CN114366466B - A walking care robot that integrates health information monitoring and prediction - Google Patents

Abstract

The invention proposes a walking care robot that integrates health information monitoring and prediction, and belongs to the technical field of medical care equipment. This equipment includes a wheelchair body, which is characterized by: including a detection unit, the detection unit includes a display module, a detection box and a control box; the display module is used to control and display monitoring information; the detection box includes an integrated cuff type Blood pressure monitoring device, heart rate detection module, and blood oxygen detection module; the control box includes a control unit and a data processing unit, which can realize storage and data analysis of motion control and detection information, establish health files, and upload them through a mobile phone. This invention can effectively solve the transportation problem of the elderly disabled population and enhance life satisfaction. At the same time, it stores daily monitored heart rate, blood pressure, blood oxygen and other data, and relies on big data interpolation algorithm analysis to establish the user's personal health information file, which can Help doctors correctly grasp patients' daily physical conditions.

Description

A walking care robot that integrates health information monitoring and prediction

Technical field

The invention belongs to the technical field of medical care equipment, and in particular relates to a walking care robot that integrates health information monitoring and prediction.

Background technique

With the development of economy and society, the problem of population aging is becoming increasingly serious globally. Taking my country as an example, the elderly population will increase to 264 million in 2020, accounting for 18.7% of the total population. In addition, the incidence and disability rates of central cerebrovascular diseases, orthopedic diseases, etc. among the elderly are increasing, which puts great pressure on the health and care care of the disabled elderly. According to Burgess Consulting data, nearly 24% of residents in my country do not know about rehabilitation care, and only 26% of residents have a correct understanding of rehabilitation care. The data shows that my country’s per capita consumption of rehabilitation care in 2017 was US$5.5, which was much lower than the same year. With per capita rehabilitation medical consumption in the United States at US$54, my country has great potential to tap on the demand side. Therefore, it is necessary to design a rehabilitation nursing robot that integrates heart rate, blood pressure, blood oxygen and other health information monitoring, stores and analyzes the data at the same time, forms a health file, and can be used as a home transportation device to reduce the pressure on caregivers.

The health monitoring companion robots currently on the market mainly focus on human-computer interaction. They mainly focus on the elderly who are unable to receive timely treatment when they suffer from physical illness or are in danger due to lack of companionship. Therefore, they mainly use acceleration sensors to monitor whether the elderly have fallen, etc. , the human-computer interaction function is limited by current technical limitations, and the effect is poor in speech recognition and actual question and answer experience. Health monitoring mainly relies on the smart bracelet worn, failing to take into account that the shape, size and wearing requirements of the bracelet are limited, resulting in detection accuracy. Restricted. At present, the detection of heart rate by smart bracelets mostly adopts the principle of PPG photoplethysmography, which measures the pulse in an optical way. After years of testing and experimentation, it has become more mature. However, the detection of blood pressure by the bracelet cannot be accurate. According to " JJG 692-2010 Calibration Regulations for Non-Invasive Automatic Measurement of Blood Pressure Monitors" stipulates that if the error exceeds 4mmHG, it will be judged as unqualified. Therefore, if the measured blood pressure does not match the actual blood pressure, it is likely to affect the amount of medication used by patients with hypertension, thereby affecting their health. Secondly, the existing health monitoring companion robots mainly focus on the early warning function, fail to store and analyze the monitored data, and miss the opportunity to form the user's health profile. Therefore, it is necessary to improve the design to address these problems.

Contents of the invention

In response to the above technical problems, the present invention proposes a walking care robot that integrates health information monitoring and prediction. On the one hand, this invention can effectively solve the problem of the elderly population being disabled, unable to walk independently, and relying too much on nursing staff; on the other hand, On the one hand, it can also solve the problem of late detection of cardiovascular and cerebrovascular diseases due to their reliance on health examinations, and the inability of doctors to obtain patients’ daily health information.

