sensors-based wearable systems for monitoring of human...

IJMTES | International Journal of Modern Trends in Engineering and Science ISSN: 2348

Volume: 01 Issue: 03 March 2014

Sensors-Based Wearable Systems for Monitoring ofHuman Movement and Falls

1(UG Scholar, Dept.of.ECE, Christ the king E2(Asst. Professor, Dept.of.ECE, Christ the King Engin

________________________________________________________________________________________________________Abstract Multiple tri axial acceleration sensor devices for joint sensing of injured body parts, when an accidental fall occurs. The

model transmitted the information fed by the sensors distributed over various body parts to the computer through wireless trafor further analysis and judgment, and employed cognitive adjustment method to adjust the acceleration range of various body movements .The concept of a fall is in the common sense, it is difficult to describe it precisely, aperson was injured, he/she will take some time to recovery and then they will try to walk, at that time there are lots of poswhich we can monitor in receiving side by RF receiver. The athe receiving side .The main focuses on designing a system that prevents a patient from falling down when he/she is unconscioaccelerating sensors are used to detect the position of patient who has met an accident and these sensors can be placed at different places of the body. If any abnormal activity occurs in any region, AIRBAG will be opened to prevent the person from falling. This informatidisplayed in Liquid crystal display as well as PC. Front and Back fall detection is detected using x axis of accelerating sensor and LeRight fall detection is detected using y axis of accelerating sensor.

Keywords—component;AIRBAG,BUZZER,ACCELARTOR SENSOR

________________________________________________________________________________________________________

1. INTRODUCTION

GLOBAL demographic trends demonstrate a clear rise in

the proportion of elderly and chronically ill is partly the result of improved healthcare systems, especially in developed nations. However, this trend ultimately poses considerable public health challenges, as most health systems are already stretched to their resource limits and dopossess the capacity or framework to cater for the increasingneeds of an aging population [1].

Concurrently, the occurrence of falls is also on the rise, with more than a third of adults over 65 years of age said to fall at least once per year. The consequences of falls, particularly of those which are followed by a “long lie” of an hour or more, are detrimental, both on a personal and a broader community level, and include hospitalization, morbidity, and mortality [2]. This stresses the importance omonitoring individuals at risk of falling, detecting falls, and implementing preventative strategies to minimize falls risk by allocating suitable and timely intervention when possible. Telehealth – the provision of health services involving information, communications, measurement, and monitoring from a distance to patients’ homes – is forecast to become a mainstream application. Its intended purpose is to alleviate the expected pressure on healthcare systems and facilitate an improved level of care via long-term follow-and intervention purposes [3]. Tele health applications are largely based on the use of sensor technologies [4]. Consolebased systems have been at the core of telehealth for many years. Such systems typically include sensors to capture vital physiological measurements (heart rate, electrocardiogram, etc.) and are often equipped with a user viewing screen. In


www.ijmtes.com

Based Wearable Systems for Monitoring ofHuman Movement and Falls

N.Dharungeeran1, J.JafarAli2 Christ the king Engineering college, Tamilnadu, India, dharungeeran92

Christ the King Engineering College, Tamilnadu, India, jafar86_______________________________________________________________________________________

Multiple tri axial acceleration sensor devices for joint sensing of injured body parts, when an accidental fall occurs. The model transmitted the information fed by the sensors distributed over various body parts to the computer through wireless trafor further analysis and judgment, and employed cognitive adjustment method to adjust the acceleration range of various body movements .The concept of a fall is in the common sense, it is difficult to describe it precisely, and thus to specify its means of detection. If person was injured, he/she will take some time to recovery and then they will try to walk, at that time there are lots of poswhich we can monitor in receiving side by RF receiver. The accelerating sensors values goes abnormal then alert system (buzzer) will alert in the receiving side .The main focuses on designing a system that prevents a patient from falling down when he/she is unconscio

ct the position of patient who has met an accident and these sensors can be placed at different places of the body. If any abnormal activity occurs in any region, AIRBAG will be opened to prevent the person from falling. This informati

in Liquid crystal display as well as PC. Front and Back fall detection is detected using x axis of accelerating sensor and LeRight fall detection is detected using y axis of accelerating sensor.

component;AIRBAG,BUZZER,ACCELARTOR SENSOR

_____________________________________________________________________________________________________

LOBAL demographic trends demonstrate a clear rise in

the proportion of elderly and chronically ill individuals. This is partly the result of improved healthcare systems, especially in developed nations. However, this trend ultimately poses considerable public health challenges, as most health systems are already stretched to their resource limits and do not possess the capacity or framework to cater for the increasing

Concurrently, the occurrence of falls is also on the rise, with more than a third of adults over 65 years of age said

consequences of falls, particularly of those which are followed by a “long lie” of an hour or more, are detrimental, both on a personal and a broader community level, and include hospitalization, morbidity, and mortality [2]. This stresses the importance of monitoring individuals at risk of falling, detecting falls, and implementing preventative strategies to minimize falls risk by allocating suitable and timely intervention when possible.

