Remote Monitoring Systems For Fall Detection In Elderly Population

Challenges in Monitoring Elderly Population

The Elderly populace has been one of the principle worries in most of the nations amid the most recent decade. Most elderly individuals experience the ill effects of more extensive range of different sicknesses and more crisis circumstances, for example, fall are probably going to happen. Subsequently, they should be taken to the nearby clinic, where these injured elderly people shall be watched along with medicinal aid if wellbeing state to be maintained. In the meantime, the measure of old individuals keeping up their autonomous ways of life is developing quickly, that creates it tougher for restorative experts to take after variations in patient’s wellbeing conditions when the fall happens. Be that as it may, remote observing for elderly fall in toilet can be anticipately portrayed situation, essentially it shall help in decreasing medicinal services costs and in the meantime, keep up patient’s autonomous way of life (Maged N Kamel Boulos, Steve Wheeler, 2010). Hence, there is a reasonable request in solid multi-practical remote observing frameworks for elderly individuals, which gather and join distinctive wellsprings of therapeutic information comparing to ordinary routine of the checked patient. By and large, extraordinary segments including the frameworks are crumbled and working independently from each other. Be that as it may, if we consolidate observing parts that is toilet alarm fall sensors into keen situations, then we will have the capacity to complete perceptions for individuals with different ceaseless circumstances at home. It shall enhance old person’s level of flexibility in addition to wellbeing conditions, this being the principle issues in monitoring the elderly people fall especially in toilets or other places (Maged N Kamel Boulos, Steve Wheeler,2010). 

The elderly people treatment costs are relied upon to develop quickly amid the next decades because of the maturing of “infant boomer’s” era. In the meantime, crisis circumstances related with older people fall are thought to be important issue in today’s circumstances. In this project, we shall define structure that utilizes smartphone innovation mutually with functional information checking to recognize falls. The framework does gathering, putting away and preparing of increasing speed information with additionally caution producing and exchanging every one of the estimations to remote parental figure. To perform assessment, some fall detection characteristics, framework for toilet sensors to determine falls and methodology along with algorithm to acquire reasonable fall information are all defined in this report. A drop discovery calculation has been intended to adapt to vast varieties of development in the middle. The on-line calculation working demonstrated execution consequences of 90% specificity, 100% affectability and 94% exactness (C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2012). 

Integrating Monitoring Components in Smart Environments

In the meantime, fall episodes are thought to be a standout amongst the most well-known and unsafe dangers among elderly populace, with almost 50 per cent of elderly inhabitants along with 40 per cent of freely active individuals dropping every year in the bathroom or other areas (Marcin Bajorek and Jedrzej Nowak, 2011). In this way, present day social insurance frameworks tend to incorporate dependable fall location usefulness into general checking structure. With the current advancement on information technology advertise toilet sensors are frequently sent in combination with ecological gadgets to enhance drop identification rates along with limit false alerts. For this situation, a multi-modular framework necessitates an extraordinary combination calculation to join all the dynamic segments. Another wellspring of inspiration is an absence of relevant information in a greater part of current social insurance frameworks. In this way, shrewd homes, with a capacity to subtly gather relevant information are basic wellsprings of data. This information can be handled a short time later and induce genuine exercises, giving an additional knowledge on physiological procedures occurring with elderly. In any case, there are no conspicuous answers for coordinating medicinal sensors into a brilliant home condition, which makes this zone open for additionally explore examinations (Damien Brulin and Estelle Courtial, 2010).

Consistently, 33% to one-portion of the populace over 65 years of age have experience falls in the bathroom areas. Of these elderlies who do drop in the restroom, half of them has happened more than once. Falls are the main source of damage in more established grown-ups and the main source of unplanned demise in those 75 years old or above. Over 90% of hip broke due to fall in the toilets, with a large portion of these cracks happening in people more than 70 years old. Treatment of the wounds and intricacies related with falls is huge almost in billion dollars yearly. It is normal after a fall that an elderly individual can’t get up independent from anyone else or summon offer assistance. There is hence a requirement for a programmed fall recognition framework in which a patient can summon help regardless of the possibility that they are oblivious or unfit to get up after the fall. There are as of now a few items available endeavoring to address this issue have achieved commercialization. Notwithstanding, these items require the fall discovery gadget to interface to a stationary base station, which is frequently an independently obtained item. This sordid position, set midway in anyone’s home along with connected to Bluetooth enabled mobile phones, at that point smartphone a call place for offer assistance or message alerts were sent to friends and relatives. The weakness of these items is that they all require a go-between call benefit which adds up to a robust month to month expense. Additionally, they are altogether constrained to the scope of one’s home since they rely upon the focal station for outdoor communication. On the non-marketed part of advancement, the major share of investigation in fall detection in toilets for elderly people is unified around outlining novel productive calculations for deciding tumbles from non-drops in the bathrooms. The labor about fall identification is recognized with help of gear and with help of components removed from toiler sensor information. The principal method depends on accelerometers, it is a gadget that can recognize the size as well as speeding up of device along a specific hub; typically, accelerometers are utilized are 3 hub (Pannurat N., Thiemjarus S., Nanta jeewarawat, 2014). They can likewise figure one’s point in connection to the ground by identifying the quickening of surface because of gravity in the ground. The second primary approach utilizes spinners, which measure introduction. A whirligig comprises of a turning wheel whose hub can take some quantification that shall introduce along one or numerous tomahawks. Utilizing spinners, it is conceivable to decide one’s introduction and changes in introduction, which can be utilized to figure rakish speed and quickening action (Pannurat N., Thiemjarus S., Nanta jeewarawat, 2014).

Existing Products and Technologies for Fall Detection

 With a maturing populace, our gadget intends to fulfill the developing neglected requirement of drop location as well as counteractive action. The goal of toilet sensor is to outline plus make a Fall Recognition Structure for the older people. The framework is a fall checking gadget that can connect remotely through a pre-modified Tablet with Bluetooth enabled cell phone. After recognizing a fall, the gadget discusses remotely with the portable smartphone to emergency services as well as crisis contacts. The gadget likewise identifies unusual slope as well as cautions the near one to redress their stance to limit the danger of falling in the toilet. This framework is extraordinary from current business gadgets for a few ins and outs. This exercise e shall be fulfilled by joining the framework with the person’s current phone/PC, which likewise limits and extra setup expenses for Bluetooth device, sensors, batteries along with other equipment. This is subject to the way that the up and coming era of elderly  are generally alright with innovation and doubtlessly effectively possess a mobile phone as well as PC. The gadget is likewise special in that it bids a physical cancelation catch in case of a incorrect caution or minor drop that the client could recuperate from. Another favorable position our framework has is that it permits portability past the scope of the house. Remote connecting with a phone takes into account security anyplace there is cell benefit. For more protracted avoids home, about as soon as going by a family member home, remote connecting with one’s own portable smartphone or tablet gives versatile insurance that a settled base station shall not be able to permit. This toilet senor gadget shall offer an extensive variety of selectable ready techniques should the elderly person having hearing-disability (Mubashir M., Shao L., Seed L,2013).

As a reaction to the maturing populace, current social insurance showcase gives an extensive variety of restorative gadgets for remote measuring of indispensable wellbeing parameters. Most of the gear is customized and abuse for position checking, also it is not ready to give a consistent review of the person’s wellbeing situations. In addition, diverse strictures are restrained independently since the observing procedure are not coordinated. In the meantime, present day cell phones are furnished with cutting edge sensor usefulness, which has an incredible potential for the medicinal services, however is generally abused in amusement industry. Expecting specified conditions and beforehand directed investigation concentrated on android based checking framework for patients with interminable prominent illness we can plan two principle difficulties of exhibited inquire about (Mubashir M., Shao L., Seed L,2013):

• An absence of learning and adequate expertise for persistent therapeutic information examinations.
• Wasteful knowledge on the connection between deliberated parameters. 

Use of Smartphone Technology for Data Monitoring and Processing

Specific conditions incorporating maturing of populace in created nations, expanding expenses of essential human services and solid request in free living, have effectively evoked a serious study work in distant checking zone. Typically, these sorts of frameworks are sectioned into 3 noteworthy areas: sensor sheet, correspondence sheet and guardian. This shall be especially intrigued by the initial 2 classes in charge of gathering, exchanging and handling of the gushing information. Generally checking process includes both wearable and ecological sensors, gathering information for additionally preparing and envisioning. Be that as it may, these two sorts of data channels are overseen independently and seldom exhibited as a consolidated structure. 