In order to achieve the above object, the technical solution adopted by the present invention is: a walking care robot that integrates health information monitoring and prediction, including a wheelchair main body, which is characterized in that it includes a detection unit, and the detection unit includes a display module, a detection box and a control unit. box; the display module is used to control and display monitoring information; the detection box is located under the armrest of the wheelchair body, and includes an integrated cuff-type blood pressure monitoring device, a heart rate detection module, and a blood oxygen detection module; the blood pressure monitoring box The device uses an electronic sphygmomanometer; the heart rate detection module detects heart rate through a receiver installed on the cuff; blood oxygen monitoring uses a reflective blood oxygen sensor chip to calculate blood oxygen saturation through the difference in light intensity between emission and reception. degree; the control box is located under the seat of the wheelchair body and includes a control unit and a data processing unit. The control unit integrates Ardnino and Bluetooth modules and can be paired with a smartphone to download and operate the APP through the mobile phone to implement a mobile platform-based The mobility control of the mobility care robot, the APP control interface realizes forward, backward, left turn, right turn and motor speed adjustment functions, while setting prompts for health monitoring; the data processing unit stores and data analyzes the received information, and passes The control unit uploads data to the mobile phone and establishes a health file; the data analysis uses a multi-weighted interpolation algorithm to ensure the smoothness of the user's multi-parameter health record curve, and predicts the user's health status, specifically as follows:

(1) After measuring data for more than one week, preprocess multiple sets of data: perform an arithmetic mean of a certain quantity:

According to Bessel's formula, the standard deviation of the measurement is:

in: is the residual error;

(2) Determine whether there are gross errors in the processed data information based on the 3σ criterion, eliminate the data with gross errors, and retain the remaining data;

(3) Use interpolation algorithms for data;

After the multiple sets of measured data are cleaned and filtered, based on Markov theory, data densification is performed between the two measurement data. The overall number of densification is twice, and the value point of the first densification is Bn, which satisfies the following formula:

The value points of the second densification are Cn and Dn, where Cn is located on the equally spaced middle side of An and Bn, and Dn is located on the equally spaced middle side of Bn and An+1. After two densifications, the overall data volume is the original 3 times, the calculation formula is as follows:

Substituting Bn to represent the function we get:

C _n =KA _n +(i*k+j)A _n-1 +i ² A _n-2 +i*jA _n-3

In the same way, the calculation formula of Dn is:

Substituting Bn to represent the function we get:

D _n =(K ² +i)A _n +k(i+j)A _n-1 +j(k+i)A _n-2 +j ² A _n-3

Considering the smoothness of the fitting curve, quadratic term interpolation is used to fit the function:

On the basis of quadratic function fitting to enhance data reliability, the prediction function is implemented to provide early warning of health diseases when the predicted value deviates significantly from the normal value.

Further, the walking care robot includes a movement unit, which includes a drive motor, a control rocker and a battery pack; the control rocker has multiple gears to select and adjust the movement speed, which is convenient for the elderly and disabled people to use; The battery pack powers the robot.

Furthermore, the control rocker adopts a 24V universal rocker intelligent controller with EABS slope retention function. The human-machine interface has LED prompts. The driving mode is PWM duty cycle adjustment and electronic differential, with a controller. Alarm sound. The battery pack uses a lithium battery pack with a nominal capacity of 40Ah and a standard voltage of DC24V, ensuring a maximum range of about 6 kilometers on a single charge; the charger input voltage is 220V, 50HZ, the output voltage is 29.2V, and the output current is 5A.

Furthermore, the control box adopts STM32F103RCT6 series control chip.

Further, the display module includes a measurement information display screen, indicator lights and operation buttons for measurement control, information display and alarm prompts.