the provision of health services involving communications, measurement, and monitoring

is forecast to become a mainstream application. Its intended purpose is to alleviate the expected pressure on healthcare systems and facilitate an

-up for diagnostic and intervention purposes [3]. Tele health applications are largely based on the use of sensor technologies [4]. Console-based systems have been at the core of telehealth for many

sors to capture vital physiological measurements (heart rate, electrocardiogram, etc.) and are often equipped with a user viewing screen. In

some instances, ubiquitous sensors may be placed around the home or a residential care facility, as in the “smart hconcept [5]. Remote observation of activity can then be conducted with various objectives in mind, including safety monitoring or to ensure proper care. Another application category involves body-fixed sensors that are referred to as wearable sensors (or systems, since in many cases more than one sensor is utilized). Regardless of the specific application, information collected via such telehealth sensors is transmitted over some distance and possibly further analyzed at its destination, before being viewed by predetermined recipientswhose role it is to monitor or act upon the data when required[3], [4]

In relation to possible nonhealth applications, wearable sensors used for monitoring physical activity can also be used by professional athletes to imof their training. Rescue operations may keep track of an emergency crew’s location by using sensorsystems including global positioning systems. Individuals executing tasks that require high levels of alertness, such astruck drivers or heavy machinery operators, may be monitoredto ensure their consciousness [6]. It is therefore evident that the field of wearable sensors for telehealth is both broad and rapidly emerging. This paper focuses on sensorwearable systems in relation to ambulatory monitoring and presents an overview of recent developments in the field. Movement monitoring and classification are examined, along with a range of clinical applications of these ambulatory sensor technologies, with particular detection and falls risk assessment. 2.AMBULATORY MA. The Significance of Studying Human Motion

IJMTES | International Journal of Modern Trends in Engineering and Science ISSN: 2348-3121

64

Based Wearable Systems for Monitoring of

India, [email protected]) India, [email protected])

_______________________________________________________________________________________ Multiple tri axial acceleration sensor devices for joint sensing of injured body parts, when an accidental fall occurs. The

model transmitted the information fed by the sensors distributed over various body parts to the computer through wireless transmission devices for further analysis and judgment, and employed cognitive adjustment method to adjust the acceleration range of various body parts in different

nd thus to specify its means of detection. If person was injured, he/she will take some time to recovery and then they will try to walk, at that time there are lots of possibilities to fall down,

ccelerating sensors values goes abnormal then alert system (buzzer) will alert in the receiving side .The main focuses on designing a system that prevents a patient from falling down when he/she is unconsciousness. Two

ct the position of patient who has met an accident and these sensors can be placed at different places of the body. If any abnormal activity occurs in any region, AIRBAG will be opened to prevent the person from falling. This information will be

in Liquid crystal display as well as PC. Front and Back fall detection is detected using x axis of accelerating sensor and Left and

_____________________________________________________________________________________________________ some instances, ubiquitous sensors may be placed around the home or a residential care facility, as in the “smart home” concept [5]. Remote observation of activity can then be conducted with various objectives in mind, including safety monitoring or to ensure proper care. Another application

fixed sensors that are referred to as (or systems, since in many cases more than

one sensor is utilized). Regardless of the specific application, information collected via such telehealth sensors is transmitted over some distance and possibly further analyzed at its

iewed by predetermined recipients whose role it is to monitor or act upon the data when required

In relation to possible nonhealth applications, wearable sensors used for monitoring physical activity can also be used by professional athletes to improve the efficiency of their training. Rescue operations may keep track of an emergency crew’s location by using sensor-based wearable systems including global positioning systems. Individuals executing tasks that require high levels of alertness, such as truck drivers or heavy machinery operators, may be monitored to ensure their consciousness [6]. It is therefore evident that the field of wearable sensors for telehealth is both broad and rapidly emerging. This paper focuses on sensor-based

ms in relation to ambulatory monitoring and presents an overview of recent developments in the field. Movement monitoring and classification are examined, along with a range of clinical applications of these ambulatory sensor technologies, with particular emphasis on falls detection and falls risk assessment.

MONITORING

A. The Significance of Studying Human Motion



Human motion is a highly complex concept, which depends on and is, in turn, influenced by many factors, including physiological, anatomical, psychological, environmental, and social effects [6]. Movement reduction or modification can stem from various conditions, including stroke, osteoarthritis, and aging. Once mobility is impaired or reduced, a cycle of deterioration is commonly generated, whereby a person’s capacity to move is further diminished, mostly due to physical and emotional reasons [8]. This highlights the need for proper and timely interventions that address the specific issues which hinder movemindividual and provide the necessary encouragement to keep frail and sick people as physically active as possible. Clearly, to achieve these objectives, we must first be able to monitor and quantify movement, identify reduced or impaired movement, and estimate the value of administered interventions.

