KR20120108335A - Emergency monitoring system based on newly developed fall detection algorithm - Google Patents

Abstract

PURPOSE: An emergency monitoring system is provided to accurately deal with emergency situations by improving falling sensing performance. CONSTITUTION: An acceleration measuring part(110) measures acceleration of 3-shafts. An angular speed measuring part(120) measures angular velocity of 2-shafts. A heart beat measuring part(130) measures heart rate of a user. A falling determination part(151) determines the falling of the user using acceleration data of the 3-shafts and the angular velocity of the 2-shafts. An emergency state determination part(152) determines emergency situations using the change of the acceleration data of the 3-shafts and the angular velocity of the 2-shafts after falling. A user response checking part(153) determines whether real emergency situations are generated or not by checking a response of the user. A heart rate abnormal determination part(154) compartmentalizes the kind of emergency situations using the heart rate of the user. A communication part(160) transmits the heart rate and an emergency signal to outside. [Reference numerals] (110) Acceleration measuring part; (120) Angular speed measuring part; (130) Heart beat measuring part; (140) Data storing part; (151) Falling determination part; (152) Emergency state determination part; (153) User response checking part; (154) Heart rate abnormal state determination part; (160) Communication part; (200) Gateway; (AA) Data

Description

EMERGENCY MONITORING SYSTEM BASED ON NEWLY DEVELOPED FALL DETECTION ALGORITHM}

The present invention relates to an emergency monitoring system, and in particular, by applying an improved algorithm compared to the conventional method for detecting falls on a sensor node using a three-axis accelerometer and a two-axis gyroscope, the user's heart rate and falls measured by the sensor node The present invention relates to an emergency monitoring system using a fall detection algorithm, which collects detection parameters to quickly determine a fall and an emergency.

Recently, the change in the social structure of the elderly population has increased the demand for ubiquitous healthcare along with a new lifestyle.

Falls are a very important problem in the health of older people and are particularly common among older people 65 years or older due to poor physical function.

17.2% of the 65-year-olds in Korea experienced a fall more than once a year, and 23.7% of the elderly experienced a fall more than three times. Falls are a problem for older people because older people are less likely to recover from physical damage and dysfunction.

Falls can lead to death if you are unconscious or unable to move quickly because of a fall. In addition, elderly people who have experienced a fall are in a state of psychological anxiety due to fear of falling again.

Therefore, the fall is a very important problem for the elderly, it is recognized as a social problem and measures are required accordingly.

In addition, if a message is sent to a guardian or a health center by automatically detecting an emergency situation in which the person is unconscious or unable to move after the fall, the time for medical intervention may be shortened in case of an emergency from the fall.

An object of the embodiment of the present invention for improving the above-mentioned problem is superior to the emergency monitoring system using the fall detection algorithm by applying the optimal algorithm for the fall detection to the sensor node using the three-axis accelerometer and two-axis gyroscope It is to provide an emergency monitoring system using the fall detection algorithm to show the fall detection performance.

Another object of the embodiment of the present invention for improving the above-described problem is to determine the emergency situation through the measurement of the signal and heart rate of the fall detection sensor node fall detection algorithm to determine more accurately than the emergency monitoring system using only the fall detection algorithm It is to provide an emergency monitoring system.

Another object of the present invention for improving the above-described problem is that the gateway receiving a signal from the sensor node of the user when the falling situation occurs by transmitting the location information of the user identified by the signal and GPS to the health care center or guardian It is to provide an emergency monitoring system using a fall detection algorithm to easily determine the current location of the user.

Emergency monitoring system using the fall detection algorithm according to an embodiment of the present invention for achieving the above object is an acceleration measuring unit for measuring the acceleration of three axes, an angular velocity measuring unit for measuring the angular velocity of two axes, heart rate for measuring the heart rate of the user A falling-measurement unit for determining a fall of the user by using a measurement unit, the acceleration data of the three axes and the angular velocity of the two axes, and if the fall determination unit determines that the fall, the acceleration data of the three axes and the angular velocity of the two axes after the fall In the emergency determination unit for determining an emergency situation by using a change, if it is determined that the emergency situation in the emergency determination unit, the user response check unit for determining whether the user's response is a real emergency, the user response check unit If it is determined that the actual emergency situation, the user's heart rate is used to Heart rate abnormality determination unit that separates the flow, the screen signals the heart rate and the emergency situation includes a sensor node with a communication unit for transmitting to the outside.