Framework Components

The framework comprises of a ultrasonic sensor as well as an infrared sensor with Arduino panel. Ultrasonic sensor is a nearness sensor which can quantify separation of articles, inside the predefined go and with no physical contact. Infrared Ray (IR) sensor comprises of a producer and a beneficiary. The producer radiates infrared beam consistently. The diverted beam shall be gotten by the collector. The beneficiary goes about as a transistor with its base current that will be controlled by the force of light got. Arduino is an open-source physical registering stage considering a basic microcontroller board, as well as an advancement situation for comprising software development for the board.

Applied Design  

A model house was built in view of the proportion 1:11 of genuine protest as appeared in diagram 1 below. Every one of the figuring’s depended on the model line. The toilet zone was 12 cm, can region was 10 cm and bowl is 8.5 cm. The thought is to outline a device that can recognize fall as well as to recognize conduct without giving false caution. By utilizing toys of perfect stature, proficiency of ultrasonic sensor in recognizing fall alongside the best point has been done. Both of the sensors were put opposite to each other to build its effectiveness as appeared in diagram 2. 

Diagram 1

 

Diagram 2

Fall Detection

The falls are amongst significant issues in current service providers and a genuine danger for elderly populace. Thus, the vast majority of the remote checking frameworks have a tendency to incorporate programmed fall recognition into their usefulness. Present day cell phones are frequently outfitted with an arrangement of intense sensor innovation and begin to assume a huge part in social insurance advancement. Late investigations demonstrated that accelerometer, gyrator and magnetometer can include a free fall recognizing instrument or be a piece of the fall location structure. Normally, increasing speed information is gathered and put away on the cell phone with ensuing on-line or disconnected preparing relying upon the present conditions. Then again, a portion of the examinations propose calculations where logical or visual information gathered by ecological devices is sent to recognize a drop. For this situation prominence of the procedure is moderately low since elderly people don’t require to wear any gadgets. In the meantime, these kinds of frameworks are regularly confronting protection issues and require extra moral affirm. Because of these reasons and intricacy of the fall procedure as a rule a few endeavors were made to consolidate the two sorts of information to enhance general execution of fall location frameworks. In the accompanying segment we give principle fall qualities, portray well known methodologies and clarify how drop location shall be incorporated into a over-all observing model executed in a savvy home environment or at any hospital or care clinic. 

Challenges of Integrating Medical Sensors into Smart Home Environment

Characteristics  

A drop is ordinarily characterized as “unexpectedly stopping on the ground, floor, or other lower level”. Trailing the adjust as well as ensuing dropping with the assistance of a right hand additionally considered as a fall (Koshmak G., Loutfi A., Linden M, 2016). In view of conceivable situations 4 primary sorts of falls shall be recognized:

• tumble from resting,
• tumble from sitting,
• tumble from strolling/standing and
• tumble from remaining on help instruments,

 Each sort has its own one of a kind attributes, which can help engineers to adjust fall indicator stages to a more extensive range of client prerequisites. Normally all the cutting edge fall discovery frameworks can be part into 3 principle classes relying upon the sensor innovation conveyed for observing: wearable sensors, surrounding sensors and vision-based sensors. Most of the wearable fall identifiers depend on accelerometer information and working with stance and movement of the patients’ body (Koshmak G., Loutfi A., Linden M,2016).

They can also be split into thresholding strategies as per the handling calculation they send. Speeding up information gathered amid the fall in various ways is shown on Figure. Each line speaks to crude, pitch or yaw of the cell phones arrange hub, has its novel variety and can unmistakably portray three diverse upsets of each fall movement: (1) pre-fall, (2) affect, (3) after-fall stage (Koshmak G., Loutfi A., Linden M, 2016).

Convey present day vision or surrounding strategies to identify a fall. For this situation gathered estimations are transferred to a remote gadget and investigated for conceivable crisis circumstances related with falls. In by far most of the setting mindful or image based frameworks drops are identified disconnected with the assistance of factual calculations. Every one of the introduced methods still gives a lot of false optimistic cautions despite the fact it iss working autonomously. It is thusly essential to coordinate extra sensor usefulness keeping in mind the end goal to enhance unwavering quality of fall location frameworks. This pattern is getting to be noticeably well known and tended to as multi-sensor combination based fall location. For this situation, a few sensor frequencies are sent to gather information that is later intertwined on a handling level. In the accompanying areas, we keep on describing each sort of fall recognition method specifically and talk about conceivable answers for a multimodal system melding the two procedures (C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2011). 

Alarm Sensor  

Sensor cushion gets set over the feet on the underside of the latrine situate. The line on the cushion, at that point gets associated with the fall counteractive action screen. At the point when the patient or occupant endeavors to get up from the can and weight is expelled from the sensor cushion, the screen will caution to alarm the elderly people (Maged N Kamel Boulos, Steve Wheeler, Carlos Tavares, and Ray Jones,2011). 

• Gives elderly persons or inhabitant security while utilizing the restroom.
• Reduces fall by alarming when patient or occupant endeavors to get up from the latrine.
• The weight detecting cushion triggers the caution when weight is discharged from the sensor cushion. 

Context-Aware Fall Detection 

Logical information has been as of late sent to do fall identification along with activity acknowledgment of old individuals in their homebased condition. It shall depict an approach, with the primary thought to intertwine dissimilar types of information source stations into an exceptional engineering, getting information from basic identifiers such as Bluetooth based device or camera. This framework are also able to accomplish on-line or back handling to determine stance or introduction of the client plus trigger an alert if there should arise an occurrence of fall hazard. Different endeavors were made to enhance this procedure by presenting extra wellsprings of datalike encompassing sound caught by amplifier exhibits (Vavoulas G., Pediaditis M., Spanakis E.G., Tsiknakis M, 2014) or current location of the person. On the other hand, the device is deployed in ponder for healing center fall counteractive action. Additionally, new contemplate exhibits how drop avoidance framework types utilization of collected data from devices with a specific end goal to control as well as counsel elderly people or even to offer guidelines to treat a strange condition plus lessen the fall risk. For this situation checking and handling information from sensors is performed by a cell phone that will send notices to the persons’ phone and in emergency circumstances send them to a parental figure. Additionally, association between speeding up of body’s focal point of gravity amid sit to walk motion along with a procedure of falling. The consequence of investigation by utilizing files with a significant difference uncovered a 90 percent right forecast rate for falling. Nonetheless, there is as yet a moderately abnormal state of false alert created associated with the setting based recognition and aversion frameworks, which can be potentially enhanced by incorporating with a wearable device. It was beforehand examined that wearable sensors with inbuilt accelerometer can fill in as a compelling human services gadget. The clear majority of them have as of late been sent for precise fall identification demonstrating dignified comes about amid assessment handle. This shall be intrigued in accelerometer-based device both as an individual segment and as separated of the multi-modular framework, where it is consolidated with environmental devices. For situation, a Bluetooth enabled gadget is supplanted with a mobile phone, which can fill in as fall indicator sensor and entryway mode at the same time (Naruhiro Shiozawa and Shin Arai, 2012). 

Method of Fall detection via using Accelerometer

The accelerometer shall be utilized as a part of different examinations and applications to unbiasedly screen a scope of human development, for instance to gauge metabolic vitality use, physical movement levels, adjust and postural influence, step, and to identify falls. The observing framework appear in diagram shown below. These framework having equipment part .It shall be append able on the midriff of the individual plus a microcontroller is utilized for arrange individual’s activities as well as distinguishing any conceivable falls. In the event of the falls, the framework likewise conveys a caution to the fitting reaction unit. The initial segment of the framework incorporates a fall-distinguishing band for separating and handling signals gotten from the triaxial accelerometer. In this framework, an ADXL335 accelerometer utilized for the drop discovery as shown on X, Y and Z pivot. After that these sensors shall send the simple flag to the microcontroller for its coherent control for distinguishing the ongoing position of the body to the server to refresh and show the actual data. The gadget additionally incorporates a crisis enable catch to show the fall alarm and crisis flag. The second piece of the framework comprises of a GSM modem outline that shall relate to the microcontroller for sending the message to the individual places alongside the area of client. The server initially got a fall caution, and afterward it produces alert. In fall recognition procedure, the estimations of accelerometer are the contribution to microcontroller to Arduino board for computational reason. The three tomahawks shall each deliver an alternate speeding up esteem, in light of these esteem utilize arrange examination to assess clients development signals. Developments of individual will be ordered into ordinary developments and anomalous developments. As indicated by speeding up esteems from accelerometer, ordinary activities will be nonstop and cadent development signals, while unpredictable activities should be perceived as fall flag. In this strategy check the estimation of three tomahawks with particular characterized limit. In the event that esteem is beneath the edge at that point produce caution. It’s capacity to decide diverse exercises under various conditions. Accelerate the action state distinguishing proof process and prompt quicker fall recognition.