Interestingly, though a real fall would never be an intentional or positive event, in the context of ambulatory monitoring, a fall can be regarded as a subcategory of human movement [9]. Its occurrence is related to the same factors that affect movement in general and it often leads to reduced movement and an increased risk for subsequent fall episodes [8]. Monitoring and preventing falls may therefore be possible using similar concepts to those applied to movement monitoring.

B. Assessment Techniques

The study of human motion and falls employs manytechniques, including visual observations, video capture, interviews, diaries, questionnaires, physical measurements, and wearable ambulatory sensors. Self-report tools are simpleto administer, but capture partial information and suffer frominherent bias due to inaccurate recall, whether intentional or not. Objective measurements use a variety of physical tools such as force plates, gait mats, and balance testing apparatus. Such tests are designed to be conducted in a clinical setting, usually in dedicated gait and falls clinics, and are relatively costly and inappropriate for long-term monitoring of large patient cohorts under real-life conditions [10]. sensors or sensor systems that can be worn on the body offer another means of gathering physical activity and falls data in a way that is suitable for clinical settings but has immense potential for long-term use, especially in the community [11]Additional advantages related to the use of such wearable ambulatory monitors (WAMs) are: a) capturing objective measurements of everyday or structured movements, including aspects that cannot be obtained by other assessment tools and b) custom-tailored measurements can be developed to enable improved interventions and to quantify their effect overtime [12]. C. Types of Wearable Ambulatory Sensors

Accelerometers are used to measure acceleration along a sensitive axis and over a particular range of frequencies. Since they measure acceleration due to gravity and movement, the actual component of movement


www.ijmtes.com

Human motion is a highly complex concept, which depends on and is, in turn, influenced by many factors, including physiological, anatomical, psychological, environmental, and social effects [6]. Movement reduction or modification can stem from various conditions, including stroke, osteoarthritis, and aging. Once mobility is impaired or

terioration is commonly generated, whereby a person’s capacity to move is further diminished, mostly due to physical and emotional reasons [8]. This highlights the need for proper and timely interventions that address the specific issues which hinder movement in each individual and provide the necessary encouragement to keep frail and sick people as physically active as possible. Clearly, to achieve these objectives, we must first be able to monitor and quantify movement, identify reduced or impaired

nt, and estimate the value of administered

Interestingly, though a real fall would never be an intentional or positive event, in the context of ambulatory monitoring, a fall can be regarded as a subcategory of human

nce is related to the same factors that affect movement in general and it often leads to reduced movement and an increased risk for subsequent fall episodes [8]. Monitoring and preventing falls may therefore be possible

ied to movement

The study of human motion and falls employs many techniques, including visual observations, video capture, interviews, diaries, questionnaires, physical measurements,

report tools are simple to administer, but capture partial information and suffer from inherent bias due to inaccurate recall, whether intentional or not. Objective measurements use a variety of physical tools

e testing apparatus. Such tests are designed to be conducted in a clinical setting, usually in dedicated gait and falls clinics, and are relatively

term monitoring of large life conditions [10]. Miniature

sensors or sensor systems that can be worn on the body offer another means of gathering physical activity and falls data in a way that is suitable for clinical settings but has immense

term use, especially in the community [11]. Additional advantages related to the use of such wearable ambulatory monitors (WAMs) are: a) capturing objective measurements of everyday or structured movements, including aspects that cannot be obtained by other assessment tools and

measurements can be developed to enable improved interventions and to quantify their effect over

are used to measure acceleration along a sensitive axis and over a particular range of

encies. Since they measure acceleration due to gravity and movement, the actual component of movement-related

acceleration needs to be separated from the gravitational. The gravitational component is nevertheless useful in defining a subject’s postural orientation. There are several types of accelerometers available based on piezoelectric, piezoresistive, or variable capacitance methods of transduction. They all employ the same principle of operation of a mass that responds to acceleration by causing a sprian equivalent component to stretch or compress proportionally to the measured acceleration (Hooke’s law). Early available accelerometer sensor devices were of a uniaxial design; however, further advances in MEMS technology have lead to the availability, at low-cost, of biaxial and triaxial devices, with their sensitive axes mounted orthogonally to one another.Vibrating gyroscopes measure angular velocity by taking advantage of the Coriolis Effect. MEMSa small vibrating mass within the sensor that undergoes a slight displacement when the gyroscope is rotated. If measured over time, a change of angle in relation to an initial known angle can be detected. These sensors have known limitations, which include output drift over time, output offsets when the device is stationary, and a sensitivity which is limited to a particular range of angular velocities. used to measure the orientation of a body segment in relation to the earth’s magnetic north, utilizing electromagnetic induction. In order to work effectively, the orientation of the sensitive axis of the device must be aligned with the magnetic field lines; composite devices containing multiple devices on orthogonal axes are nowused to crequirement. Goniometers are fairly rudimentary devices, based on a potentiometric element which is attached to a joint’s rotation point to measure joint angle, although more advanced flexible electrogoniometers employ strain gauge elements. These sensors (along with used to measure the slope of an object with respect to gravity using an artificial horizon) are mainly employed in the determination of the range of motion of human body joints. D. Design and Usability Considerations