The fall determination unit, when the AVsvm below exceeds the first preset threshold value and the ACCsvm value below the preset second threshold value within a predetermined number of samples, examines the mAngle value below and presets the predetermined value. If the predetermined third threshold value is exceeded within a range of samples, it is determined that a fall occurs.

Here, i denotes the i th sample, X acc , Y acc , and Z acc denote acceleration data of each axis, and X av , Y av denote angular velocity data of each axis.

In addition, mAngle means the angle of the sensor node.

After the fall judgment is determined, the emergency state determination unit maintains a state in which mACCpp and mAVpp are smaller than a predetermined fourth threshold value and a fifth threshold value, respectively, and the mAngle is greater than the third threshold value for a predetermined time or more within a predetermined time period. If it is an emergency.

In addition, after the user response check unit is determined to be an emergency, it is preferable to check the response of the user to determine that the user does not respond to the actual emergency, and to determine that the emergency is not an emergency.

In addition, the heart rate abnormality determiner is determined as an emergency, and after checking the user's heart rate is classified as a first emergency within a predetermined normal range, if not within the normal range is classified as a very emergency second emergency. More preferably.

In addition, the emergency monitoring system using the fall detection algorithm receives the heart rate and the emergency signal from the sensor node, the GPS data calculated using the embedded GPS to the predetermined health care center using wireless communication The transmission gateway may further include.

The emergency monitoring system using the fall detection algorithm according to an embodiment of the present invention is applied to the emergency monitoring system using the existing fall detection algorithm by applying the optimal algorithm for the fall detection to the sensor node using the 3-axis accelerometer and the 2-axis gyroscope. Compared with the fall detection performance, it is possible to accurately deal with emergency situations.

Emergency monitoring system using the fall detection algorithm according to an embodiment of the present invention to determine the emergency situation through the measurement of the signal and heart rate of the fall detection sensor node to make more accurate judgment than the emergency monitoring system using only the fall detection algorithm. There is an effect that can be more specific response to.

In the emergency monitoring system using the fall detection algorithm according to an embodiment of the present invention, the gateway receiving the signal from the sensor node of the user transmits the location information of the user identified by the signal and the GPS to the health care center or the guardian together with the fall situation. When it occurs, it is possible to easily grasp the current location of the user to have a quick response to emergencies.

1 is an illustration of an emergency monitoring system according to an embodiment of the present invention.
2 is an exemplary view of a database for acceleration and angular velocity measurement and analysis according to an embodiment of the present invention.
3 is a block diagram of an emergency monitoring system according to an embodiment of the present invention.
4 is a flowchart illustrating an algorithm operation of an emergency monitoring system according to an exemplary embodiment of the present invention.

The present invention as described above will be described in detail with reference to the accompanying drawings and embodiments.

1 is an illustration of an emergency monitoring system using a fall detection algorithm according to an embodiment of the present invention.

The emergency monitoring system using the fall detection algorithm includes a sensor node 100, a gateway 200, and a health care center 300 as shown.

The sensor node 100 may measure an electrocardiogram, and an accelerometer and a gyroscope may be connected to measure acceleration and angular velocity of the user 1.

The sensor node 100 uses an R-R interval after extracting an electrocardiogram from three electrodes attached to a user's left chest for measuring heart rate.

In this case, a 16-bit microcontroller for applying an instrumentation amplifier and a digital filter may be used to extract the ECG.

In general, large falls and changes in angular velocity occur in the fall, unlike normal behavior.

Therefore, the sensor node 100 is provided with an accelerometer and an angular speedometer to detect a fall, thereby measuring the direction and intensity of the movement of the user 1 in real time.

In this case, the accelerometer is preferably a three-axis accelerometer, the angular rate is preferably a two-axis gyroscope. The sensor node 100 collects three-axis acceleration and two-axis angular velocity data from the three-axis accelerometer and two-axis gyroscope to detect the fall of the user 1.

The accelerometer and gyroscope are very small in size and can be easily attached to a wireless module, making them suitable for falling detection. Therefore, the fall can be detected by measuring the direction and the intensity of the movement of the user 1 through this.

On the other hand, a predetermined reference value is required to distinguish the movement and the fall in a normal situation through the acceleration and the angular velocity values measured using the accelerometer and the gyroscope.