The present invention also provides a method for integrated health information detection and prediction. Based on the above-described integrated health information monitoring and prediction walking care robot, it is characterized by including the following steps:

(1) Use the mobility care robot to monitor blood pressure, heart rate and blood oxygen:

Blood pressure detection: Using an electronic sphygmomanometer, the user takes out the cuff from the detection box, connects the air nozzle hose, and performs blood pressure detection operations through the buttons on the display module;

Heart rate detection: Green light is emitted through the photoelectric receiver installed on the cuff; after the light beam is emitted, the skin, muscle tissue and blood will absorb part of it, and the rest will be reflected back to the receiver. When the heart contracts and relaxes, it will Reflected light of different colors is produced, and the received electrical signal also changes with the pulse; this signal is demodulated through an algorithm and the heart rate is calculated;

Blood oxygen monitoring: A reflective blood oxygen sensor chip is used. There are two light-emitting diodes in the chip, which emit red light with a wavelength of 669nm and infrared light of 880nm respectively to the wrist. The photodiode on the other side is used to receive the reflected light. , calculate the saturation of blood oxygen through the difference in light intensity between emission and reception;

(2) The mobility robot is used to store monitoring data. All detected information is transmitted to the control box, and the data processing unit stores the data. The information includes the type of detected data, including blood pressure, heart rate and blood oxygen; followed by The time information of the detection is specific to the day; the last is the size of the data value, and it is judged whether it is within the normal range, specifically the blood pressure value: systolic blood pressure 90-140mmHg, diastolic blood pressure 60-90mmHg, heart rate 60-100 beats/minute, When the blood oxygen concentration exceeds 90%, it will be automatically prompted by the display module when it exceeds the set limit;

(3) Use the mobility robot to process the stored data to ensure the smoothness of the user's multi-parameter health record curve and predict the user's health;

(3.1) After measuring data for more than one week, preprocess multiple sets of data: perform an arithmetic mean of a certain quantity:

According to Bessel's formula, the standard deviation of the measurement is:

in: is the residual error;

(3.2) Based on the 3σ criterion, determine whether there are gross errors in the processed data information, eliminate the data with gross errors, and retain the remaining data;

(3.3) Use interpolation algorithms for data;

After the multiple sets of measured data are cleaned and filtered, based on Markov theory, data densification is performed between the two measurement data. The overall number of densification is twice, and the value point of the first densification is Bn, which satisfies the following formula:

The value points of the second densification are Cn and Dn, where Cn is located on the equally spaced middle side of An and Bn, and Dn is located on the equally spaced middle side of Bn and An+1. After two densifications, the overall data volume is the original 3 times, the calculation formula is as follows:

Substituting Bn to represent the function we get:

C _n =KA _n +(i*k+j)A _n-1 +i ² A _n-2 +i*jA _n-3

In the same way, the calculation formula of Dn is:

Substituting Bn to represent the function we get:

D _n =(K ² +i)A _n +k(i+j)A _n-1 +j(k+i)A _n-2 +j ² A _n-3

Considering the smoothness of the fitting curve, quadratic term interpolation is used to fit the function:

On the basis of quadratic function fitting to enhance data reliability, the prediction function is implemented to provide early warning of health diseases when the predicted value deviates significantly from the normal value;

(4) Save all results and upload them to the mobile phone through the control unit to establish a health file.

Furthermore, in order to obtain more accurate interpolation results, the three interpolation coefficients are preferably set to: k=0.5, i=0.3, j=0.2.

Description of drawings

Figure 1 is a schematic structural diagram of a walking care robot of the present invention.

Figure 2 is a schematic diagram of the right control rocker of the present invention.

Figure 3 is a schematic diagram of the left detection box of the present invention.

Figure 4 is a schematic diagram of the bottom control unit of the present invention.

Figure 5 is a schematic diagram of the primary densification of the present invention.

Figure 6 is a schematic diagram of multi-weighted interpolation and prediction according to the present invention.

Among them: 1 headrest, 2 display module, 3 detection box, 4 control box, 5 battery pack, 6 rubber rear wheels, 7 drive motor, 8 seat unit, 9 pedals, 10 omnidirectional wheels, 11 leg pads, 12 seats Pads, 13 control rockers, 14 soft armrests, 15 seat back pads.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings and examples.