On a technical level, the most important factors in designing an ambulatory sensing system are reliability (no random variance in measurements over time), durability, portability, continuous recording, high resolution at the desired frequencies, and an ability to filter the bandwidth as required. Wireless communication is another fundamental feature; in some cases, real-time data processing via embedded intelligence

3.MOVEMENT MONITORING AND

The monitoring of human movement research field, with applications in gait analysis, rehabilitation, orthotic prescription, prosthesis adjustment, and orthopedic interventions. Although movement classification may be regarded as a clinical application in its own right, for the purpose of this review, it is presented in the following section because it actually forms much of the basis required to achieveambulatory monitoring in general. The work done in this field


65

acceleration needs to be separated from the gravitational. The gravitational component is nevertheless useful in defining a

entation. There are several types of accelerometers available based on piezoelectric, piezoresistive, or variable capacitance methods of transduction. They all employ the same principle of operation of a mass that responds to acceleration by causing a spring or an equivalent component to stretch or compress proportionally to the measured acceleration (Hooke’s law). Early available accelerometer sensor devices were of a uniaxial design; however, further advances in MEMS technology have lead to

cost, of biaxial and triaxial devices, with their sensitive axes mounted orthogonally to one another.

measure angular velocity by taking advantage of the Coriolis Effect. MEMS-based gyroscopes use a small vibrating mass within the sensor that undergoes a slight displacement when the gyroscope is rotated. If measured over time, a change of angle in relation to an initial known angle can be detected. These sensors have known limitations,

output drift over time, output offsets when the device is stationary, and a sensitivity which is limited to a particular range of angular velocities. Magnetometers can be used to measure the orientation of a body segment in relation

c north, utilizing electromagnetic induction. In order to work effectively, the orientation of the sensitive axis of the device must be aligned with the magnetic field lines; composite devices containing multiple devices on orthogonal axes are nowused to compensate for this

are fairly rudimentary devices, based on a potentiometric element which is attached to a joint’s rotation point to measure joint angle, although more advanced flexible electrogoniometers employ strain gauge

nts. These sensors (along with inclinometers that are used to measure the slope of an object with respect to gravity using an artificial horizon) are mainly employed in the determination of the range of motion of human body joints.

Considerations On a technical level, the most important factors in

designing an ambulatory sensing system are reliability (no random variance in measurements over time), durability, portability, continuous recording, high resolution at the

cies, and an ability to filter the bandwidth as required. Wireless communication is another fundamental

time data processing via embedded

ONITORING AND CLASSIFICATION

The monitoring of human movement is a vast research field, with applications in gait analysis, rehabilitation, orthotic prescription, prosthesis adjustment, and orthopedic interventions. Although movement classification may be regarded as a clinical application in its own right, for the urpose of this review, it is presented in the following section

because it actually forms much of the basis required to achieve ambulatory monitoring in general. The work done in this field



is paramount to providing a more in-depth understanding of movement on the whole, as well as generating classification algorithms that accurately identifyWAMmovements, without which one could not apply WAMs for the clinical purposes of falls detection, falls risk assessment and energy expenditure measurement, all of which are discussed in a subsequent section of this review. A. Posture and Transition Identification

A basic level of movement monitoring entails a differentiation between activity and rest. Accelerometry is suitable for this task since during rest only the gravitational component is recorded; the addition of a movement component is relatively clear-cut when activity takes place, assuming correct signal preprocessing via filtering is executed. The premise employed is often that human motion will occupy activities with a repetition rate above 0.1 Hz, or one repetition per 10 s. This is a simple, but useful approximation which allows the orientation of the device to be extracted, even in the presence of other movement; however, changes in orientation faster than this rate are not recognized, and what is calculatedamounts to an average device orientation. B. Gait Analysis

Building on the basic ability to identify walking using WAMs, attention was diverted to more advanced gait analysis. A gyroscope-based device was used in young and old subjects to estimate a range of spatioparameters; reported results were comparable to those obtained with foot pressure sensors, which were regarded as the criterion standard [22]. Using a triaxial accelerometer on the lower trunk, several teams were also able to identify important gait parameters such as walking speed, stride length, gait symmetry, and regularity [23], [24]. Additional popular locations for accelerometry-assisted gait analysis have been the upper trunk, head, tibia, and waist [25]. The ability to obtain objective measurements of gait parameters has been instrumental in comprehending gait-relbetween healthy individuals and those suffering with Parkinson’s disease, neuropathy, or other disorders that affect gait and balance. Moreover, a better understanding of typical changes that occur in gait with aging (e.g., reduced walkingspeed and a shorter step length) has also been achieved [25]. These research areas present an ongoing challenge and have unique considerations. For example, measuring gait parameters in individuals with sensorimotor deficits should take into account abnormal segmental orientation. A possiblesolution may, once again, include a combination of sensormodalities; for example, in one case adding a gyroscope toan accelerometry-based WAM provided additional insightregarding segmental orientation and angular ve C. Balance and Sway Testing