 

Following figure displays how the acceleration standards of accelerometer works:

 

Fall Detection using Smartphone  

With the current advancement on versatile marketplace, cell phones begin to play an imperative part in present day medicinal services frameworks. Most recent versions equipped with an accelerometer device are usually utilized as drop detection apparatuses. For this situation, they supplant both processing mode plus a specialized instrument while keeping up moderately small size. A decision of preparing calculation relies upon definite application of the framework and fluctuates in various investigations. A portion of the current implementation methods put on Gaussian appropriation of grouped information, neural system ( Gjoreski H., Kozina S., Gams M., Lustrek M, 2014) and machine learning procedures. Nevertheless, furthermost of them are at first in view of three fundamental parameters associated with falls: effect, speed and stance. As indicated by the current scenario, combining effect and stance while examining the fall case is sufficient to create a solid calculation. Based on explore questions planned before, we incline to grow a fall recognition framework and research its additional combination into a life-sized checking framework with extra sensor usefulness including medicinal gadgets plus ecological sensors. Comparable method was accepted in a few investigations with goal to join logical data with basic accelerometer estimations misusing latency and location sensors (Federico Pecora, Marcello Cirillo, Francesca Dell Osa, Jonas Ullberg, and Alessandro Saffiotti, 2013 ). This shall explore a novel fall detection method that uses increasing speed, stance and setting information, where setting can be displayed by natural sensors and individual profiles. Remote accelerometer, 3-D camera and amplifier are being simultaneously handled by Leone et al to achieve a superior outcome in fall hazard appraisal. All the introduced ponders, be that as it may, are inadequate with regards to a solid melding procedure to join preparing comes about because of free parts. In work by an off-the-rack programmable detecting stage called Sun SPOT is utilized for information recording. Setting  data is exhibited in a few classifications, covering primary parts of elderly living:

• Physical movement;
• Physiological condition;
• Individual wellbeing record;

Every classification speaks to a different variable or preparing result (counting fall caution from detached calculation) and converged into a Bayesian system for facilitate measurable investigations. In the flow examine, we make an endeavor to build up an autonomously working fall recognition framework, which can be effortlessly converged with different sorts of sensors and incorporated into a bigger checking condition. The figure shown below shows the co-ordinates send for person fall on smartphone through the message (Maged N Kamel Boulos, Steve Wheeler, Carlos Tavares, and Ray Jones,2011):

Dependable fall recognition is one of the means towards general observing framework for elderly individuals with various sorts of infection. It ought to be consolidated with different wellsprings of therapeutic data diverts keeping in mind the end goal to give a superior understanding on person restorative conditions. This can be effectively executed by means of current brilliant home conditions as a multi-modular stage with melding abilities. Savvy homes by and large are a piece of the Ambient Assisted Living range, in charge of persistent observing of elderly individuals in agreeable home condition. As of late, an expanding number of activities have been founded on this approach suggesting different parts and applications. Exceptional sustenance guide was proposed as an endeavor to enhance physical condition for elderly individuals with diabetes. It shows the likelihood to make wholesome administration a great deal more successful through conveying Ambient Intelligence frameworks. In addition, this shall give their Necessity framework versatile checking capacities and comprehensive assessment technique which was incorporated in the improvement procedure. It shall make an endeavor to manufacture a shrewd elderly locally established on an android gadget, which is used as a 3-hub accelerometer gadget to distinguish a fall of the bearer. We have a tendency to use savvy home condition with a specific end goal to manufacture a long term observing stage for elderly individuals. In the meantime, some of its capacities can be utilized to gather relevant information, which makes it an extra wellspring of data. It shall make an endeavor to fuse picture handling and simulated methods in light of Platform for the Automatic Construction of Organizations of Intelligent Agents (PANGEA) platform (Gabriel Villarrubia, Juan F De Paz, Javier Bajo, and Juan M Corchado, 2014). The framework is introduced for a situation examine planned utilizing diverse specialists and sensors in charge of giving client bolster at home in case of occurrences or crises. Designers shall take day in and day out observing of solid more established individuals above and beyond by refining it and making it accessible to more established individuals with various endless infection. A specific exertion is made on correspondence with the client sending different sorts of intuitive gadgets like Tablet, Smartphone, Notebooks etc.. The key thought is to expand autonomous life at home shall not evade hospitalization for longer periods. The quantity of productions in social insurance space is relied upon to develop quickly. There are both fruitful executions and studies which require extra research exertion. The conspicuous pattern seen in most of the current distributions can be portrayed by consolidating disconnected wellspring of information into incorporated system. Be that as it may, it is as yet not clear how to prepare gathered estimations or which set up will be best in the greatest measure of observing examples. In this way, we trust it is essential:

• To consolidate both wearable and logical information to have the capacity to modify created framework for various sorts of clients and observing situations,
• To make an adaptable mix stage to have the capacity to include/evacuate sensors depending specific client necessities.

For this situation, a proficient and adaptable calculation is required for combining different wellsprings of information before the preparing stage. In this way, as a piece of the exploration procedure, we directed a writing seek brought about a investigations inside the multi-sensor combination based fall location. Mix of relevant and wearable information considering savvy home stage for successful location of crisis circumstances related through drop is a stage in the direction of the full-scale observing framework for elderly.

Procedure 

As it was beforehand detailed above, fundamental difficulties of the proposed think about incorporate:

• an absence of information and adequate expertise for nonstop therapeutic information examinations and
• an absence of proficient intertwining calculation for overseeing irrelevant information sources
• wasteful understanding on the connection between’s different parameters.

All difficulties speak to various sorts of examinations associating between each other, and along these lines can be drawn closer with cross-disciplinary research. An absence of learning for ceaseless information examinations can be overwhelmed by acquainting extra information sources with the observing procedure. With a specific end goal to give a superior understanding ona relationship between’s deliberate parameters we can convey multi-sensor combination calculation and join distinctive sorts of information into a solitary source channel (Bagala F., Becker C., Cappello A., Chiari L., Aminian K., Hausdorff J.M., Zijlstra W., Klenk J.,2012). The decision of calculation for this situation can be founded on writing pursuit and survey of the current investigations inside multi-modular observing zone. Subsequently, we initiated a multi-disciplinary research work including segments like sensor administration, flag handling, android improvement, counterfeit consciousness and restorative information expertise. Each range compares to a specific phase of the checking procedure and along these lines requires autonomous research approach, equipment and programming setups. The imperative segment of any remote demonstrative framework is a sensor layer, which oversees the ceaseless information accumulation. Quickly developing business sector of restorative gadgets enable us to send distinctive sorts of sensors and extend the quantity of vitals sings to catch. Additionally, to set up dependable correspondence channel amongst sensors and handling gadget, Bluetooth innovation is used including Health Device Profile convention particularly intended for therapeutic applications. Various calculations and plans from “flag preparing” zone are conveyed for overseeing sensor information on both pre-and post-preparing level with a specific concentrate on sensors combination and fall location techniques. In request to build up the most proficient calculation for incorporation of gathered estimations into a typical system we direct a writing study with a specific concentrate on multi-sensor combination procedures for fall recognition. It will help us to show signs of improvement understanding on as of late requested approach and systematize the present endeavors in this novel region.