FIG. 2 is an exemplary diagram of a database configuration for measuring and analyzing acceleration and angular velocity of the accelerometer and gyroscope, before the user measures the acceleration and angular velocity for offset error correction of the accelerometer 110 and gyroscope 120. In the standing state, 100 samples are taken and the average of these values is set as the reference value.

In this case, the acceleration and the angular velocity of the user received from the sensor node 100 are monitored in real time through the LabVIEW 410 (LabVIEW) in the personal computer 400 (PC) as shown in the drawing and store data through the database 420.

Referring back to FIG. 1, the sensor node 100 checks the state of the user 1 through data input from the sensor, and transmits a heart rate and an emergency signal to the gateway 200 when an emergency occurs. .

To this end, the sensor node 100 preferably includes a communication unit for short-range wireless communication with the gateway 200. For example, the sensor node 100 may include a wireless communication module such as Zigbee.

The gateway 200 may be a mobile phone or an embedded board possessed by a user. The gateway 200 transmits the heart rate and the emergency signal received from the sensor node 100 to the health care center 300 or the guardian (3), and stores data on the ECG and the fall before and after the emergency. .

In addition, the gateway 200 may include a global positioning system (GPS) to transmit user location information (3).

The health care center 300 may receive an emergency signal of the user 1 from the gateway 200 and check the heart rate to take necessary measures 4.

In addition, the health care center 300 may track 4 the location of the user 1 through the location information of the GPS.

As such, the location of the user 1 whose gateway 200 receives the signal 2 from the sensor node 100 of the user 1 is identified by the signal and GPS to the health care center 300 or the guardian. By transmitting the information together (3) it is possible to easily grasp the current position of the user (1) when the fall situation occurs, there is an advantage that can quickly respond to an emergency situation.

Conventionally, various methods of identifying an emergency situation by measuring an acceleration value and an angular velocity value using an accelerometer or an accelerometer and a gyroscope have been proposed.At present, studies on the fall detection using an accelerometer and a gyroscope are actively conducted. .

The present invention proposes an emergency monitoring system using a three-axis accelerometer and a two-axis gyroscope to apply a fall detection algorithm far superior to the conventional emergency monitoring system. In addition, we propose an emergency monitoring system to more specifically determine the emergency situation by detecting heart rate and falling.

A detailed description of the emergency monitoring system and the emergency situation determination algorithm using the fall detection algorithm will be described with reference to FIGS. 3 to 4.

3 is a block diagram of an emergency monitoring system according to an embodiment of the present invention, Figure 4 is a flow chart of the operation of the emergency monitoring system according to an embodiment of the present invention.

Referring to FIG. 3, the emergency monitoring system using the fall detection algorithm includes an acceleration measuring unit 110 measuring acceleration of three axes, an angular velocity measuring unit 120 measuring an angular velocity of two axes, and a heart rate measurement of a user's heart rate. The unit 130, a fall determination unit 151 for determining the fall of the user by using the acceleration data of the three axes and the angular velocity of the two axes, if the fall determination unit 151 determines that the fall, the fall of the three axes after When it is determined that the emergency situation is determined by the emergency determination unit 152 and the emergency determination unit 152 using the acceleration data and the change in the angular velocity of the two axes, the user's response is checked to determine the actual emergency situation. If it is determined that the user response check unit 153, the user response check unit 153 is the actual emergency situation, using the heart rate of the user of the emergency situation Heart rate abnormality determination section 154 that separates the acids, the screen signals the heart rate and the emergency situation includes a sensor node 100 equipped with a communication unit 160 for transmitting to the outside.

In this case, the emergency monitoring system using the fall detection algorithm receives the heart rate and the emergency signal from the sensor node 100, and uses the wireless communication to calculate the GPS data calculated using the built-in GPS. The gateway 200 may further include a transmission to the care center.

In addition, the sensor node 100 includes a fall determination unit 151, an emergency situation determination unit 152, a user response confirmation unit 153, and a heart rate abnormality determination unit 154. It may further include a data storage unit 140 for storing data necessary for each detection and determination.

The sensor node 100 uses five parameters to detect a fall and an emergency of the user.

The fall detection parameters are ACC svm , AV svm and mAngle, and the emergency detection parameters are mACC pp , mAV pp and mAngle.

