WO2022041225A1 - Server for use in assessing cardiovascular state, wearable device, and method for cardiovascular state assessment - Google Patents
Server for use in assessing cardiovascular state, wearable device, and method for cardiovascular state assessment Download PDFInfo
- Publication number
- WO2022041225A1 WO2022041225A1 PCT/CN2020/112638 CN2020112638W WO2022041225A1 WO 2022041225 A1 WO2022041225 A1 WO 2022041225A1 CN 2020112638 W CN2020112638 W CN 2020112638W WO 2022041225 A1 WO2022041225 A1 WO 2022041225A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- cardiovascular
- vital sign
- sign data
- user terminal
- user
- Prior art date
Links
- 230000002526 effect on cardiovascular system Effects 0.000 title claims abstract description 242
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000004891 communication Methods 0.000 claims abstract description 80
- 238000012545 processing Methods 0.000 claims description 32
- 238000012549 training Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 16
- 230000005540 biological transmission Effects 0.000 claims description 13
- 230000008859 change Effects 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 7
- 239000003550 marker Substances 0.000 claims description 3
- 239000004984 smart glass Substances 0.000 claims description 2
- 210000000707 wrist Anatomy 0.000 claims description 2
- 239000008280 blood Substances 0.000 description 7
- 210000004369 blood Anatomy 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000006866 deterioration Effects 0.000 description 5
- 230000036996 cardiovascular health Effects 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 230000036772 blood pressure Effects 0.000 description 3
- 230000036760 body temperature Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 210000004204 blood vessel Anatomy 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 150000002632 lipids Chemical class 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000029058 respiratory gaseous exchange Effects 0.000 description 2
- 208000024172 Cardiovascular disease Diseases 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 235000020940 control diet Nutrition 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000002542 deteriorative effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 230000000276 sedentary effect Effects 0.000 description 1
- 230000002792 vascular Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
Definitions
- the present application relates to health, medical technology, and more particularly, to servers, wearable devices, systems and methods for analyzing and evaluating a user's cardiovascular status.
- a server for assessing cardiovascular status includes a communication module and a processor.
- the communication module is configured to be able to communicate with different user terminals to receive vital sign data of the respective user associated with each user terminal, the vital sign data including at least cardiovascular parameters.
- the processor is configured to perform inputting of the vital sign data of the respective user associated with each user terminal into a cardiovascular prediction classifier, and the classifier may assign the corresponding user to the corresponding user based on at least the cardiovascular parameters in the vital sign data. Cardiovascular status is marked; and the marking is returned to the user terminal.
- the cardiovascular prediction classifier is trained and updated at predetermined time intervals with the corresponding user's vital sign data associated with each user terminal.
- a wearable device comprising a plurality of sensors configured to sense vital sign data of a wearer, the vital sign data including at least a cardiovascular parameter; a processor , which is configured to at least convert the sensed vital sign data into data suitable for transmission; and a communication module, which is configured to transmit the vital sign data processed by the processor.
- the communication module may include communication modules such as a Bluetooth module, a wireless communication module, and the like.
- a user terminal configured to communicate with a wearable device and a remote server, respectively.
- the user terminal is associated with the wearable device such that both are used for the same user.
- the user terminal is configured to receive vital sign data transmitted from the wearable device, where the vital signs include at least cardiovascular parameters; process the vital sign data to obtain changes in the cardiovascular parameters over time; process the sensed the measured vital sign data to obtain the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle; standardize the vital sign data, and perform one-hot coding on the standardized processing result to obtain the one-hot value of each parameter.
- the present application also provides a cardiovascular assessment system, which includes a wearable device, a user terminal, and a server.
- the wearable device then includes a plurality of sensors configured to sense vital sign data of the wearer, the vital sign data including at least cardiovascular parameters; a processor configured to at least convert the sensed vital sign data into data suitable for transmission; a communication module configured to transmit the processor-processed vital sign data.
- the user terminal includes a communication module configured to establish communication with the communication module of the wearable device and receive vital sign data sent by the wearable device; a processor configured to process the sensed vital sign data to obtain The difference between the cardiovascular parameters and the corresponding parameters of the previous cycle; standardize the vital sign data, and perform one-hot encoding on the standardized processing result to obtain the one-hot encoding of each parameter; according to the vital signs
- the specific distribution of data classifies each cardiovascular parameter to obtain classification information; and the communication module is configured to classify the change of the obtained cardiovascular parameter over time, the difference between the obtained cardiovascular parameter and the corresponding parameter of the previous cycle
- the value, the obtained one-hot encoding of each parameter and the obtained evaluation are sent to the server.