This design is a walking care robot that integrates health information monitoring and prediction (as shown in Figure 1): the overall structure includes three parts: the seat body, the movement unit and the detection unit. The seat body meets the actual riding needs, including the head Pillow 1, rear rubber wheel 6, seat frame unit 8, pedal 9, front omnidirectional wheel 10, leg pad 11, seat cushion 12, soft armrest 14 and chair back pad 15. The motion unit is composed of a battery pack 5, a drive motor 7 and a control rocker 13. Among them, the front wheel omnidirectional wheel 10 has a small turning radius and low lateral movement damping, and is suitable for indoor use. The control rocker is convenient for the elderly and disabled people, and the moving speed can be Adjustable, it can be adjusted to 1st gear indoors and to 5th gear for outdoor movement to speed up travel. The motion system can effectively replace the elderly disabled population's dependence on caregivers for mobility and reduce the pressure on caregivers. The detection unit consists of a display module 2, a detection box 3 and a control box 4. The detection box 3 can be easily opened and integrates a cuff-type blood pressure detection device and heart rate and blood oxygen monitoring. The display module 2 has a measurement information display screen, indicator lights and operations. button.

The materials selected in this design should meet halogen-free requirements and meet ROHS certification standards. The main structure of the seat frame unit is welded and formed from 6061 aluminum profiles, which has good cold working performance, high corrosion resistance and toughness, and is convenient for subsequent color coating. The front wheel arm, front beam, plug, screw mounting block, bearing sleeve, etc. are all made of 6061 aluminum alloy material; the rotating front wheel shaft is made of chrome-plated smooth round, which has high hardness and is wear-resistant and corrosion-resistant; the pedal mounting plate, spacer sleeve, and bearing The retaining ring, shaft end retaining ring, and pedal push seat are all made of stainless steel 304; the main seat plate, bottom support, and rear wheel support welding group are all made of Q235-A carbon steel, which has good overall performance; the pedal 9 is made of plastic. Simple and durable; the leg pads 11, seat pad 12, chair back pad 15 and soft armrests 14 are custom-made with leather exterior to ensure comfort.

The right control joystick 13 adopts a 24V universal joystick intelligent controller with EABS slope retention function. The human-machine interface has LED prompts. The driving method is PWM duty cycle adjustment and electronic differential, and has a controller alarm sound. .

On the left is the detection box 3. The detection box 3 can be opened easily. It contains a cuff-type blood pressure detection device and a heart rate and blood oxygen monitoring module. The blood pressure detection uses an electronic sphygmomanometer with high accuracy used in hospitals. The user puts the cuff After taking it out from the detection box 3, connect the air nozzle hose and perform blood pressure detection operation through the buttons on the display module 2. The test results will also be displayed on the display screen. The principle is to indirectly measure blood pressure with the oscillometric method. During the process of deflating or inflating the cuff, the pulse wave information is sensed, and the blood pressure data is obtained through a series of complex conversions and calculations. The heart rate detection adopts the principle of PPG photoplethysmography, which emits green light through the photoelectric receiver installed on the cuff. After the beam is emitted, the skin, muscle tissue and blood absorb part of it, and the rest is reflected back to the receiver. Among them, the reflection of light by structures such as human bones, skin, and fat is a fixed value, while capillaries, arteries and veins constantly change with the beating of the heart, so the reflection value keeps fluctuating, and the blood tends to reflect red light and absorb it. Green light, so the heart will produce reflected light of different colors when it contracts and relaxes, which is reflected at the receiving end in that the received electrical signal also changes with the pulse. When the heart contracts, the peripheral blood volume is at its maximum and light absorption is at its maximum, and the detected light intensity is minimum. During diastole, on the contrary, the detected light intensity is the largest, causing the light intensity received by the light receiver to change pulsatingly. Through algorithms, we can demodulate this signal, and then use a certain algorithm to calculate the heart rate. Blood oxygen monitoring uses a reflective blood oxygen sensor chip. There are two light-emitting diodes in the chip, which emit red light with a wavelength of 669nm and infrared light of 880nm respectively to the wrist. The photodiode on the other side is used to receive the reflected light. Light, the blood oxygen saturation can be calculated through the difference in intensity of emitted and received light. All detected information will be transmitted to the storage unit in control box 3, waiting for subsequent processing.