The ACC svm and AV svm are SVM (Sum Vector Magnitude) of acceleration and angular velocity, respectively, as follows.

Where i is the i-th data, X acc , Y acc , and Z acc are the acceleration data of each axis, and X av and Y av are the angular velocity data of each axis.

  mAngle represents the angle of the sensor node.

The angle uses the acceleration of three axes measured from the acceleration measuring unit, and is as follows.

The angle measured by Equation 3 is an error in the angle measurement when the acceleration changes due to the user's movement. Therefore, in order to reduce the error of the angle measurement, the average of the dog samples of the measured angle is used and is represented by Equation 4. In this embodiment, n is 70 pieces.

MACC pp and mAV pp are parameters for detecting a user's movement after a fall occurs.

At this time, if the change of the acceleration or the angular velocity of the user after the fall is very small, it is determined as an emergency.

The mACC pp and mAV pp represent the peak-peak for n samples of acceleration and angular velocity, respectively, and the mean for m samples of this value.

The following shows the peak-peak of acceleration and angular velocity for n samples, respectively.

Here, ACC max (i) , ACC min (i) , AV max (i) , and AV min (i) are as follows.

Here, Max represents the maximum value of n samples, and Min represents the lowest value of n samples. n is 50 pieces.

Equations 5 and 6 can be used to obtain the peak-peak for n samples, and the average for m samples is as follows.

The fall determination unit 151 if the AV svm exceeds the preset first threshold value, and after the ACC svm value exceeds the preset second threshold value within a predetermined number of samples, the mAngle value thereafter. If it is determined that the predetermined third threshold value is exceeded within a predetermined range of samples, it is determined that a fall occurs.

In addition, the emergency state determination unit 152 may determine that the mACC pp and the mAV pp are smaller than a preset fourth threshold value and a fifth threshold value, respectively, and the mAngle is the third threshold value within a preset time after the fall is determined. If a larger state persists for more than a predetermined time, it is considered an emergency.

After the user response check unit 153 is determined to be an emergency, it is preferable to check the response of the user to determine that the user does not respond to the actual emergency, and to determine that the emergency is not an emergency.

In addition, the heart rate abnormality determiner 154 determines that the user's heart rate after being determined to be an actual emergency, and classifies it as a first emergency if it is within a preset normal range, and if it is not the normal range, a very urgent system. 2 Can be classified as an emergency.

Alternatively, although not shown, in another embodiment of the present invention, the emergency monitoring system using the fall detection algorithm may include an acceleration measuring unit measuring acceleration of three axes, an angular velocity measuring unit measuring two angular velocities, and a user's heart rate. A sensor node having a heart rate measuring unit, a communication unit for transmitting the acceleration data, the angular velocity data and the heart rate to the outside, and a data receiving unit for receiving the acceleration data, the angular velocity data and the heart rate from the sensor node, the acceleration data of the three axes, and the second. A fall judgment unit that determines the fall of the user using the angular velocity of the axis, if the fall determination unit determines that the fall, the emergency situation to determine the emergency by using the acceleration data of the three axes and the change of the angular velocity of the two axes after the fall Judgment unit, the emergency judgment unit sold as an emergency If the user response check unit to determine whether the actual emergency situation by checking the response of the user, if the user response check unit is determined to be an actual emergency situation, heart rate abnormality determination using the heart rate of the user to distinguish the type of emergency situation The gateway may include a gateway having a transmitter for transmitting the GPS data calculated by using the signal for the emergency situation and the built-in GPS to a predetermined healthcare center using wireless communication.

As such, the emergency monitoring system using the fall detection algorithm is superior to the emergency monitoring system using the fall detection algorithm by applying the optimal algorithm for detecting the fall to the sensor node using the 3-axis accelerometer and the 2-axis gyroscope. Sensitive performance allows accurate response to emergencies.

4 is a flowchart illustrating an algorithm operation of an emergency monitoring system using a fall detection algorithm according to an embodiment of the present invention, wherein the emergency monitoring system collects data (S410), falls detection (S420), emergency situation detection (S430), The user response check (S440) and heart rate abnormality detection (S450) is performed.

Referring to FIG. 4, the emergency monitoring system determines whether the user has a fall through the fall detection step S420.