In the meantime, all on-line examinations and different strategies are executed on android gadget utilizing android API and most recent accessible programming. Extra exertion will be devoted to the decision of potential beneficiary and misuse of the yield (i.e. outside capacity, caution warning, information representation) and elderly view of the created innovations. Finally, evaluation of the present examinations is executed through a progression of reproductions and genuine tests for both on-line and disconnected information collection. These particular concentrate is made on dynamic contribution of elderly populace into assessment prepare in a joint effort with the neighborhood therapeutic specialists. This activity will advance separation human services for elderly individuals and enhance the level of participation. After preparatory conferences and common assent, we propelled a progression of internet checking process including elderly individual with revealed danger of falling and a gathering of parental figures. We figured out how to set up a solid correspondence between the patient and therapeutic staff, perform day by day checking and produce cautions in the event of crisis. Gathered information can be later sent for advance examinations and give a superior understanding on physical action of the elderly individual who has a place with a hazard gathering (Becker C., Schwickert L., Mellone S., Bagalà F., Chiari L., Helbostad J.L., Zijlstra W., Aminian K., Bourke A., Todd C, 2014).

Venture  

The principle motivation of the venture corresponds with our examination points and requires sending of physiological and ecological sensors for nonstop observing of elderly individuals in their homes. A few extra functionalities are intended to be executed to improve the framework including persistent information gathering of day by day movement and physiological parameters from circulated sensors, long haul incline investigation, information introduction by means of a customized interface and social communication between essential clients (elderly individuals) and optional clients (parental figures). 

This connection is actualized via robot which is teleoperated by auxiliary clients also shall move to get the information from home and converse with elderly people. The rundown of optional clients incorporate family, companions, casual and formal guardians and wellbeing experts, who can get to the framework through their smartphone. An exceptional classification is spoken to by medicinal work force, who can get to the data gathered on-line and examine drifts in information.

Information accumulation is sorted out with the assistance of physiological and natural devices, which are carefully put inside the condo. The rundown of sent ecological actuators can shift contingent upon checking necessities, however typically incorporate RFID labels, weight mats, switcher sensors and so forth. Gathered information is ceaselessly prepared with setting acknowledgment system, which can construe specific patient’s movement (cooking, working out, cleaning up, dozing), performed amid the day. The decision of physical devices depends completely on client’s wellbeing circumstances and, in this way, entirely person. Right now conveyed gadgets are beat oximeter, circulatory strain sensor as well as electronic weights, which perform intermittent spot checking and send gathered information to the worldwide database for additionally preparing. We use framework as an essential stage and make an  endeavor to enhance its abilities by presenting android-based persistent observing of physiological limits. This examination philosophy can enable us to join disconnected wellsprings of information immature design, to test different combination calculations on genuine dataset and survey real parameters of made framework (i.e. unwavering quality, interoperability, client acknowledgment). The primary concentrate is made on fuse of the fall identification calculations in light of increasing speed information into an observing structure. Essentially, we are required to perform joining into a formerly created engineering, which can likewise survey adaptability of picked approach as far as information synchronization. As a further work, we intend to build up a cell phone based calculation, sending relevant information from the savvy home in conjunction with physiological measurements.

The goal of this venture shall be to plan and make a Toilet sensor for Fall recognition in the bathrooms for older people. The framework must be equipped for remote correspondence with a portable workstation phone. The gadget must have the capacity to recognize unsafe slope besides if a drop had happened. In case of a drop or hazardous slope, the gadget must have the capacity to caution the client and others. To have the size to distinguish falls, the gadget initially must have the size to detect movement plus the distinctive computable qualities required with movement. Detecting in the device starts with a computerized tri-pivot accelerometer, which measures speeding up along the three arrange tomahawks (C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2012). A simple bi-hub spinner measures the pitch and move of the gadget as precise speeds. To utilize these sensors to identify falls, the sensor readings must be yielded to a microcontroller for handling and application to calculations. For this to happen, first the sensor readings are changed over from a simple voltage flag to a discretized bit an incentive for the microcontroller to have the capacity to utilize them(C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2012). This is expert by passing the sensor yields through an ADC (Analog to Digital Converter) before entering the microcontroller. The microcontroller needs to take the discretized bit information from the Analog to Digital Converter and apply diverse equations and transformation variables to ascertain the vital elements. Utilizing these variables, the microcontroller nourishes them into a calculation, contrasting the sources of info and different limit esteems, starting triggers when certain edges are met or surpassed. After recognizing unsafe tilt, the microcontroller needs to start a short capable of being heard, material, and visual cautioning to the client to amend their stance. The client is cautioned by means of sight, sound, and touch if there should arise an occurrence of weakness of one of the faculties. After recognizing a fall, the gadget starts a constant capable of being heard, material, and visual cautioning. The client is then given a time of 20 seconds in that to cross out the caution in the example that the drop is not genuine plus the client can recover their levelheadedness all alone. If, left uncancelled, the fall is observed as genuine and an alarm is conveyed (C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2012).

All through the procedure of sensor readings, calculation triggers, fall discovery, and so forth the gadget is persistently synced with a portable smart phone. This is finished utilizing a Bluetooth device appended to the microcontroller for yield. The synced portable smart phone always screens the status of the gadget; it gets plus stores all sensor information against its hard drive besides is refreshed with any calculation trigger deviations. The portable workstation is then ready to send a caution to the specialists and additionally crisis contacts (C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2012).

Design Process 

Hardware to be Used: The subsequent plan for the systems shall be included in deciding the fitting Bluetooth module, sensor sheets, peripherals, microcontroller board, , plus the proper packaging to plinth all the hardware in (C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2012).

• Microcontroller:

To play out every important handling, we shall be using the ATmega328 microcontroller equestrian on the Arduino Duemilanove panel. A portion of the applicable details for this microcontroller panel are unstated underneath (C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2012). 

The Arduino Duemilanove is an extremely well-known board amongst specialists along with that is the microcontroller foremost group of decision while constructing little model activities. Along these lines, there are broad instructional exercises and open source cases accessible to encourage learning and acclimating oneself with the board (C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L, 2012). Also, we picked this board as a result of the accompanying attributes: 

• Working voltage: The working voltage shall be 5V along with 3.3V alternative fitting  since these sensor sheets as well as Bluetooth shall work under 5V power which shall yield understandings in the scope of 1-5V. 

• Input voltage :  The Arduino board has a worked in voltage controller that permits an information voltage scope of 7-12V, which is appropriate on the grounds that we intend to control the board with a 9V battery (Maged N Kamel Boulos, Steve Wheeler, Carlos Tavares, and Ray Jones,2011). 

• Number of pins – The extensive no of advanced I/O pins (14) is fitting since we required to interface with a few peripherals (He J., Hu C., Wang X.Y,2014). The quantity of simple information pins (6) is adequate on the grounds that we need numerous pins to get contribution from our sensor sheets (He J., Hu C., Wang X.Y,2014). 

• Memory – The blaze memory is proper in light of the fact that our calculation projects shall be genuinely long and require a tolerable measure of memory on the microcontroller to store them (He J., Hu C., Wang X.Y,2014). The SRAM (2KB) is a petite on the low side, though the calculations shall work around this by not putting away an excessive number of factors, to not debilitate the SRAM limit (He J., Hu C., Wang X.Y,2014). 

• Specialty pins – The Arduino Duemilanove accompanies RX/TX pins, that shall be utilized for serial correspondence with the Bluetooth device. The board moreover accompanies I2C perfect pins, that shall be vital to border with the advanced accelerometer (He J., Hu C., Wang X.Y,2014). 

• USB correspondence and programming condition – The Arduino conveys as well as stacks plans from a system by means of a USB connection; it can likewise be controlled by USB. This is beneficial and less complex than different microcontrollers which shall require serial port correspondence. The Arduino user interface design dialect and condition is exceptionally easy to understand (He J., Hu C., Wang X.Y,2014).

Sensors: For proper fall recognition, our gadget required two distinct sorts of sensors:

• Accelerometer

We chose the ADXL345 triple hub computerized accelerometer to use in our device. It has a wide settable g run (up to ±16g), which is appropriate on the grounds that falls shall reach up to 8 g’s. Since it is computerized, its determination 13 can likewise be set, there is less voltage clamor, and there is less focusing/adjustment mistake: a 0 speeding up perusing will really give a 0, instead of simple accelerometers that shall gave a non-zero power notwithstanding for 0 increasing speed. Additionally, the ADXL345 requires 3.3V power and is I2C perfect, so that microcontroller panel would interface be able to with it accurately (He J., Hu C., Wang X.Y, 2014). 

• Gyroscope

We chose the LPR530AL double pivot spinner to utilize in the device. It has a tall rakish speed run (±300°/s), that is suitable for the circumstances qualified while falling (He J., Hu C., Wang X.Y,2014). This gyro panel additionally has worked in separating and enhancement (4x), and requires 3.3V power and yields in the 0-5V territory, this microcontroller panel shall interface accurate result ((He J., Hu C., Wang X.Y,2014)).