Postural sway and balance assessment using WAMs is another prolific research area. A proofprovided in [29], where an accelerometer, placed in the back close to the center of mass, resulted in simila


www.ijmtes.com

depth understanding of t on the whole, as well as generating classification

algorithms that accurately identifyWAM-recorded movements, without which one could not apply WAMs for the clinical purposes of falls detection, falls risk assessment and

ll of which are discussed in

A basic level of movement monitoring entails a differentiation between activity and rest. Accelerometry is suitable for this task since during rest only the gravitational component is recorded; the addition of a movement

cut when activity takes place, nal preprocessing via filtering is executed.

The premise employed is often that human motion will occupy activities with a repetition rate above 0.1 Hz, or one repetition per 10 s. This is a simple, but useful approximation which

the device to be extracted, even in the presence of other movement; however, changes in orientation faster than this rate are not recognized, and what is calculated

o identify walking using WAMs, attention was diverted to more advanced gait

based device was used in young and old subjects to estimate a range of spatio-temporal gait parameters; reported results were comparable to those

h foot pressure sensors, which were regarded as the criterion standard [22]. Using a triaxial accelerometer on the lower trunk, several teams were also able to identify important gait parameters such as walking speed, stride length,

larity [23], [24]. Additional popular assisted gait analysis have been

the upper trunk, head, tibia, and waist [25]. The ability to obtain objective measurements of gait parameters has been

related differences between healthy individuals and those suffering with Parkinson’s disease, neuropathy, or other disorders that affect gait and balance. Moreover, a better understanding of typical changes that occur in gait with aging (e.g., reduced walking speed and a shorter step length) has also been achieved [25]. These research areas present an ongoing challenge and have unique considerations. For example, measuring gait parameters in individuals with sensorimotor deficits should

mal segmental orientation. A possible solution may, once again, include a combination of sensor modalities; for example, in one case adding a gyroscope to

based WAM provided additional insight regarding segmental orientation and angular velocity [26].

Postural sway and balance assessment using WAMs is another prolific research area. A proof-of-concept was provided in [29], where an accelerometer, placed in the back close to the center of mass, resulted in similar and even better

results than a force platform, which is typically used for balance and sway measurements.

D. Directed Movements or Routines

The movements covered so far fit into the category ofactivities of daily living (ADL). More structured routines often used in order to control some of the unavoidable variability in the way people conduct ADL, and to enable a more informed classification process. For example, subjects may be asked to perform a devised routine which is based on a set number of repetitions of sitting, standing, walking, and lying down that are performed in a certain order [11] or they may be asked to walk on a certain path for a set distance [30].Whether involving WAM use or not, many researchers examine how subjects perform specsuch as: 1) Sit-to-Stand Test (STS or STS5) no armrests, a seated person must stand and sit down again as fast as possible, with their arms folded. This is performed either one or five times, the latter common. STS(5) is considered a measure of lower limb strength, speed and coordination [30].2) Alternate Step Test (AST) standard size platform (19 cm high, 40 cm wide), place each foot on the platform, and replace it back onto the floor as quickly as possible. This may be repeated several times, most commonly four times with each foot. The AST may provide a measure of lateral stability [30].3) Timed Up-and-Go Test (TUGT) standing position, walks three meters, turns around, returns to the chair and sits back down as quickly as possible. Thus, overall this test combines aspects of STS transfers,walking and turning [31]. 4.CLINICAL USE OF WEARABLE

Some of the main clinical applications of WAMs include falls detection, falls prevention and risk assessment, as well as physical activity and energy expenditure monitoring. The following section describes the motivation for each of these applications and sets the conthe more notable or recently reported work in each area.A. Falls Detection

The importance of detecting falls in order to prompt rapid assistance and prevent the consequences associated with“long lie” situations provides a fertclassic personal alarm is not practical in situations where a person is unable to press the emergency button due to loss of consciousness, injury, or emotional distress. Hence, an automatic fall detection system, whether bodyof a smart home setup Fig. 1. Anteroposterior, mediolateral, and vertical accelerations, and barometric pressure, from a young healthy subject performing a simulated fall in theonto a mattress, whereby some attempt is made toimpact of the fall using the hands and arms [13]. The simulated fall is initiated at approximately 25 s, and the fall


66

results than a force platform, which is typically used for balance and sway measurements.