Under the seat are the battery pack 5 and the control box unit 4. The battery pack 5 uses a lithium battery pack with a nominal capacity of 40Ah and a standard voltage of DC24V, which can ensure a maximum range of about 6 kilometers on a single charge. The input voltage of the charger is 220V, 50HZ, and the output voltage is 29.2V. The output current is 5A.

The control chip in the control box 4 adopts the STM32F103RCT6 series, which has high performance, low cost and low power consumption. Its main functions are to control the forward, backward and left and right rotation of the walking care robot and control the opening and closing of the detection module. The data of each detection can also be summarized in this module to facilitate subsequent calls. The control box 4 also includes a data processing unit. The data processing unit first stores data. The information includes the type of detected data, such as blood pressure, heart rate and blood oxygen. The second is the detection time information, which can be specific to the day. The last thing is the size of the data value. When it exceeds the normal set range, the system will automatically alarm and prompt, specifically the blood pressure value: systolic blood pressure 90-140mmHg, diastolic blood pressure 60-90mmHg. Heart rate is 60-100 beats/minute. Blood oxygen concentration >90%. When the set time limit is exceeded, the machine will automatically prompt and the user can consider re-measurement or repeat the measurement at another time to ensure accurate results.

In the data processing stage, multiple sets of data are processed after measuring the data for more than one week. A series of equal-precision measurements of a certain quantity are performed. Due to the existence of random errors, the measured values are different. The arithmetic mean of all measured values should be used as the final measurement result. Assume l ₁ , l ₂ ,..., l _n are the values obtained by n measurements, then the arithmetic mean

The arithmetic mean is closest to the true value. According to the law of large numbers in probability theory, if the number of measurements increases infinitely, the arithmetic mean It must approach the true value L ₀ . Among them/> is called the residual error. Considering that the measurement is an equal-precision measurement, the standard deviation of the measurement can be obtained according to Bessel's formula:

Converts to:

According to the 3σ criterion, it is judged whether there are gross errors in the processed data information, and the data that may have gross errors are eliminated, and the remaining data are retained.

Based on the different influences of the weight coefficients of multiple measurement values, after the multiple sets of measured data are cleaned and filtered, based on the Markov theory, the user's recent three measurement data will have a greater impact on the subsequent results, so the two measurements are Data densification is performed in the middle of the data. The overall number of densification times is two, which can increase the data volume to 3 times. The schematic diagram of the first densification is shown in Figure 5. The densification value point is Bn, which satisfies the following formula:

The value points of secondary densification are Cn and Dn. Among them, the position of Cn is the middle side of the equal distance between An and Bn, and the position of Dn is the middle side of the equal distance between Bn and An+1. After two densifications, the overall data volume is 3 times the original. The calculation formula is as follows:

Substituting Bn to represent the function we get:

C _n =KA _n +(i*k+j)A _n-1 +i ² A _n-2 +i*jA _n-3

In the same way, the calculation formula of Dn can be obtained as:

Substituting Bn to represent the function we get:

D _n =(K ² +i)A _n +k(i+j)A _n-1 +j(k+i)A _n-2 +j ² A _n-3

Considering the smoothness of the fitting curve, using quadratic term interpolation and fitting the function with the above nodes, we can get:

As shown in Figure 6, on the basis of quadratic function fitting to enhance data reliability, the prediction function can be realized, and early warning of health diseases can be achieved when the predicted value deviates significantly from the normal value.

Taking into account the different weight coefficients of multiple measurements of health information, after multiple experimental tests, in order to obtain more accurate interpolation results, the three coefficients were set to: k=0.5, i=0.3, j=0.2.