Bluetooth: 

For remote communication, we used Bluetooth technology as the technique. This is because numerous present day convenient devices (PCs, telephones, GPRS, and so forth.) are promptly perfect with Bluetooth (Maged N Kamel Boulos, Steve Wheeler, 2010). The Bluetooth modem picked is BlueSMiRF gold with a RN41 Class 1 Bluetooth module plus implicit receiving wire. We picked this modem to be specific since its pins as well as power are 5V perfect. It likewise underpins RX/TX serial correspondence from 9600 to 115200bps (bits every second, baud rate), which makes it completely good with our Arduino board (He J., Hu C., Wang X.Y,2014).

Peripherals:

To furnish interface with the client, our gadget has various peripherals: 

• Two catches – These two catches give contribution from the client and enable them to control the gadget. The picked catches were straightforward 4-stick pushbuttons perfect with 5V control. We utilized 10kω resistors as draw down resistors to ground (Maged N Kamel Boulos, Steve Wheeler,2010). 

• Two LEDs – These two LEDs give visual criticism to the client (Maged N Kamel Boulos, Steve Wheeler,2010). They are low voltage LEDs, so expected resistors to cut down the 5V control supply properly.

Siren & Buzzer – These devices are capable of being heard plus material criticism to the user, separately. The siren transmits a high pitch tone and the ringer vibrates. Both were decided for their similarity with 5V control (He J., Hu C., Wang X.Y,2014).

This design was utilized for troubleshooting purposes just to ensure every one of our sensors, Bluetooth, plus peripherals were working accurately. To achieve real fall testing along with information gathering, a more vigorous lodging arrangement shall be required. 

To finish some preparatory drop testing and information accumulation, we continued to utilize an off-the-rack lodging box intended for the Arduino. The case shall be elastic and rectangular fit as a fiddle with a metal cover. To house these sensors, we shall be straddling them on a littler breadboard and connected it to the metal top of the container. With this mechanical assembly, Housing Mk. I, we could assemble a lot of fall information. Be that as it may, this setup was just for information gathering; there was no space for our fringe ready segments and catches, or a Bluetooth module. We concluded that we expected to build a custom lodging for the working model.

The framework must be wearable gadget, fit for remote correspondence with a portable workstation phone. This put a few plan imperatives on the lodging of the gadget. The gadget and lodging must be sufficiently little and sufficiently light to be worn serenely and not restrain typical every day exercises.

Because of the reason for the device, the lodging must be molded in such a path in order to limit damage to the client should they fall on it. Additionally, the lodging material and configuration must be sufficiently hearty to retain affect and bear the client’s weight falling on it. To be totally remote, the gadget additionally needed to have its own power source that shall be sufficiently solid to control all the electronic segments for a drawn out stretch of time. We utilized a 9V battery since its voltage is perfect with our microcontroller board, it has a reasonably long battery life underneath these circumstances (Bourke A.K., Van De Ven P., Gamble M., O’Connor R., Murphy K., Bogan E., McQuade E, 2010). 

Mulling over these focuses, we thought of our second lodging, outlined in SolidWorks seen beneath . In this lodging, we see base is bended to form what’s more, fit the body line, and an escape clause is adjusted along the base to slide in a belt. There are openings with the goal that our microcontroller board and sensor sheets can slide in and be held safely; there are lodging gaps for the pushbuttons and LEDs situated for ideal client availability and perceivability. The rooftop slides off for simple get to. We had this outline made by a 3-dimensional prototyping machine in a lightweight, solid material known as ABS (acrylonitrile butadiene styrene); a similar material utilized as a part of Lego blocks (Bourke A.K., Van De Ven P., Gamble M., O’Connor R., Murphy K., Bogan E., McQuade E, 2010). 

Programming  

All microcontroller writing computer programs shall be completed in the Arduino software development dialect along with condition. The smartphone based plans shall speak with the Arduino are written in the Dealing out dialect.

Data Collection 

For testing as well as information gathering resolutions, the gadget plus microcontroller panel shall be fastened to a portable workstation phone a link to guarantee predictable information exchange and maintain a strategic distance from the conceivable intricacies of remote correspondence (Tong L., Song Q., Ge Y., Liu M,2012). On the Arduino end, the microcontroller must be modified to peruse information from the dual sensors as well as guide that information via serial port to the smartphone interface. Perusing on or after the spinner shall be finished utilizing the analog work, which receipts the energy at a simple information stick and changes over it within a 0 to 1023 for 10-bit number. Perusing from the computerized accelerometer shall somewhat more included; it s shall be mandatory utilizing I2C correspondence conventions to write to the gadget deliver plus getting the registers to place it in the correct method before you could read from the gadget. I2C convention shall be utilized for imparting among the advanced accelerometer as well as microcontroller rather than having I2C more straightforward to actualize and the good twigs on the microcontroller panel which are available. Altogether, there shall be three or more readings from the accelerometer as well as two more readings from the bi-hub gyrator (gx and gy). When the information from the two sensors shall be legitimately perused in, then there shall be change over to cords, linked, at that point directed to the Bluetooth device that shall conclude the sequential association. A deferral of 100ms shall be embedded to keeping in mind the end goal to not stop up the port. This implies we seized readings from the sensors 10 times each second, which demonstrated adequate. Each arrangement of fall detection of elderly reading shall be finished with a newline character to recognize them on the PC side. The complete Arduino information gathering system can be found the dashboard provided. Since the Arduino program end shall be running, the system side needed to compile a package to get the information. This sequencer shall open a serial correspondence port at that point perused in the information. The information cords shall need to be concatenated to separate the significant information esteems once again into whole number frame (Kohei Arai, 2013).

All the fall calculations which were formed considerately ought to be reformed concluded from a through workflow configuration to Arduino software development, which preserves consecutively in a constantly spinning design. Because of the constrained memory of the microcontroller, data interpretations cannot be persistently missing in memory along with erstwhile data past a precise grade cannot be gotten into. For assurance, the sequencer distinguishes the ‘previous state’ of the calculation or this occurrence which limits historical circle implementation, a “trigger” outline was applied. In a manner of speaking, the point which limits the fragmented a Boolean “start” variable shall be set valid; in the subsequent circle implementation this shall prompt additional high-quality enunciations as well as conceivable trigger enactment/deactivation till the point that prompts are established valid in addition to fall is recognized or every starts are not initiated (Medrano C., Igual R., Plaza I., Castro M., Fardoun H.M,2015).

First Algorithm

The main drop recognition calculation is founded off the idea that amongst a fall, a elderly person encounters a flashing free drop or lessening in fast-moving up, straggled by an extensive point in increasing speed, at that point an adjustment in introduction. The work flow for first algorithm is seen underneath in diagram beneath (Carlo Tacconi, Sabato Mellone, and Lorenzo Chiari, 2011). We perceive the calculation verifies whether the increasing speed extent interruptions a usual lower boundary. If this lower limit is wrecked, the design at that point confirms whether AM discontinuities a established upper edge within 10s. On the off chance when the higher edge is wrecked, the computation at that point confirms whether the individual’s overview has altered in a usual range inside 10s, which shall establish a man has fallen or toppled over. In the event that, the individual’s introduction has changed, the calculation at that point looks at to checked whether introduction breaks after 10s, which shall show the specific is powerless in their fallen position on the pulverized. On the off chance that this relics for this solution shall be continuous, the calculation perceives this as a fall. A dissatisfaction of each of the halfway optimal situations shall rearrange the triggers as well as send you back to the begin (Juan a. Botia, Ana Villa, and Jose Palma, 2012). The excellence of this calculation is that it needs an achievement to breakdown two AM restrictions as well as have an introduction change. In a perfect world this extra lower edge would decrease the quantity of false positives. The shortcoming of this calculation is that it requires the tumble to include an introduction change. If there is certain elderly falls, similar to those dropping down near the toilet seat in the bathroom, will not inspire the essential introduction alteration (Juan Cheng, Student Member, Xiang Chen, and Minfen Shen, 2012). 