Fall detection algorithm example 1.if AVsvm> 170 ° / s
then
2.if ACCsvm> 2g
(among 50 samples after satisfying the condition in Line 1)
then
If mAngle> 60 °
(among 30-100 samples after satisfying the condition in Line 2)
then
4. return Fall Detection
5. return No Fall Detection

Table 1 is an example of a fall detection algorithm according to an embodiment of the present invention, wherein the emergency monitoring system first checks whether a threshold value of AV svm has been exceeded to detect a fall, and then ACC within 50 samples after the condition of line 1 is satisfied. Check that the svm value exceeds the threshold.

If the mAngle exceeds the threshold within 30 to 100 samples after the condition in line 2 is satisfied, it is determined that a fall has occurred to the user.

If it is determined that the fall in the above, the emergency situation detection step (S430) to determine whether the emergency situation actually occurred.

Emergency detection algorithm example 1.if Fall Detection
then
2.if mACCpp <0.5g and mAVpp <50 ° / s
and mAngle> 60 °
(continuously during 3 second)
then
3. return Emergency detection
4. return No Emergency detection

For example, as shown in Table 2, it is determined as an emergency when mACC pp and mAV pp are smaller than the threshold and mAngle is larger than the threshold for more than 3 seconds within 10 seconds after the fall is detected.

In addition, after the emergency situation is detected, the user's response is determined to determine whether the actual emergency situation (S440). That is, if the user does not respond after the emergency is detected, it is determined to be an actual emergency. If the response is made, it is not an emergency and is determined to be an incorrect detection.

User Response Algorithm Example 1.if emergency detection
then
2.if user's response
then
3. return No Emergency
4. return Emergency

Afterwards, if it is confirmed as an emergency, heart rate is measured, and the emergency is classified to transmit a signal to the gateway (S450).

For example, after an emergency is detected, the heart rate is checked to classify the emergency into A and B. In other words, if your heart rate is not between 60 and 100, which is your normal heart rate, classify it as an emergency A to alert you that you are in a very urgent state. Can be distinguished (S460, S470).

Example heart rate anomaly checking algorithm 1.if emergency
then
2.if Heart Rate <60 or Heart Rate> 100
then
3. return Emergency A
4. return Emergency B

As such, the emergency monitoring system using the fall detection algorithm according to an embodiment of the present invention can determine the emergency situation by measuring the signal and the heart rate of the fall detection sensor node, and more accurately determine the emergency monitoring system using only the fall detection algorithm. This allows more specific responses to emergencies.

Table 6 to Table 8 are the results of the experiment through the emergency monitoring system using the fall detection algorithm according to an embodiment of the present invention, the verification and analysis of the data was performed through the labview and database shown in FIG.

Each experimental item is shown in Table 5, and referring to FIG. 5, ADL (Activities of daily living) means daily actions, and Fall means fall actions. Here, ADL tested the behavior with acceleration and angular velocity of magnitude similar to the fall behavior.

In this experiment, specificity and sensitivity were measured according to each combination of parameters for falling detection. Tables 6 to 7 show the results of measuring specificity and sensitivity measured in daily and falling behaviors, respectively. Table 8 is an emergency situation detection rate of the emergency monitoring system according to an embodiment of the present invention.

Experiment item  behavior ADL-A
ADL-B
ADL-C
ADL-D
Fall-a
Fall-b
Fall-c jump
jump
Sitting in a chair and waking up
Lying in bed
Tripped forward
Unbalanced posture and falling to the side
Slip and fall back

Specificity (%) of each algorithm measured in daily life algorithm ADL-A (51) ADL-B (51) ADL-C (50) ADL-D (49) Total (201) Algorithm to be compared1 100 100 100 81.63 95.41 Algorithm to compare 2 0 0 88 42.86 32.71 Algorithm according to an embodiment of the present invention 100 100 100 95.92 98.98

Sensitivity (%) of each algorithm measured in the fall behavior algorithm Fall-A (51) Fall-B (50) Fall-C (51) Total (152) Algorithm to be compared1 100 100 100 100 Algorithm to compare 2 98.04 100 100 99.35 Algorithm according to an embodiment of the present invention 100 100 100 100

Emergency detection rate (%) Fall  Emergency detection Fall-A (51) 100 Fall-B (50) 100 Fall-C (51) 100 Total (152) 100

Referring to Tables 6-7, Algorithm 2 to be compared had a low specificity of 32.71% in everyday behavior. In particular, 0% specificity was measured in ADL-A and ADL-B. ADL-A and ADL-B cause acceleration and angular velocity of magnitude similar to falling, and Algorithm 2 does not accurately distinguish ADL from falling.