D. Directed Movements or Routines The movements covered so far fit into the category of

activities of daily living (ADL). More structured routines are often used in order to control some of the unavoidable variability in the way people conduct ADL, and to enable a more informed classification process. For example, subjects may be asked to perform a devised routine which is based on a

petitions of sitting, standing, walking, and lying down that are performed in a certain order [11] or they may be asked to walk on a certain path for a set distance [30]. Whether involving WAM use or not, many researchers examine how subjects perform specific controlled movements,

Stand Test (STS or STS5) – using a hard chair with no armrests, a seated person must stand and sit down again as fast as possible, with their arms folded. This is performed either one or five times, the latter version being more common. STS(5) is considered a measure of lower limb strength, speed and coordination [30]. 2) Alternate Step Test (AST) – subjects stand in front of a standard size platform (19 cm high, 40 cm wide), place each

replace it back onto the floor as quickly as possible. This may be repeated several times, most commonly four times with each foot. The AST may provide a measure of lateral stability [30].

Go Test (TUGT) – a seated subject rises to a ing position, walks three meters, turns around, returns to

the chair and sits back down as quickly as possible. Thus, overall this test combines aspects of STS transfers,walking

EARABLE AMBULATORY SENSORS

main clinical applications of WAMs include falls detection, falls prevention and risk assessment, as well as physical activity and energy expenditure monitoring. The following section describes the motivation for each of these applications and sets the context for discussing some of the more notable or recently reported work in each area.

The importance of detecting falls in order to prompt rapid assistance and prevent the consequences associated with “long lie” situations provides a fertile research ground. The classic personal alarm is not practical in situations where a person is unable to press the emergency button due to loss of consciousness, injury, or emotional distress. Hence, an automatic fall detection system, whether body-worn or as part

Fig. 1. Anteroposterior, mediolateral, and vertical pressure, from a young healthy

subject performing a simulated fall in the forward direction onto a mattress, whereby some attempt is made to break the impact of the fall using the hands and arms [13]. The

at approximately 25 s, and the fall



event ends at around 32 s, with the time of impact at about 28 s.. Multimodal sensor fusion of this type is becoming an emerging trend for unobtrusive and unsupervised monitoring devices. B. Falls Risk Assessment and Prevention

In light of the major negatives associated with falls, much effort is being dedicated towards their prevention. Clearly, the first step would be to single out those individuals most at risk. In addition, in light of the multifactorial nature of falling, it is also beneficial to identify the specific risks affecting each individual in order to provide the most suitable intervention. Falls risk assessment is a vast research area with widely disparate approaches being used. There are various scoring systems intended for use in hospitals, nursing homes, or outpatient settings. The available indices are designed to beused by different professionals (e.g., geriatric doctors, nurses, or physical therapists), and are based on questionnaires, observations, physical examinations, or their combination. Menz et al. examined a more comprehensive array of gait andstability patterns in the elderly, as manifested

pelvis levels, via accelerometry. Using the Physiological ProfileApproach (PPA) [50] for falls risk assessment, the authors were able to define gait parameters on smooth and irregular surfaces, in particular the harmonic ratio, that differed between individualswith varied falls risk. However, once again, an optimalcutoff point for this parameter was not provided for independent risk classification [51]. The PPA was also used by Narayanan et al., who utilized a different approach of allowing elderly subjects to perform a directed routine (STS5, AST and TUGT) in a semi-supervised manner while wearing a waist-mounted accelerometer.With the addition of reaction time testing, a significant correlation of 81% was found between the overall PPA fallscertain set of extracted time-domain features. These features were also correlated against the PPA subcomponents that include knee extension strength, body sway, vision acuity, and proprioception [52]. The importance of this preliminary wlies in the fact that elderly subjects were able to selfadminister a structured non-ADL-based movement routine and that conclusions can be drawn regarding individual deficits that increase one’s risk of falling, instead of just allocating a


www.ijmtes.com

of impact at about 28 sensor fusion of this type is becoming an

unsupervised monitoring

In light of the major negatives associated with falls, much effort is being dedicated towards their prevention.

single out those individuals most at risk. In addition, in light of the multifactorial nature of falling, it is also beneficial to identify the specific risks affecting each individual in order to provide the most suitable

ent is a vast research area with widely disparate approaches being used. There are various scoring systems intended for use in hospitals, nursing homes, or outpatient settings. The available indices are designed to be

geriatric doctors, nurses, or physical therapists), and are based on questionnaires, observations, physical examinations, or their combination.

examined a more comprehensive array of gait and stability patterns in the elderly, as manifested at the head and

pelvis levels, via accelerometry. Using the Physiological ProfileApproach (PPA) [50] for falls risk assessment, the authors were able to define gait parameters on smooth and irregular surfaces, in particular the harmonic ratio, that

d between individualswith varied falls risk. However, once again, an optimalcutoff point for this parameter was not provided for independent risk classification [51]. The PPA