It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solution of the present invention can be carried out. Modifications or equivalent substitutions without departing from the spirit and scope of the technical solution of the present invention shall be included in the scope of the claims of the present invention.

Claims (8)

1. A mobility care robot that integrates health information monitoring and prediction, including a wheelchair main body, which is characterized in that: it includes a detection unit, and the detection unit includes a display module (2), a detection box (3) and a control box (4); The display module (2) is used to control and display monitoring information; The detection box (3) is located under the armrest of the wheelchair body, and includes an integrated cuff-type blood pressure monitoring device, a heart rate detection module, and a blood oxygen detection module; the blood pressure monitoring device uses an electronic sphygmomanometer; and the heart rate detection module The heart rate is detected through the photoelectric sensor installed on the cuff; the blood oxygen monitoring uses a reflective blood oxygen sensor chip to calculate the saturation of blood oxygen through the difference in light intensity between emission and reception; The control box (4) is located under the seat of the wheelchair main body and includes a control unit and a data processing unit. The control unit integrates Ardnino and Bluetooth modules and can be paired with a smartphone to download and operate the APP through the mobile phone to realize operation based on the mobile phone. The platform's nursing robot mobile control, APP control interface realizes forward, backward, left turn, right turn and motor speed adjustment functions, while setting prompts for health monitoring; the data processing unit stores and data analyzes the received information, and passes The control unit uploads data to the mobile phone and establishes a health file; the data analysis uses a multi-weighted interpolation algorithm to ensure the smoothness of the user's multi-parameter health record curve, and predicts the user's health status, specifically as follows: (1) After measuring data for more than one week, preprocess multiple sets of data: perform an arithmetic mean of a certain quantity: According to Bessel's formula, the standard deviation of the measurement is: in: is the residual error; (2) Determine whether there are gross errors in the processed data information based on the 3σ criterion, eliminate the data with gross errors, and retain the remaining data; (3) Use interpolation algorithms for data; After the multiple sets of measured data are cleaned and filtered, based on Markov theory, data densification is performed between the two measurement data. The overall number of densification is twice, and the value point of the first densification is Bn, which satisfies the following formula: The value points of the second densification are Cn and Dn, where Cn is located on the equally spaced middle side of An and Bn, and Dn is located on the equally spaced middle side of Bn and An+1. After two densifications, the overall data volume is the original 3 times, the calculation formula is as follows: Substituting Bn to represent the function we get: C _n =KA _n +(i*k+j)A _n-1 +i ² A _n-2 +i*jA _n-3 In the same way, the calculation formula of Dn is: Substituting Bn to represent the function we get: D _n =(K ² +i)A _n +k(i+j)A _n-1 +j(k+i)A _n-2 +j ² A _n-3 Considering the smoothness of the fitting curve, quadratic term interpolation is used to fit the function: On the basis of quadratic function fitting to enhance data reliability, the prediction function is implemented to provide early warning of health diseases when the predicted value deviates significantly from the normal value. 2. The walking care robot integrating health information monitoring and prediction according to claim 1, characterized in that the walking care robot includes a movement unit, and the movement unit includes a drive motor (7) and a control rocker (13) and a battery pack (5); the control rocker (13) has multiple gears to select and adjust the moving speed, making it convenient for elderly and disabled people to use; the battery pack (5) supplies power to the robot. 3. The mobility care robot integrating health information monitoring and prediction according to claim 2, characterized in that the control rocker (13) adopts a 24V universal rocker intelligent controller with an EABS slope-holding function. The machine interface has LED prompts, the driving method is PWM duty cycle adjustment and electronic differential, and has a controller alarm sound. 