Second Algorithm 

The second fall location calculation is centered around recognizing and dispensing with tedious exercises like strolling and successively from drops. The work flow for second algorithm is seen in the beneath diagram. We understand the calculation confirms whether an consequence has chanced. On the off chance that a consequence occurred, the calculation stores the introduction 10s preceding the primary affect (Maged N Kamel Boulos, Steve Wheeler, Carlos Tavares, and Ray Jones, 2012). The calculation at that point checks if another effect happens inside 20s as well as preserves on spinning via till the chain of possessions halts. The calculation at that point forms whether this restraint of effects surpassed 20s, which shall demonstrate a dreary action corresponding to falling. In the event that, the chain of effects is under 10s, the calculation stores the introduction 1s afterwards the last effect and thinks about it to the introduction 1s preceding the primary effect. If an adjustment in introduction has occurred, the individual shall be expected to be tumbled over amongst the unsettling influence and a fall is identified especially in the case of elderly. The quality of this calculation shall allow to perceive the progressive effects that shall be normal for typical exercises such as strolling; it shall likewise perceive plus distinguish muddled falls which shall include a battle or falter in the bathroom or toilet. The shortcoming of this calculation is that it needs the tumble to include an introduction variation (Salgado P., Afonso P.,2015). 

Third Algorithm

This is a recognition calculation, which is founded off the possibility that amongst a drop, a elderly man or woman encounters an widespread point in increasing speed, at that point a timeframe wherever they might battle to recover levelheadedness. The workflow for this third Algorithm shall be seen below. Although the calculation verifies whether the increasing speed size breaks a set upper edge (Kangas M., Vikman I., Wiklander J., Lindgren P., Nyberg L., Jämsä T,2009). If this higher limit is wrecked, the calculation at that point sits close-fitting up to 10s for the fall detection to originated back to a generally ordinary level. If the fall detection in the sensors comes back to an ordinary level, this would demonstrate the individual has possibly quit battling and is immobilized. The calculation at that point confirms whether the person’s outline has altered in a established range, which shall show elderly man is unmoving in a fallen situation on the bathroom floor. On the off chance this remnants are endless, the calculation perceives this as a drop. A dissatisfaction of the central choice situations shall reorganize the triggers as well as refer you back to the beginning. The quality of this calculation is that it considers post-fall battle plus come back to typical speeding up to affirm the client is powerless on a level plane. The shortcoming of this calculation is that it requires the tumble to include an introduction change (Wang J., Zhang Z., Li B., Lee S., Sherratt R.S.,2014). 

After Fall Process  

Subsequent to the location of a drop of elderly with help of any of the calculations, the drop discovery plans at that point experience what is known as a upright fall prepare. The workflow for the procedure can be found in the diagram shown beneath (Kangas M., Vikman I., Wiklander J., Lindgren P., Nyberg L., Jämsä T,2009). This procedure shall explain, every falls identified by a calculation, this shall also consider possible drops for elderly people’. The user is then cautioned along with that it shall give 10 seconds window to drop the alarm in the event that it was a false location or they could recover their self-control. On the off chance that the alarm is most certainly not crossed out inside 10s, the sequencer shall go into affirmed fall method where the user along with its crisis associates are alarmed of an affirmed drop occurrence (Pang C. H., Lee J. V., Chuah Y. D., Tan Y. C. and Debnach N.,2013). In the case of this Arduino sequencer which completely executes the toilet sensors, the third algorithm, catches, alarm inside sensors, and remote Bluetooth correspondence through a tablet. The toilet sensor dealing out program that completely executes remote Bluetooth information exchange through the gadget, information stockpiling to content record, onscreen show messages synced with gadget modes, and a continuous increasing speed size plot.

Result

The target of this toiler sensor for this elderly fall detection system in the bathrooms is to connect remotely the smart tablet based Bluetooth enabled smart phone. By the completion of this venture, we had accomplished our essential objective of making a working model ready to perceive both hazardous stance and tumbles from non-falls, while remotely synced with a Bluetooth enabled tablet (Tong L., Song Q., Ge Y., Liu M,2012). Taking a gander at the hidden recognition handle, our Fall Detection System enhances past frameworks and plans. We joined half breed fall recognition calculations gotten from prevailing calculations to locate the one with the most noteworthy affectability in addition to specificity. What’s more, we could utilize our broad test informational collection to configuration, prepare, and execute a basic program which shall be ready to analyze information from toilet sensors which shall decide whether a drop has happened or not in the premise that beat each of the customary calculations (Mubashir M., Shao L., Seed L,2013). The toilet sensor falls detection venture, in that there are some regions for the impending advancement. In addition business part of things, upgrades shall consists of  consuming pre-recorded speech strategies for the user, the expansion of a receiver thus allow the user to record their very own voice communications to be sent to emergency acquaintances amid crises, setting up crisis contacts however the server side by sending instant messages, Voice over via VOIP techniques decreasing the whole size of the gadget utilizing convention developed circuit sheets along with Lithium particle batteries, thus porting the personnel cross software development onto smart phone for totally versatile correspondence. This thing shall bring hypothetical improvement in the IOT based solution proposed for toilet sensors project, furthermost is the requirement for extra trying of project current calculations; additionally, tuning of limit esteems and high-quality circumstances might create new, far superior calculations (Tong L., Song Q., Ge Y., Liu M,2012). What’s more, more guineas pigs from a more extensive scope of physical classes performing more sorts of fall or non-fall exercises would give a great deal more entire informational collection to effort from. This would altogether enhance the preparation of the this sensor for elderly fall detection, and with enough information one shall even actualize tallness/weight/sex particular query based  tables/calculations to tailormade the gadget to every client’s personnel requirements (Stevens J.A., Haas E.N., Haileyesus T,2011).

References

Shumway-Cook A., Ciol M.A., Hoffman J., Dudgeon B.J., Yorkston K., Chan L. Falls in the Medicare Population: Incidence, Associated Factors, and Impact on Health Care. Phys. Ther. 2009;89:324–332. doi: 10.2522/ptj.20070107.

Stevens J.A., Haas E.N., Haileyesus T., (2011) “Nonfatal bathroom injuries among persons aged ≥15 years–United States, 2008”, J Safety Res. Vol. 42, No. 4, pp 311-5.

 Mubashir M., Shao L., Seed L., (2013) “A survey on fall detection: Principles and approaches”, Neurocomputing, Vol. 100, pp 144-152.

 Joe J., Demiris G., (2013), Older adults and mobile phones for health: A review”, Journal of Biomedical Informatics, Vol. 46, No. 5, pp 947-954.

 Chuah, Y. D., Komiya, R., Goi, B. M., (2012) “Abnormal walking styles definition”, Applied Mechanics and Materials, Vol. 145, pp 374-378.

 Zhao Z., Chen Y., Wang S., Chen Z., (2012) “FallAlarm: Smart phone based fall detecting and positioning system”, Procedia Computer Science, Vol. 10, pp 617-624.

 Nyan M.N., Tay F.E.H., Murugasu E., (2008) “A wearable system for pre-impact fall detection”, Journal of Biomechanics, Vol. 41, No.16, pp 3475-3481.

 Lee J. V., Chuah Y. D, Chieng K. T. H., (2013) “Smart Elderly Home Monitoring System with an Android Phone”, International Journal of Smart Home, Vol. 7, No. 3, pp 17-32.

 Lee J. V., Chuah Y. D., Chieng K. T. H., (2013) “Development of an Elderly Telecare System”, International Journal of Control and Automation, Vol. 6, No. 3, pp 57-68.

 Lee J. V., Chuah Y. D., Chai C. T.,, (2013) “A Multilevel Home Security System (MHSS)”, International Journal of Smart Home, Vol. 7, No. 2, pp 49-60.

 Pang C. H., Lee J. V., Chuah Y. D., Tan Y. C. and Debnach N., (2013) “Design of a Microcontroller based Fan Motor Controller for Smart Home Environment”, International Journal of Smart Home, Vol. 7, No. 4, pp 233-246.

 Tan K. L., Hing K.C.T., Chai C. T., Cheun A.H.L., Tan R. W., Chuah Y. D., (2012) “Development of an automated multiple power sources switching and monitoring system”, IEEE Conference on Sustainable Utilization and Development in Engineering and Technology (STUDENT), pp 271-276.

Lord S.R., Sherrington C., Menz H.B., Close J.C. Falls in Older People: Risk Factors and Strategies for Prevention. Cambridge University Press; Cambridge, UK: 2007. p. 470.