- the server includes a communication module for receiving vital sign data sent with the processor, the vital sign data including at least cardiovascular parameters; a processor configured to input the received vital sign data into the heart In the blood vessel prediction classifier, the classifier marks the cardiovascular state of the user based on at least the cardiovascular parameters in the vital sign data; and the communication module sends the mark to the user terminal; and the The user terminal is configured to emit the indicia visually and/or vocally.
- a method for evaluating a cardiovascular state performed on a server side comprising: receiving vital sign data of a corresponding user associated with each user terminal transmitted from a user terminal, the vital sign The data includes at least cardiovascular parameters; the vital sign data of the corresponding user associated with each user terminal is input into a cardiovascular prediction classifier, and the classifier is used to at least evaluate the cardiovascular parameters of the corresponding user based on the cardiovascular parameters in the physical sign data.
- the vessel status is marked; the marking is sent to the corresponding user terminal so that it can display the marking in a visual manner.
- a method for assessing cardiovascular status comprising: sensing cardiovascular parameters of a user through a wearable device worn by the user; the wearable device transmitting the sensed cardiovascular parameters of the user To the user terminal; the user terminal processes the sensed vital sign data to obtain the variation of the cardiovascular parameter with time and the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle, and normalizes the sensed vital sign data processing, and one-hot coding is performed on the standardized processing result to obtain the one-hot coding of each parameter, and each cardiovascular parameter is classified according to the specific distribution of the vital sign data to obtain classification information; the user terminal will obtain the obtained The change of cardiovascular parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot coding of each parameter, and the classification information are sent to the cloud server; the cloud server sends the received cardiovascular parameters.
- the changes of parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot coding of each parameter, and the classification information are input into the cardiovascular prediction classifier, which analyzes the cardiovascular state of the user. mark; the cloud server sends the mark to the user terminal; the user terminal shows the mark to the user.
- FIG. 1 is a schematic structural diagram of a remote server according to an example of the present application.
- FIG. 2 is a schematic structural diagram of a wearable device according to an example of the present application.
- FIG. 3 is a schematic structural diagram of a cardiovascular detection system according to an example of the present application.
- FIG. 4 is a flowchart of a method for evaluating a cardiovascular state performed on a server side according to an example of the present application.
- FIG. 5 is a flowchart of a method of assessing cardiovascular in accordance with an example of the present application.
- the examples given in this application are used to determine the cardiovascular state of the human subject according to the vital sign data of the human subject.
- vital sign data it mainly refers to including cardiovascular parameters related to heart rate, pulse, respiration, body temperature, exercise volume, blood sugar, blood pressure, blood sample, blood lipids, but does not exclude other Vital sign data to help determine cardiovascular status.
- user wearer, patient all refer to human subjects whose cardiovascular status is monitored using the protocols described in this application, referred to as user, wearer, patient according to the context.
- FIG. 1 is a schematic structural diagram of a server for assessing cardiovascular status according to an example of the present application, wherein the server 10 receives vital sign data of a human subject transmitted to it.
- the server 10 includes a communication module 100 , a processor 102 and a memory 104 .
- Server 10 is an illustrative and non-limiting example. Although shown as one server device, it can be multiple server devices.
- the processor 102 and the memory 104 may be distributed in multiple devices, and each device may be provided with the communication module 100 .
- the server 10 is exemplified as a cloud server.
- the communication module 100 is, for example, a wireless communication module, but does not exclude a wired communication module that communicates through a cable.
- the server 10 can establish connections with different user terminals and implement communication.
- the server 10 receives, through the communication module 100, the vital sign data of the corresponding user associated with each user terminal, where the vital sign data at least includes cardiovascular parameters.
- a cardiovascular prediction classifier 1020 is configured in the processor 102 . After the vital sign data of the corresponding user associated with each user terminal received through the communication module 100 is input to the processor 102 , the cardiovascular state of the corresponding user is marked by the cardiovascular prediction classifier 1020 .
- Labeling here means that the classifier gives an evaluation of the cardiovascular health status of the respective user in the form of a score based on the received data.
- the processor 102 is also configured to transmit the indicia via the communication module 100 for reception by a user terminal associated with the user represented by the indicia.