, who utilized a different elderly subjects to perform a directed

supervised manner mounted accelerometer.With the

addition of reaction time testing, a significant correlation of 81% was found between the overall PPA falls risk score and a

domain features. These features were also correlated against the PPA subcomponents that include knee extension strength, body sway, vision acuity, and proprioception [52]. The importance of this preliminary work lies in the fact that elderly subjects were able to self-

based movement routine and that conclusions can be drawn regarding individual deficits that increase one’s risk of falling, instead of just allocating a

lowor high risk classification. before features can be extracted from the waveforms to estimate the risk of falling. In order for a fully unsupervised assessment to be achievable, this segmentation process must be automated [53]. C. Energy Expenditure

Another important trend in society is an increased awareness to problems like obesity and lack of physical exercise, which have reached epidemic proportions [1]. Monitoring and intervention are sought not only for weight reduction and increasing activity levels, both iadults; they are crucial in preventing an even further increase in the prevalence of chronic illnesses such as diabetes and cardiovascular disorders. Physical activity quantification is also essential in other clinical contexts (for example,rehabilitation following surgery or stroke).potentially replace the costly and cumbersome methods of doubly labeled water and indirect calorimetry, which have been considered the gold standard techniques for EE application but are not practical for everyday use. WAMS also offer an objective and easy-activity questionnaires. However,WAM application for EE still suffers from some limitations, such as correctly accounting for: EE by upper body and arms, EE duringsedentary activities (unless proven negligible), EE when carrying physical objects, and EE when walking on irregular surfaces. An interesting approach was adopted by Weiss who developed an algorithm for automated detection of near falls using a pelvic-trunk accelerometer. The hypothesis behind this work is that near falls may hold some predictive value of a subject’s risk of falling. Since near falls situations occur more frequently than actual falls, monitoring of near falls may prove advantageous required to reach conclusions; well as progress in movement classification, has improved the accuracy of detecting the type, duration, and intensity of the physical activity performed [57]. Muchhas focused on walking, since it is viewed as the single largestcontributor to daily EE, even in the elderly [58].important, as more energy is expended when walking up or down stairs than when walking on a flat terrain [61


67

risk classification. before features can be extracted from the waveforms to estimate the risk of falling. In order for a fully unsupervised assessment to be achievable, this segmentation process must be automated [53].

tant trend in society is an increased awareness to problems like obesity and lack of physical exercise, which have reached epidemic proportions [1]. Monitoring and intervention are sought not only for weight reduction and increasing activity levels, both in children and adults; they are crucial in preventing an even further increase in the prevalence of chronic illnesses such as diabetes and cardiovascular disorders. Physical activity quantification is also essential in other clinical contexts (for example, rehabilitation following surgery or stroke). Thus, WAMs may potentially replace the costly and cumbersome methods of doubly labeled water and indirect calorimetry, which have been considered the gold standard techniques for EE

tical for everyday use. WAMS also -to-use alternative to physical

activity questionnaires. However,WAM application for EE still suffers from some limitations, such as correctly accounting for: EE by upper body and arms, EE during sedentary activities (unless proven negligible), EE when carrying physical objects, and EE when walking on irregular

An interesting approach was adopted by Weiss et al. who developed an algorithm for automated detection of near

trunk accelerometer. The hypothesis behind this work is that near falls may hold some predictive value of a subject’s risk of falling. Since near falls situations occur more frequently than actual falls, monitoring of near falls may prove advantageous due to the shorter time period

Additional work since then, as well as progress in movement classification, has improved the accuracy of detecting the type, duration, and intensity of the physical activity performed [57]. Much of EE-related research has focused on walking, since it is viewed as the single largest contributor to daily EE, even in the elderly [58]. This is important, as more energy is expended when walking up or down stairs than when walking on a flat terrain [61].




www.ijmtes.com

TABLE:1

SUMMARY OF FALL DETECTION


68



5.DISCUSSION

Telehealth solutions comprise the basis of an attractive model for the treatment of chronic diseases, such as chronic heart failure or chronic obstructive pulmonamaturation of such applications [4]. However, due to the complexity of the targeted diseases and existing treatment networks which surround them, there will likely be a considerable lead time in the uptake of new care models. Sensor-based detection and possible prevention of falls, as the monitoring of physical activity in general, are alssuited to the telehealth care strategy and may be more quickly integrated into the clinical arena due to a reduced complexity in the care delivery model required. In this paper, we have provided a review of ambulatory monitors and their clinical applications, with an emphasis on those associated with the monitoring of falls and the estimation of falls risk. 6.CONCLUSION

