonging the autonomous living at home, while assuring a high level of assistance when needed. These kinds of systems are usually based on spreading different kinds of sensors and actuators over the environment in order to capture relevant information and to act accordingly inside the ambient. Several methods have been being developed focusing on motion capture and gait measurement since research has shown that the parameters which describe locomotion are indispensible in the diagnosis of frailty and fall risk [29]. It should be noted that there aren’t any integrated software systems or standalone tools towards the early diagnosis and management of the balance disorders.

Recent research in activity monitoring of older adults has focused on the use of passive infrared (PIR) motion sensor in the home. These sensors generate information about the daily activity of monitored subjects, and arrays of such sensors have been used to obtain velocity measurements on a continuous basis in domestic environments.[2,11] In addition wearable devices for measuring gait, based on accelerometers are an area that has received significant attention. The acceleration data is often augmented with data from sensors such as gyroscopes [12], magnetometers [2], ambient light [6] or ambient and skin temperature [13], aiming to extract a more detailed environmental user context. Another widely researched area that has been applied to gait assessment is vision-based monitoring systems. Human motion analysis using vision technology addresses the need for passive, environmentally mounted hardware that does not require those being monitored to wear any devices.

The goal of this paper is to present a provisional architecture of an intelligent and highly automated home environment, specifically targeted for elderly people susceptible to loss of balance disorders. The underlying core of this environment is based on the AmI concept, therefore resulting in a threefold contribution: i) significant enhancement and amelioration of mobility, ii) seamless alignment in everyday life with minimal effort and involvement of the patient, iii) alarming of critical situations. In the sections that follow, we present the design of this supportive environment, describe the technical details and discuss its components in terms of functionality.

System overview

Based on current findings there are varied factors that affect mobility and balance of elderly people and also there are factors that are affected by the existence of the mobility and balance loss problem. For instance, people with specific problem have a strong feel of insecurity and fall, feelings that lead to increase blood pressure, heart rates and changes in skin conductance. In addition, reduced or excessive lighting, unsymmetrical or extremely high or narrow steps, humidity, slippery surfaces, unmarked edges, poor surroundings around home such as garden paths and walks that are cracked or slippery from rain, snow or moss is some of the factors that can pose risks.

Although many aspects of balance and gait can influence risk of falling, a critical factor is the ability to react effectively to balance disturbances. A critical factor is the ability to respond effectively to balance perturbations. These perturbations can arise from: (a) slips and missteps; (b) collisions or other physical interactions with objects (animate or inanimate) in the environment; or (c) the destabilizing effects of volitional movement [8, 9]. A ‘hostile’ leaving environment can increase potential perturbations. In order to accomplish this, inhabitant’s actions would have to be monitored continuously, because the individual environment may change due to a change in the patient’s functional or cognitive activity, or an alteration in their routines.

The intelligent ambient proposed in this paper is composed by several audio and video sensors spread in the environment combined with a wearable sensor attached to the wrist and a wireless pressure insole sensor. A first prototype scenario has been formed, in which a user has placed an ambient system at home that is able to monitor his or her activity to detect incidents related to loss of balance. The prototype apartment has a bedroom, a hall, a toilet, a kitchen and a living room. The system predicts the inhabitants’ needs using patterns in the inhabitants’ behavior. The current measurements will be gathered through the sensors in order to obtain information about the environment and vital signs of the elderly people. Passive infrared motion sensors are installed at each location and, in addition, the bedroom has a pressure sensor positioned in bed, the sofa in the living room hosts another pressure sensor and a magnetic sensor has been settled to the doorframe to detect the opening and closing of the entrance door. All sensor boards have a complementary temperature and humidity sensor, while in parallel measuring ambient light. The system is open to receive information from a triaxial accelerometer integrated in the wristband sensor and physiological signs included heart rate, blood pressure and body temperature. Furthermore

Data streams from these sensors are collected and processed and only when an event is detected it is transmitted through the wireless network to a base station using the IEEE-802.15.4 communication standard. A PC-based AmI station receives the detected events from the base station through a USB interface and decides what action to take.

As described earlier, devices and sensors in the smart home environment produce context information in diverse forms. In order to provide flexibility and scalability, Integrating context data from different sources to perceive the overall environmental state requires a significant programming effort because devices’ source code must be updated whenever a new device or a new integration rule is added. Middleware architecture plays an important role here since the connection between devices are not fixed. We need a flexible and light-weight operating system not only for connecting sensors and actuators, but for providing seamless connection from devices to users.

The final platform will be able to provide efficient feedback to elderly users as well as augmented and personalized alarming and recommendations. The platform will be adapted to the personal needs and preferences of the users and will be characterized by user friendliness and unobtrusive existence of sensors.

Analysis consists in collecting, in real time information coming from different sensors. The selection of the sensors will be based on criteria such us small size, light weight, unobtrusiveness, potential usefulness for our target audience, aesthetics, safety, privacy, operational capacity, ease of Integration. Real-time information is being recorded with innovative sensors obtaining various heterogeneous data such us:

• Physiological signs(sweat, blood pressure, heart rate, etc.).

• mobility status: the gait of the elderly people,

the motion and position of the elderly people, the morphology of the ground, the layout of the

room where the patient moves.

• environmental measurements : ambient light, humidity, soil texture ,weather status, morphology of the ground, map of the room where the person.

The smart home platform hardware consists of a back-end and network infrastructure, and a home gateway that controls a wireless sensor network, network communications and the delivery of a range of services including but not limited to: health, security and smart energy. In fact, by using the historical information from its database, the system will be able to decide whether to trigger an alarm signal, which in turn will alert a concerned person (a close relative or a caregiver) for an immediate medical assistance.

CONCLUSIONS

Assisted living home environments, monitor a wide variety of factors and although there are many examples of ambient home environments only a small number of them implement mobility monitoring as part of their functions. This is because it is quite difficult to measure mobility only by using ambient sensors. Passive infrared sensors and switches placed in doorframes are the most common sensors applied to measure mobility. These sensors measure mobility by determining the location of the subject and recording their interactions in that location as well as the time spent there.

However, ambient home care systems have several disadvantages including the requirement to identify the monitored subject from others in the home and the inability to monitor the subject outside of the home environment.

Monitoring the activity of an elderly person in a smart

home is comparatively simple if the person is living alone, because all the detected activity can be attributed to that person. The ambient smart home must have the ability to identify the monitored subject, and distinguish between their location and the location of others if the monitored person is living with others or regularly receives visitors. This can be achieved using video recognition and audio recognition, height recognition, wearable id tags or footstep analysis.

Video recognition and audio recognition may be seen

as disturbing and invasive.

Ambient smart homes cannot monitor a person while they are outside of the home environment. A wearable element

would be required to measure the person’s mobility outside of the home environment but then the system would have to be reclassified as a combination system. Smart home systems are therefore not suited to monitoring the mobility levels of active persons who are frequently and irregularly, outside of the home.

ACKNOWLEDGMENT

Format the Acknowledgment and References headlines without numbering.