- the processor 102 is further configured to train the cardiovascular prediction classifier 1020 at predetermined time intervals with the vital sign data of the corresponding user associated with each user terminal, and use the trained classifier The original cardiovascular prediction classifier 1020 is updated.
- the memory 104 of the server 10 is used to store the data received by the communication module 100 and the instructions and data required by the processor 102 to perform various functions. To sum up, the memory 104 is used to store the instructions to be stored in the server 10 including the coordination processor 102 , the operation of the communication module 100 and other data to ensure the normal operation of the server 10 .
- the cloud server 10 receives data transmitted by a plurality of user terminals including the user terminal T1, the user terminal T2, . . . , the user terminal Tn through the communication module 100.
- each user terminal is associated with a user, that is, associated with a human object.
- User terminal T1 corresponds to user 1
- user terminal T2 corresponds to user 2
- user terminal Tn corresponds to n.
- the user terminal Ti (1 ⁇ i ⁇ n, n is a natural number) is associated with the user i, which means that the characteristic data of the communication between the user terminal Ti and the cloud server 10 is about the user i, not other users.
- the way of association can be realized by, for example, binding user ID information, identifying the user before collecting user data, etc., as long as the corresponding relationship between the user terminal Ti and the user i can be ensured.
- one user terminal is associated with one human object, the situation where one user terminal is associated with more than one user is not excluded, that is, one user terminal can be associated with multiple users. In this case, when transmitting data, the user terminal needs to identify the cardiovascular parameter of which user the transmitted cardiovascular parameter is.
- the user terminal Ti in this example may be a wearable device, such as a wristband, a wrist watch (such as an Apple Watch, a Huawei Watch, etc.), smart glasses, a waist-worn part, a chest-worn part, etc., in any device that can be worn or worn by a human subject. one or a combination thereof.
- the user terminal Ti may also include a wearable device and a portable electronic terminal such as a smart phone and an IPAD; in this example, the wearable device is communicatively connected with a portable electronic terminal such as a smart phone and an IPAD.
- the smart watch used for i and the smart phone of user i are connected in communication with each other to form a user terminal Ti associated with user i.
- the wearable device collects and collects the vital sign data of the associated user.
- the user terminal T1 transmits the cardiovascular parameters of the user 1 to the server 10
- the user terminal T2 transmits the cardiovascular parameters of the user 2 to the server 10
- the user terminal Tn transmits the cardiovascular parameters of the user n to the server 10.
- the received cardiovascular parameters of each user can be stored in the memory 104 of the server 10, and the processor 102 reads the data in the memory 104 and performs analysis.
- the cardiovascular prediction classifier 1020 in the processor 102 is a pre-built model. It is trained using a labeled dataset that includes M users (patients). Each training user i is represented by a vector of user data (eg vital signs, patient history) and a label Yi. The elements of the training user data vector are also referred to in the application as "features" commonly used in classifier training techniques. Label Yi indicates whether training user i is diagnosed with a cardiovascular state deterioration for which the classifier is being trained. The flag can be a binary value representing good and bad, or it can be a range, such as between 0 and 3, where 0 to 3, for example, in turn represent the level of state deterioration. Other ways of marking can also be considered.
- the cardiovascular prediction classifier 1020 is pre-trained based on a large number of existing data sets, for example, the labeled data sets of M users (patients). Once loaded into the processor 102, the cardiovascular prediction classifier 1020 processes the cardiovascular parameters entered into it for each user to give a signature characterizing their state. According to some examples of the present application, the cardiovascular prediction classifier 1020 in the processor 102 is trainable online, ie, after being set into the processor 102, it can be further trained for optimization.
- the cardiovascular parameters of user 1 , user 2 , and user n respectively enter the cardiovascular prediction classifier 1020 .
- the classifier 1020 will analyze and label the cardiovascular parameters of user 1, and finally output the label value Y1; analyze and label the cardiovascular parameters of user 2, and finally output the label value Y2; The vascular parameters are analyzed and labeled, and the labeled value Yn is finally output.
- the cloud server 10 is further configured to send the tag value for each user output by the processor 102 to the user terminal of the corresponding user. For example, the flag value Y1 is sent to user terminal T1 of user 1, the flag value Y2 is sent to user terminal T2 of user 2, and the flag value Yn is sent to user terminal Tn of user n.
- a preset threshold value representing a deterioration situation is set for the cardiovascular health state, and when the value represented by the marker or the marker value given by the cardiovascular prediction classifier 1020 exceeds the preset threshold value , then a warning signal is generated and sent to the corresponding user equipment.