The hardware and software design of an embedded monitoring system for real time applications is presented in this paper. Vibration signals have been analyzed to detect the mechanical faults. The implementations of analysis technique in time and frequency domain are given. The proposed system imbalance detection technique is verified with different level of severity. The potential of using wearable sensor systems for a wide range of clinical applications has moved beyond the theoretical scope and has even reached the commercial stage for falls detection and physical activity monitoring. Further research is nonetheless required in order to resolve an array of outstanding issues and enable mainstream utilization of such systems, particularly in an unsupervised longwhich is viewed by most as the leading objective due to anticipated global demographic trends. REFERENCES [1] World Health Organization, [Online]. Available: http://www.who.int/topics/en/ [2] S. R. Lord, C. Sherrington, and H. B. Menz, Falls in Older People: RiskFactors and Strategies for Prevention. Cambridge, U.K.: Cambridge Univ. Press, 2001. [3] B. G. Celler, N. H. Lovell, and D. Chan, “The potential impact of hometelecare on clinical practice,” Med. J. Australia, vol. 171, pp.1999. [4] P. van de Ven, A. Bourke, C. Tavares, R. Feld, J. Nelson, A. Rocha,and G. O. Laighin, “Integration of a suite of sensors in a wireless healthsensor platform,” in Prof. IEEE Sensors Conf., 2009, pp. 1678 [5] M. Chan, E. Campo, D. Estève, and J.-Y. Fourniols, “Smart homes– Current features and future perspectives,” Maturitas, vol. 64, pp.90–97, 2009. [6] A. Godfrey, R. Conway, D. Meagher, and G. ÓLaighin, “Direct measurement of human movement by accelerometry,” Med. Eng. Phys.30, pp. 1364–1386, 2008. [7] P. Bonato, “Wearable sensors/systems and their impact on biomedicalengineering,” IEEE Eng. Med. Biol. Mag., vol. 22, pp. 18 [8] S. L. Murphy, “Review of physical activity measurement using accelerometers in older adults: Considerations for research design andconduct,” Preventive Med., vol. 48, pp. 108–114, 2009. [9] S. J. Preece, J. Y. Goulermas, L. P. J. Kenney, D. Howard, K. Meijer,and R. Crompton, “Activity identification using body-mounted sensors


www.ijmtes.com

Telehealth solutions comprise the basis of an attractive model for the treatment of chronic diseases, such as

or chronic obstructive pulmonathe maturation of such applications [4]. However, due to the

es and existing treatment networks which surround them, there will likely be a considerable lead time in the uptake of new care models.

ossible prevention of falls, well as the monitoring of physical activity in general, are also well suited to the telehealth care strategy and may be more quickly integrated into the clinical arena due to a reduced complexity in the care delivery model required. In this paper, we have provided a review of ambulatory monitors and their clinical

lications, with an emphasis on those associated with the monitoring of falls and the estimation of falls risk.

The hardware and software design of an embedded monitoring system for real time applications is presented in this paper. Vibration signals have been analyzed to detect the mechanical faults. The implementations of analysis technique

omain are given. The proposed system imbalance detection technique is verified with different level

The potential of using wearable sensor systems for a wide range of clinical applications has moved beyond the

ached the commercial stage for falls detection and physical activity monitoring. Further research is nonetheless required in order to resolve an array of outstanding issues and enable mainstream utilization of such

long-term context, which is viewed by most as the leading objective due to

, [Online]. Available: http://www.who.int/

, Falls in Older People: Risk . Cambridge, U.K.: Cambridge Univ.

[3] B. G. Celler, N. H. Lovell, and D. Chan, “The potential impact of home , vol. 171, pp. 518–521,

[4] P. van de Ven, A. Bourke, C. Tavares, R. Feld, J. Nelson, A. Rocha, and G. O. Laighin, “Integration of a suite of sensors in a wireless health

, 2009, pp. 1678–1683.

Y. Fourniols, “Smart homes , vol. 64, pp.

[6] A. Godfrey, R. Conway, D. Meagher, and G. ÓLaighin, “Direct Med. Eng. Phys., vol.

[7] P. Bonato, “Wearable sensors/systems and their impact on biomedical , vol. 22, pp. 18–20, 2003.

[8] S. L. Murphy, “Review of physical activity measurement using older adults: Considerations for research design and

[9] S. J. Preece, J. Y. Goulermas, L. P. J. Kenney, D. Howard, K. Meijer, mounted sensors-

a review of classification techniques,” R1–R33, 2009. [10] M. J. Mathie, A. C. F. Coster, N. H. Lovell, and B. G. Celler, “Accelerometry: Providing an integrated, practical method for longambulatory monitoring of human movement,” pp. R1–R20, 2004.


69

eview of classification techniques,” Physiol. Meas., vol. 30, pp.

[10] M. J. Mathie, A. C. F. Coster, N. H. Lovell, and B. G. Celler, “Accelerometry: Providing an integrated, practical method for long-term,

ement,” Physiol. Meas., vol. 25,

sensors-based wearable systems for monitoring of human...

Documents