In addition, in comparison with Algorithm 1 to be compared, the fall detection algorithm according to the embodiment of the present invention showed better characteristics in specificity.

Referring to Table 8, the emergency detection rate of the emergency monitoring system according to an embodiment of the present invention was measured to 100% in each fall behavior.

Therefore, if a user loses consciousness or does not move after falling, 100% of an emergency may be automatically detected.

As described above, in the emergency monitoring system according to the embodiment of the present invention, 100% sensitivity and 98.98% specificity were measured by applying the fall detection algorithm, and 100% detection rate was measured in the emergency detection algorithm.

In addition, the emergency monitoring system according to an embodiment of the present invention applies an algorithm for detecting an emergency situation and classifying an emergency situation using a heart rate together with the above fall detection algorithm, thereby allowing the user to detect an emergency situation caused by a fall. It can detect and classify the user's condition using heart rate.

Thus, quick and accurate actions can be taken for emergencies resulting from falls.

The foregoing embodiments and advantages are merely exemplary and are not to be construed as limiting the present invention. However, the present invention is not limited to the above-described embodiments, and various changes and modifications may be made by those skilled in the art without departing from the scope of the present invention. .

1: User 100: Sensor Node
110: acceleration measurement unit 120: angular velocity measurement unit
130: heart rate measurement unit 140: data storage unit
150: judgment unit 151: fall judgment unit
152: emergency determination unit 153: user response confirmation unit
154: heart rate abnormality determination unit 160: communication unit
200: gateway 300: health care center

Claims (6)

An acceleration measuring unit measuring acceleration of three axes;
An angular velocity measuring unit measuring an angular velocity of two axes;
Heart rate measuring unit for measuring the heart rate of the user;
A fall determination unit configured to determine a fall of the user by using the acceleration data of the three axes and the angular velocity of the two axes;
An emergency determination unit determining the emergency by using the acceleration data of the three axes and the change of the angular velocity of the two axes after the fall;
A user response confirming unit determining whether the emergency is determined by checking the response of the user when it is determined that the emergency is determined by the emergency determining unit;
A heart rate abnormality determination unit that determines a type of emergency situation by using the user's heart rate when it is determined that the user response is determined to be an actual emergency situation;
Emergency monitoring system using a fall detection algorithm comprising a sensor node having a; communication unit for signaling the heart rate and the emergency situation to transmit to the outside.
The method of claim 1, wherein the fall determination unit
If the AVsvm below exceeds the first preset threshold and the following ACCsvm value exceeds the preset second threshold within a predetermined number of samples, the following mAngle value is then examined and within the preset predetermined range of samples. The emergency monitoring system using the fall detection algorithm, characterized in that the fall is determined if the third threshold value is exceeded.

Here, i denotes the i th sample, X acc , Y acc , and Z acc denote acceleration data of each axis, and X av , Y av denote angular velocity data of each axis.
In addition, mAngle means the angle of the sensor node.
The method of claim 2, wherein the emergency determination unit
After the fall is judged, if the following mACCpp and mAVpp are smaller than the fourth threshold and the fifth threshold, respectively, and the mAngle is greater than the third threshold for a predetermined time or more, it is determined as an emergency within a preset time. Emergency monitoring system using a fall detection algorithm, characterized in that.

The method of claim 1, wherein the user response confirmation unit
The emergency monitoring system using the fall detection algorithm, characterized in that after determining that the emergency, the user's response to determine that the user does not respond to the actual emergency, and if the response is not an emergency.
According to claim 1, wherein the heart rate abnormality determination unit
After it is determined as an emergency, if the heart rate of the user is checked and falls within a predetermined normal range, it is classified as a first emergency, and if it is not the normal range, fall detection is characterized in that it is classified as a very urgent second emergency. Emergency monitoring system using algorithm.
The method of claim 1,
The gateway receives the heart rate and the emergency signal from the sensor node, the gateway for transmitting the GPS data calculated using the embedded GPS to a predetermined healthcare center using wireless communication using a fall detection algorithm Emergency monitoring system.

Images