- the flag value may include two situations, one is that the flag value is greater than the preset threshold value, and the other is that the flag value is smaller than the preset threshold value. In the former case, it indicates that the larger the mark value, the worse the deterioration or the greater the risk of deterioration. The greater the risk. Which method to choose can be set according to the actual scene.
- the cardiovascular prediction classifier 1020 is an XGBoost based cardiovascular prediction classifier.
- XGBoost is an optimized distributed boosting library that can efficiently and flexibly predict massive data, and solve prediction and regression problems by using boosted decision trees.
- the present application constructs the cardiovascular prediction classifier 1020 based on XGBoost.
- XGBoost is a known technology, and details are not described here.
- XGBoost is trained for t rounds by formula (1), wherein the training loss function of the t-th round is:
- the function Indicates the loss function between the prediction result after the first t-1 rounds of training and the given training label.
- the cross-entropy loss function is selected, t is a positive integer, x is the variable representing the user, and x i is the i-th Cardiovascular parameters of the user.
- y is a variable representing the tag, and y i is the tag of the i-th user.
- x ij appears, it indicates the jth cardiovascular parameter of the ith user.
- g i and h i are the coefficients of the second-order Taylor expansion of the function l with respect to x i .
- f t ( xi ) represents the prediction result in the t-th round.
- ⁇ (f t ) represents the regularization term of the model trained in the t-th round.
- the L2 regularization of the CART tree can be used.
- t-th round of training first obtain the loss function of formula (1), and use the gradient descent method to obtain its descending gradient with respect to the parameters, thereby optimizing the parameters of the t-th round of training, and then perform the t+1-th round of training.
- the number of rounds of XGBoost training is determined by subsequent model tuning.
- FIG. 2 is a schematic structural diagram of a wearable device according to an example of the present application.
- the wearable device 20 includes a plurality of sensors for sensing the vital signs of the wearer. Examples include sensors for sensing heart rate, pulse, body temperature, exercise volume, blood sugar, blood pressure, blood samples, blood lipids, and respiration, respectively, which sense cardiovascular parameters that can be used to determine or diagnose the wearer's cardiovascular state. Wearable device 20 may also include sensors for sensing other vital signs not listed here, such as vital signs useful in determining cardiovascular status.
- FIG. 2 only illustrates the sensor 200 and does not therefore limit the type and number of sensors.
- the processor 202 receives the sensed data transmitted by each sensor 200, and at least converts the sensed vital signs into data suitable for transmission.
- the communication module 204 transmits the data that has been converted into data suitable for transmission.
- the sensor 200 senses the wearer's vital sign data, and can also be configured to add a time stamp to the sensed data, so that the component, device, etc. that obtains the data, when using the vital sign data, can explicitly know the acquisition The time of these data, which is especially beneficial for observing the user's (or patient's) cardiovascular state through historical data. It should be pointed out that there is a corresponding relationship between the wearer and the wearable device, so as to ensure that the data sensed and uploaded by the wearable device is for a specific wearer, for example, through password, fingerprint when wearing , face or other means to authenticate or identify the wearer.
- the processor 202 when processing the data from the sensor 200, the processor 202 also converts the acquired data into numerical values in natural units, for example, converts the sensed vital sign signal into With kg, mmHg, mmol/L and other units. It should be understood by those skilled in the art that the natural unit here is only to represent the measurement of the parameter, which does not require that the parameter itself must be maintained as an analog signal or must be converted and maintained as a digital signal.
- processor 202 is further configured to process vital sign data from sensor 200 to obtain changes in cardiovascular parameters over time.
- the processor 202 processes the vital sign data from the sensor 200 to obtain the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle. cycle.
- the difference between the current cardiovascular parameter obtained here and the corresponding parameter of the previous cycle intuitively reflects the change of the parameter in this cycle, and the change can be transmitted to the cardiovascular classifier 1020 in FIG. 1 above, for example. , for its use.
- the processor 202 performs normalization processing on the vital sign data from the sensor 200, and performs one-hot coding on the normalized processing result to obtain the one-hot coding of each parameter.
- the normalization processing is implemented by scaling processing, and the scaling methods used include but are not limited to StandardScaler, MinMaxScaler, RobustScaler, and the like.