He J., Hu C., Wang X.Y. A Smart Device Enabled System for Autonomous Fall Detection and Alert. Int. J. Distrib. Sens. Netw. 2016;12:1–10. doi: 10.1155/2016/2308183.

Koshmak G., Loutfi A., Linden M. Challenges and Issues in Multisensor Fusion Approach for Fall Detection: Review Paper. J. Sens. 2016;2016:1–12. doi: 10.1155/2016/6931789.

Yu M., Rhuma A., Naqvi S.M., Wang L., Chambers J. A posture recognition-based fall detection system for monitoring an elderly person in a smart home environment. IEEE Trans. Inf. Technol. Biomed. 2012;16:1274–1286.

Yazar A., Erden F., Cetin A.E. Multi-sensor ambient assisted living system for fall detection; Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing; Florence, Italy. 4–9 May 2014; pp. 1–3.

Becker C., Schwickert L., Mellone S., Bagalà F., Chiari L., Helbostad J.L., Zijlstra W., Aminian K., Bourke A., Todd C., et al. Zeitschrift Für Gerontologie Und Geriatrie. Springer International Publishing; Berlin, Germany: 2012. Proposal for a multiphase fall model based on real-world fall recordings with body-fixed sensors; pp. 707–715.

Bai Y.-W., Wu S.-C., Tsai C.-L. Design and implementation of a fall monitor system by using a 3-axis accelerometer in a smart phone. IEEE Trans. Consum. Electron. 2012;58:1269–1275. doi: 10.1109/TCE.2012.6414995.

Salgado P., Afonso P. Lecture Notes in Electrical Engineering. Springer; Berlin, Germany: 2015. Fall Detection with Kalman Filter and SVM; pp. 407–416.

Medrano C., Igual R., Plaza I., Castro M., Fardoun H.M. Personalizable Smartphone Application for Detecting Falls; Proceedings of the IEEE-EMBS International Conference on Biomedical and Health Informatics; Valencia, Spain. 1–4 June 2014; pp. 169–172.

El-Sheimy N., Hou H., Niu X. Analysis and modeling of inertial sensors using Allan variance. IEEE Trans. Instrum. Meas. 2008;57:140–149. doi: 10.1109/TIM.2007.908635.

Pannurat N., Thiemjarus S., Nantajeewarawat E. Automatic Fall Monitoring: A review. Sensors. 2014;14:12900–12936. doi: 10.3390/s140712900.

Bourke A.K., Van De Ven P., Gamble M., O’Connor R., Murphy K., Bogan E., McQuade E., Finucane P., O´Laighin G., Nelson J. Evaluation of waist-mounted tri-axial accelerometer based fall-detection algorithms during scripted and continuous unscripted activities. J. Biomech. 2010;43:3051–3057. doi: 10.1016/j.jbiomech.2010.07.005.

Lindemann U., Hock A., Stuber M., Keck W., Becker C. Evaluation of a fall detector based on accelerometers: A pilot study. Med. Biol. Eng. Comput. 2005;43:548–551. doi: 10.1007/BF02351026.

Wang J., Zhang Z., Li B., Lee S., Sherratt R.S. An Enhanced Fall Detection System for Elderly Person Monitoring using Consumer Home Networks. IEEE Trans. Consum. Electron. 2014;60:23–28. doi: 10.1109/TCE.2014.6780921.

Li Q., Stankovic J.A., Hanson M.A., Barth A.T., Lach J., Zhou G. Body Sensor Networks. International Workshop on Wearable & Implantable Body Sensor Networks; Berkeley, CA, USA: 2009. Accurate, fast fall detection using gyroscopes and accelerometer derived posture information; pp. 138–143.

Gjoreski H., Kozina S., Gams M., Lustrek M. RAReFall—Real-Time Activity Recognition and Fall Detection System; Proceedings of the IEEE International Conference on Pervasive Computing and Communications Workshops; Budapest, Hungary. 24–28 March 2014; pp. 145–147.

Özdemir A.T., Barshan B. Detecting falls with wearable sensors using machine learning techniques. Sensors. 2014;14:10691–10708. doi: 10.3390/s140610691.

Ojetola O., Gaura E.I., Brusey J. Fall Detection with Wearable Sensors–Safe (Smart Fall Detection); Proceedings of the 7th International Conference on Intelligent Environments (IE); Nottingham, UK. 25–28 July 2011; pp. 318–321.

Zhang T., Wang J., Xu L., Liu P. Fall detection by wearable sensor and one-class SVM algorithm. Lect. Notes Control Inf. Sci. 2006;345:858–886.

Tong L., Song Q., Ge Y., Liu M. HMM-Based Human Fall Detection and Prediction Method Using Tri-Axial Accelerometer. IEEE Sens. J. 2013;13:1249–1256.

Dinh C., Struck M. A new real-time fall detection approach using fuzzy logic and a neural network; Proceedings of the International Workshop on Wearable Micro & Nano Technologies for Personalized Health; Oslo, Norway. 24–26 June 2009; pp. 57–60.

Schwickert L., Becker C., Lindemann U., Maréchal C., Bourke A., Chiari L., Helbostad J.L., Zijlstra W., Aminian K., et al. Zeitschrift für Gerontologie Und Geriatrie. Springer; Berlin, Germany: 2013. Fall detection with body-worn sensors: A systematic review; pp. 706–719.

Ligorio G., Sabatini A.M. A Novel Kalman Filter for Human Motion Tracking with an Inertial-Based Dynamic Inclinometer. IEEE Trans. Biomed. Eng. 2015;62:2033–2043. doi: 10.1109/TBME.2015.2411431.

Kangas M., Vikman I., Wiklander J., Lindgren P., Nyberg L., Jämsä T. Sensitivity and specificity of fall detection in people aged 40 years and over. Gait Posture. 2009;29:571–574. doi: 10.1016/j.gaitpost.2008.12.008.

Maryak J.L., Spall J.C., Heydon B.D. Use of the Kalman Filter for Inference in State-Space Models with Unknown Noise Distributions. IEEE Trans. Autom. Control. 2004;49:87–90. doi: 10.1109/TAC.2003.821415.

Brockwell P.J., Dahlhaus R., Trindade A.A. Modified Burg Algorithms for Multivariate Subset Autoregression. Stat. Sin. 2005;15:197–213.

Friedman N., Geiger D., Goldszmidt M. Bayesian network classifiers. Mach. Learn. 1997;29:131–163. doi: 10.1023/A:1007465528199.

Friedman N., Linial M., Nachman I., Pe’er D. Using Bayesian Networks to Analyze Expression Data. J. Comput. Biol. 2000;7:601–620. doi: 10.1089/106652700750050961. Klenk J., Becker C., Lieken F., Nicolai S., Maetzler W., Alt W., Zijlstra W., Hausdorff J.M., Van Lummel R.C., Chiari L., et al. Comparison of acceleration signals of simulated and real-world backward falls. Med. Eng. Phys. 2011;33:368–373. doi: 10.1016/j.medengphy.2010.11.003.

Bagala F., Becker C., Cappello A., Chiari L., Aminian K., Hausdorff J.M., Zijlstra W., Klenk J. Evaluation of accelerometer-based fall detection algorithms on real-world falls. PLoS ONE. 2012;7 doi: 10.1371/journal.pone.0037062.

Bourke A.K., Klenk J., Schwickert L., Aminian K., Ihlen EA., Helbostad J.L., Chiari L., Becker C. Temporal and kinematic variables for real-world falls harvested from lumbar sensors in the elderly population; Proceedings of the 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society; Milan, Italy. 25–29 August 2015; pp. 5183–5186.

Vavoulas G., Pediaditis M., Spanakis E.G., Tsiknakis M. The MobiFall dataset: An initial evaluation of fall detection algorithms using smartphones; Proceedings of the IEEE 13th International Conference on Bioinformatics and Bioengineering; Chania, Greece. 10–13 November 2013; pp. 1–4.

Department of Economic Population Division and United Nations Social Affairs. World Population Ageing. 2009.

 Gunnar Akner. Nutrition och fysisk funktion/fysisk aktivitet hosaldre personer. Technical report, ¨ Orebro Universitet, April 2009.

 Maged N Kamel Boulos, Steve Wheeler, Carlos Tavares, and Ray Jones. How smartphones are changing the face of mobile and participatory healthcare: an overview, with example from eCAALYX. Biomedical engineering online, 10(1):24, January 2011.