4. The mobility care robot integrating health information monitoring and prediction according to claim 2, characterized in that the battery pack (5) adopts a lithium battery pack with a nominal capacity of 40Ah and a standard voltage of DC24V to ensure the maximum range on a single charge. About 6 kilometers; the input voltage of the charger is 220V, 50HZ, the output voltage is 29.2V, and the output current is 5A. 5. The mobility care robot integrating health information monitoring and prediction according to claim 1, characterized in that the control box (4) adopts an STM32F103RCT6 series control chip. 6. The mobility care robot integrating health information monitoring and prediction according to claim 1, characterized in that the display module (2) includes a measurement information display screen, indicator lights and operation buttons for measurement control and information display. and alarm prompts. 7. A method of integrated health information detection and prediction, based on the integrated health information monitoring and prediction walking care robot as claimed in any one of claims 1 to 6, characterized by comprising the following steps: (1) Use the mobility care robot to monitor blood pressure, heart rate and blood oxygen: Blood pressure detection: Using an electronic sphygmomanometer, the user takes out the cuff from the detection box (3), connects the air nozzle hose, and performs blood pressure detection operations through the buttons on the display module (2); Heart rate detection: Green light is emitted through the photoelectric receiver installed on the cuff; after the light beam is emitted, the skin, muscle tissue and blood will absorb part of it, and the rest will be reflected back to the receiver. When the heart contracts and relaxes, it will Reflected light of different colors is produced, and the received electrical signal also changes with the pulse; this signal is demodulated through an algorithm and the heart rate is calculated; Blood oxygen monitoring: A reflective blood oxygen sensor chip is used. There are two light-emitting diodes in the chip, which emit red light with a wavelength of 669nm and infrared light of 880nm respectively to the wrist. The photodiode on the other side is used to receive the reflected light. , calculate the saturation of blood oxygen through the difference in light intensity between emission and reception; (2) The walking care robot is used to store monitoring data, and all detected information is transmitted to the control box (4). The data processing unit stores the data, and the information includes the type of detected data, Including blood pressure, heart rate and blood oxygen; secondly, the time information of the detection, specific to the day; finally, the size of the data value, and judging whether it is within the normal range, specifically the blood pressure value: systolic blood pressure 90-140mmHg, diastolic blood pressure 60-90mmHg , the heart rate is 60-100 beats/minute, and the blood oxygen concentration is >90%. When the set limit is exceeded, the display module (2) automatically prompts; (3) Use the mobility care robot to process the stored data, use a multi-weighted interpolation algorithm to ensure the smoothness of the user's multi-parameter health record curve, and predict the user's health condition; (3.1) After measuring data for more than one week, preprocess multiple sets of data: perform an arithmetic mean of a certain quantity: According to Bessel's formula, the standard deviation of the measurement is: in: is the residual error; (3.2) Based on the 3σ criterion, determine whether there are gross errors in the processed data information, eliminate the data with gross errors, and retain the remaining data; (3.3) Use interpolation algorithms for data; After the multiple sets of measured data are cleaned and filtered, based on Markov theory, data densification is performed between the two measurement data. The overall number of densification is twice, and the value point of the first densification is Bn, which satisfies the following formula: The value points of the second densification are Cn and Dn, where Cn is located on the equally spaced middle side of An and Bn, and Dn is located on the equally spaced middle side of Bn and An+1. After two densifications, the overall data volume is the original 3 times, the calculation formula is as follows: Substituting Bn to represent the function we get: C _n =KA _n +(i*k+j)A _n-1 +i ² A _n-2 +i*jA _n-3 In the same way, the calculation formula of Dn is: Substituting Bn to represent the function we get: D _n =(K ² +i)A _n +k(i+j)A _n-1 +j(k+i)A _n-2 +j ² A _n-3 Considering the smoothness of the fitting curve, quadratic term interpolation is used to fit the function: On the basis of quadratic function fitting to enhance data reliability, the prediction function is implemented to provide early warning of health diseases when the predicted value deviates significantly from the normal value; (4) Save all results and upload them to the mobile phone through the control unit to establish a health file. 8. A method for integrating health information detection and prediction according to claim 7, characterized in that, in order to obtain a more accurate interpolation result that conforms to the rules of health information, the three interpolation coefficients are preferably set to: k=0.5, i =0.3, j=0.2.