- the data is standardized by the processor 202, so that all parameter data maintain the same dimension, for example, the range is between 0 and 100. If standardization is not performed, it is possible that the range of parameter A (for machine learning models, also called feature A) is 1-100, and the range of feature B is 0.0001-0.0002, then the output of these models is Reference, accuracy, etc. will be greatly affected.
- the processor 202 classifies the vital sign data from the sensor 200 according to the particular distribution of the vital signs to obtain classification information for each cardiovascular parameter.
- each cardiovascular parameter can be classified according to the age of the human subject, and the cardiovascular parameters can be classified according to infants, children, adolescents, adults, adults, middle-aged, elderly, and so on.
- the processor 202 is configured to cause the processor to calculate the variation of the cardiovascular parameter over time, the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle obtained after the processing as described above , the one-hot encoding of each parameter, and the classification information are transmitted to the cloud server via the communication module 200 .
- the wearable device 20 also receives information sent from the cloud server. For example, after the cloud server marks the received vital sign data of the user through its cardiovascular prediction classifier, the cloud server sends the mark to the wearable device 20 . , which is shown to the wearer, eg, visually displayed to the user by a pattern, document, or a combination of pattern and text, or the wearable device 20 communicates the marked information to the user in a voice manner, for example.
- a preset threshold may be set in the wearable device 20, and the wearable device 20 may issue a warning message when the value represented by the received mark is greater than the preset threshold.
- the preset threshold value can also be set in the cloud server, so that the cloud server generates warning information and transmits it to the wearable device 20 for sending out warning information when the value represented by the mark is greater than the preset threshold value. user.
- the wearable device 20 may be communicatively connected to a user terminal, such as the user terminal 30 shown in the figure.
- the user terminal 30 is, for example, an electronic terminal of a user such as a smart phone and an IPAD.
- the wearable device 20 and the user terminal 30 communicatively connected thereto are associated with the same user, so that the vital sign data processed and transmitted by the wearable device 20 and the user terminal 30 are data of the same person.
- the wearable device 20 transmits the converted data suitable for transmission to the user terminal 30 through the communication module 204 .
- the user terminal 30 is configured to process the vital sign data transmitted by the wearable device 20, and obtain the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle.
- the cycle can be set as required and is not fixed. 24 hours is a cycle.
- the difference between the current cardiovascular parameter obtained here and the corresponding parameter of the previous cycle intuitively reflects the change of the parameter over this cycle, which can be transmitted, for example, to the cardiovascular classifier described in conjunction with FIG. for its use.
- the user terminal 30 performs standardization processing on the vital sign data transmitted by the wearable device 20, and performs one-hot encoding on the standardized processing result to obtain one-hot encoding of each parameter.
- the data is standardized by the processor 202, so that all the parameter data keep the same dimension, for example, the range is between 0-100.
- the user terminal 30 classifies each cardiovascular parameter on the vital sign data from the sensor 200 according to the specific distribution of the vital sign to obtain classification information.
- each cardiovascular parameter can be classified according to the age of the human subject, and the cardiovascular parameters can be classified according to infants, children, adolescents, adults, adults, middle-aged, elderly, and so on.
- the user terminal 30 is configured to make the changes of the cardiovascular parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot encoding of each parameter, and the classification information sent to the cloud server. In some cases, the user terminal 30 also receives the information sent from the cloud server.
- the mark is sent to the user terminal 30, and the It is displayed to the user in a visual manner such as a pattern, a document, or a combination of pattern and text; or the user terminal 30 can also transmit relevant information to the user in a voice manner.
- a preset threshold may be set in the user terminal 30, and when the value represented by the received flag is greater than the preset threshold, the user terminal 30 may issue a warning message.
- the preset threshold value can also be set in the cloud server, whereby the cloud server generates warning information and transmits it to the user terminal 30 for sending out warning information when the value represented by the mark is greater than the preset threshold value, thereby warning the user .
- FIG. 3 is a schematic structural diagram of a cardiovascular assessment system according to an example of the present application.
- the cardiovascular detection system 4 includes a wearable device 42, a user terminal 44 and a server 40.
- Wearable device 42 includes a plurality of sensors 420 , a processor 422 and a communication module 424 .
- User terminal 44 includes communication module 440 and processor 442 .
- Server 40 includes communication module 400 and processor 402 .
- the user terminal 44 is, for example, an electronic terminal of a user such as a smartphone or an IPAD.
- the plurality of sensors 420 of the wearable device 42 are used to sense vital sign data of the wearer, the vital sign data including at least cardiovascular parameters.