 Gregory Koshmak. An android based monitoring and alarm system for patients with chronic obtrusive disease. Master thesis, Orebro University, 2011.

Juan Cheng, Student Member, Xiang Chen, and Minfen Shen. A Framework for Daily Activity Monitoring and Fall Detection Based on Surface Electromyography and Accelerometer Signals. 2012

IEEE Journal of Biomedical and Health Informatics, pages 38 –45, December 2012.

 Marcin Bajorek and Jedrzej Nowak. The role of a mobile device in a home monitoring healthcare system. Proceedings of the Federated Conference on Computer Science and Information Systems, pages 371–374, 2011.

 Stefan Dernbach, Barnan Das, Narayanan C. Krishnan, Brian L.Thomas, and Diane J. Cook. Simple and Complex Activity Recognition through Smart Phones. 2012 Eighth International Conference on Intelligent Environments, pages 214–221, June 2012.

 Peihusan Tsai Jane W.S. Liu John K. Zao, Mei-Ying Wang. Smart phone based medicine in-take scheduler, reminder and monitor. 2010 12th IEEE International Conference on e-Health Networking Applications and Services (Healthcom), pages 162–168, June 2010.

Dominik Kobylarz and Jacek Danda. A Common Interface for Bluetooth-based Health Monitoring Devices. 2013 29th Southern Biomedical Engineering Conference, pages 153–154, May 2013.

 Kohei Arai. Wearable physical and psychological health monitoring system. Science and Information Conference (SAI), 2013, pages 133–138, 2013.

 Matthias Wagner and Benjamin Kuch. Android based Body Area Network for the evaluation of medical parameters. 10th International Workshop on Intelligent Solutions in Embedded Systems, pages 33–38, 2012.

 Zhang Aihua and Hu Linghao. The system of pulse monitoring based on Windows Mobile. Business Management and Electronic Information (BMEI), 2011.

 M Kozlovszky and L Bartalis. Personal health monitoring with Android based mobile devices. Information & Communication Technology Electronics & Microelectronics (MIPRO), pages 326–330, 2013.

Noury and A. Fleury. Fall detection-principles and methods. Proceedings of the 29th Annual International Conference of the IEEE EMBS, pages 1663–1666, January 2007.

Damien Brulin and Estelle Courtial. Multi-sensors data fusion system for fall detection. Proceedings of the 10th IEEE International Conference on Information Technology and Applications in Biomedicine, pages 1–4, November 2010.

 A Leone and G Diraco. A multi-sensor approach for people fall detection in home environment. Workshop on Multi-camera and Multi-modal Sensor Fusion Algorithms and Applications, pages 1– 2, 2008.

 Charalampos N. Doukas and Ilias Maglogiannis. Emergency fall incidents detection in assisted living environments utilizing motion, sound, and visual perceptual components. IEEE Transactions on Information Technology in Biomedicine, 15(2):277–289, 2011.

 Mitja Lu?strek, Hristijan Gjoreski, Simon Kozina, Bo?zidara Cvetkovi¨u, Violeta Mirchevska, and Matja?z Gams. Detecting Falls with Location Sensors and Accelerometers. Proceedings of the Twenty-Third Innovative Applications of Artificial Intelligence Conference, (2006):1662–1667, 2010.

Bingbing Ni, Nguyen Chi Dat, and Pierre Moulin3. RGBD-Camera based get-up event detection for hospital fall prevention. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1405–1408, 2012.

 Bruno Neves Ferreira, Vania Guimaraes, and Hugo Sereno Ferreira. Smartphone based fall prevention exercises. 2013 IEEE 15th International Conference on e-Health Networking, Applications and Services (Healthcom 2013), (Healthcom):643–647, October 2013.

Naruhiro Shiozawa and Shin Arai. Gait analysis of sit-to-walk motion by using portable acceleration monitor device for fall prevention. Proceedings of the IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI 2012), 25(Bhi):563–564, 2012.

 Carlo Tacconi, Sabato Mellone, and Lorenzo Chiari. Smartphone based applications for investigating falls and mobility. 2011 5^th International Conference on Pervasive Computing Technologies for Healthcare (PervasiveHealth) and Workshops, pages 258–261, 2011.

 Stevan Marinkovic and Riccardo Puppo. Implementation and testing of a secure fall detection system for Body Area Networks. Proc. 27th International Conference on Microelectronics (MIEL 2010), pages 16–19, 2010.

 Patrick Crilly and Vallipuram Muthukkumarasamy. Using smart phones and body sensors to deliver pervasive mobile personal healthcare. 2010 Sixth International Conference on Intelligent Sensors Sensor Networks and Information Processing, pages 291–296, 2010.

 Hamed Ketabdar and Tim Polzehl. Fall and emergency detection with mobile phones. In Proceeding of the eleventh international ACM SIGACCESS conference on Computers and accessibility ASSETS 09, page 241. ACM SIGACCESS, ACM Press, 2009.

 Bruce Moulton Mitchell Yuwono, Steven W. Su. Fall Detection using a Gaussian Distribution of Clustered Knowledge , Augmented Radial Basis. Proceedings of the 6th International Conference on Broadband Communications Biomedical Applications, pages 145–

150, November 2011.

 Christoph Dinh and Matthias Struck. A new real-time fall detection approach using fuzzy logic and a neural network. 6th International Workshop on Wearable Micro and Nano Technologies for Personalized Health (pHealth), pages 57–60, June 2009.

 Mark V Albert, Konrad Kording, Megan Herrmann, and Arun Jayaraman. Fall classification by machine learning using mobile phones. PLoS ONE, 7(5), 2012.

K. Bourke, P. van de Ven, M. Gamble, R. O’Connor, K. Murphy, E. Bogan, E. McQuade, P. Finucane, G. Olaighin, and J. Nelson. Evaluation of waist-mounted tri-axial accelerometer based fall-detection algorithms during scripted and continuous unscripted

activities. Journal of biomechanics, 43(15):3051–3057, November 2010.

 Hristijan Gjoreski, Mitja Lu?strek, and Matja?z Gams. Context-Based Fall Detection Using Inertial and Location Sensors. Ambient Intelligence, 2012.

 Qiang Li and John A Stankovic. Grammar-based, posture- and context-cognitive detection for falls with different activity levels.Proceedings of the 2nd Conference on Wireless Health WH ’11, 2011.

 Alessandro Leone, Giovanni Diraco, and Pietro Siciliano. Context-Aware AAL Services through a 3D Sensor-Based Platform. Journal of Sensors, 2013:1–10, 2013.

 J.W.Zheng, T.H.Wu, and Y.Zhang. A Wearable Mobihealth Care System support in Real-time Diagnosis and alarm. Med Bio Eng Comput, pages 877–885, September 2007.

 Topi Pulkkinen, Young-Sung Son, Joohyung Lee, Yann-Hang Lee, Mikko Sallinen, and Jun-Hee Park. Progressive monitoring and treatment planning of diabetes mellitus in smart home environment. 2013 IEEE International Conference on Consumer Electronics (ICCE), pages 206–207, January 2013.

 Juan a. Botia, Ana Villa, and Jose Palma. Ambient Assisted Living system for in-home monitoring of healthy independent elders. Expert Systems with Applications, 39(9):8136–8148, July 2012.

 JV Lee, YD Chuah, and KTH Chieng. Smart elderly home monitoring system with an android phone. Int. J. Smart Home, 7(3):17–32,2013.

 Gabriel Villarrubia, Juan F De Paz, Javier Bajo, and Juan M Corchado. Ambient agents: embedded agents for remote control and monitoring using the PANGEA platform. Sensors (Basel, Switzerland), 14(8):13955–79, January 2014.

 C Silvia, A. Cesta, L. Coraci, and L. Karlsson. GiraffPlus: Combining social interaction and long term monitoring for promoting independent living. 2013 6th International Conference on Human System Interactions (HSI), pages 578–585, June 2013.

 Lars Karsson, Lia Silva-lopez, and Jonas Ullberg. D 3 . 1 Context Inference and Configuration Planning Prototypes. Technical Report January 2012, 2014.

 Federico Pecora, Marcello Cirillo, Francesca Dell Osa, Jonas Ullberg, and Alessandro Saffiotti. A constraint-based approach for proactive, context-aware human support. Journal of Ambient Intelligence and Smart Environments, 4:347–367, 2012