- the processor 422 converts at least the sensed vital signs into data suitable for transmission.
- the communication module 424 transmits the processor-processed vital sign data.
- the sensor 420 is similar to the sensor described above in conjunction with FIG. 2 and will not be described again.
- the processor 422 of the wearable device 42 processes the data from the sensors 420, having converted it into data suitable for transmission by the communication module 424.
- the communication module 424 is a Bluetooth module, and the processor 422 converts the data into data that can be transmitted by the Bluetooth module.
- the processor 422 also converts the acquired data into numerical values in natural units, for example, converts the sensed vital sign signal into units such as kg, mmHg, mmol/L, etc., according to the parameters it represents.
- the wearable device 42 transmits the data processed by the processor 422 to the user terminal 44 through the communication module 424 .
- the communication module 440 of the user terminal 44 receives data transmitted from the communication module 424 of the wearable device 42 .
- the processor 442 of the user terminal 44 processes the received vital sign data from the wearable device 42 to obtain changes in cardiovascular parameters over time.
- the processor 442 processes the vital sign data from the wearable device 42 to obtain the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle. for one cycle.
- the difference between the current cardiovascular parameter obtained here and the corresponding parameter of the previous cycle intuitively reflects the change in the parameter over this cycle, which can be communicated, for example, to a cardiovascular classifier which will be discussed in connection with the server 40 . 1020 for its use.
- the processor 442 performs normalization processing on the vital sign data from the sensor 420, and performs one-hot coding on the normalized processing result to obtain the one-hot coding of each parameter.
- the normalization processing is implemented by scaling processing, and the scaling methods used include but are not limited to StandardScaler, MinMaxScaler, RobustScaler, and the like.
- the data is standardized by the processor 442, so that all parameter data keep the same dimension, for example, the range is between 0-100. If standardization is not performed, it is possible that the range of parameter A (for machine learning models, also called feature A) is 1-100, and the range of feature B is 0.0001-0.0002, then the output of these models is Reference, accuracy, etc. will be greatly affected.
- the processor 442 the vital sign data classifies each cardiovascular parameter according to the particular distribution of the vital signs to obtain classification information.
- each cardiovascular parameter can be classified according to the age of the human subject, and the cardiovascular parameters can be classified according to infants, children, adolescents, adults, adults, middle-aged, elderly, and so on.
- the communication module 440 transmits the variation of the cardiovascular parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot encoding of each parameter, and the classification information obtained after being processed by the processor 442. to server 40.
- the communication module 400 of the server 40 is, by way of example and not limitation, a wireless communication module that establishes communication with the communication module 440 of the user terminal 44.
- the communication module 400 of the server 40 receives the vital sign data of the corresponding user associated with each user terminal transmitted to it by the user terminal 44 .
- a cardiovascular prediction classifier 4020 is configured in the processor 402 . After the vital sign data of the corresponding user associated with each user terminal received by the communication module 100 is input to the processor 402 , the cardiovascular state of the corresponding user is marked by the cardiovascular prediction classifier 4020 .
- the processor 402 is further configured to train the cardiovascular prediction classifier 1020 with the vital sign data of the corresponding user associated with each user terminal at predetermined time intervals, and update the original cardiovascular prediction classifier with the trained classifier 1020.
- the server 40 can be implemented as the cloud server 10 as described above in conjunction with FIG. 1 , and details are not repeated here.
- the preset thresholds mentioned in the discussion in connection with FIG. 1 are not set in the server 40 .
- the preset threshold is set in the user terminal 44 .
- the user terminal 44 receives the tag sent to it by the cloud server 40, compares the value represented by the received tag with a preset threshold, and sends a warning signal when the tag value exceeds the threshold.
- the manner of issuing the warning signal includes, for example, a visual manner or a voice (including a buzzer) manner, and the like.
- the cardiovascular evaluation system of the examples of this application can include multiple wearable devices, multiple user terminals, servers, etc., and the number of each device does not vary. The limits shown in the figure are limited.
- FIG. 4 is a flowchart of a method for evaluating a cardiovascular state performed on a server side according to an example of the present application.
- step S400 vital sign data of a corresponding user associated with each user terminal transmitted from the user terminal is received, where the vital sign data at least includes cardiovascular parameters.
- step S402 the vital sign data of the corresponding user associated with each user terminal is input into the cardiovascular prediction classifier, and the classifier performs the cardiovascular status analysis of the corresponding user based on at least the cardiovascular parameters in the physical sign data. mark.
- the method further includes sending the flag to the user terminal through the communication module, as shown in step S404.
- the method for evaluating cardiovascular status performed on the server side further includes predicting the cardiovascular status at predetermined time intervals based on the vital sign data of the corresponding user associated with each user terminal.
- the classifier is trained and updated, as shown in step S406.
- the method further includes comparing the value represented by the flag with a preset threshold, and in the case of exceeding the preset threshold, generating and sending out warning information, as shown in step S408.
- the method shown in FIG. 4 has been described in detail in conjunction with the server shown in FIG. 1 .
- the method shown in FIG. 4 can be specifically implemented as a method executed in the server shown in FIG. 1 .
- step S500 a plurality of cardiovascular parameters of the user are sensed by the plurality of sensors 420 of the wearable device 42 worn by the user.
- step S502 the wearable device 42 transmits the sensed cardiovascular parameters of the user to the user terminal 44 .
- the wearable device 42 may transmit to the user terminal 44 , eg via the communication module 424 , after at least converting the sensed cardiovascular parameters into data suitable for transmission through its processor 422 .
- the user terminal 44 processes the received cardiovascular parameters to obtain the variation of the cardiovascular parameters over time and the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, and normalizes the sensed cardiovascular parameters , and perform one-hot encoding on the standardized processing results to obtain one-hot encoding of each parameter, and classify each cardiovascular parameter according to the specific distribution of cardiovascular parameters to obtain classification information; the specific distribution here refers to the age distribution.
- step S504 includes receiving the transmitted cardiovascular parameters via the communication module 440 of the user terminal 44 , and performing various processes mentioned above via the processor 422 of the user terminal 44 .
- the user terminal 44 sends the obtained changes of the cardiovascular parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot encoding of each parameter, and the classification information to the cloud server 40 .
- the cloud server 40 inputs the received changes of cardiovascular parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot encoding of each parameter, and the classification information into its processor At 402, the cardiovascular state of the user is flagged by a cardiovascular prediction classifier 4020 disposed within the processor 402.
- step S510 the cloud server 40 sends the flag to the user terminal 44 .
- the user terminal 44 shows the indicia to the user, for example by visual means to the user, or voice means to the user, or a combination of the two or other means not listed here, It is sufficient as long as the user is aware of the mark.
- step S513 is further included, comparing the value represented by the received flag with a preset threshold, and issuing a warning when the value represented by the flag exceeds the preset threshold.
- the preset threshold is a parameter threshold obtained from cardiovascular big data, for example, representing the risk of different age groups. In some examples, it is preset in the user terminal, so that the user terminal receives the information from the cloud server 40 after receiving the After marking, a comparison is made, and an alert is issued when the value represented by the marking exceeds the preset threshold.
- the preset threshold can also be optionally set in the cloud server 40, and the cloud server 40 compares the value represented by the mark with the preset threshold, and when the value represented by the mark exceeds the preset threshold, An alert signal is generated and sent to the user terminal 44 through which the user is alerted.
- the way of the user terminal 44 alerting the user may be a voice way, or a visual way such as text and pictures.
- each user's own vital sign data including cardiovascular parameters are detected by sensors provided in the wearable device, and after a series of processing, are uploaded to the cloud server.
- the vital sign data including cardiovascular parameters of each user will be stored in a cloud server as the historical data of the user, for example, and used for training cardiovascular training.
- Classifier when the cardiovascular classifier marks the user's data, the cardiovascular classifier will also refer to each user's own vital sign data including written blood vessel parameters and its changes while considering big data.
- each device, terminal, and server may include wireless communication modules, communication modules that need to be connected through cable lines to realize data transmission, and may also include Bluetooth A communication module such as a module, or one or more of the communication modules mentioned and not listed here.
- the communication module of the wearable device is, for example, a Bluetooth module.
- a user terminal of a tablet such as a smart phone and an IPAD also includes a Bluetooth module.
- the wearable device and the user terminal communicate through the Bluetooth module.
- the wearable device 42 in FIG. 3 communicates with the user terminal 44 through Bluetooth.
- the user terminal communicates with the cloud server through a wireless communication module and/or a wired communication module connected by a cable.
- the communication between the user terminal 44 and the cloud server 40 in FIG. 3 is wireless.
- the wearable device communicates directly with the cloud server
- it is a better way to communicate with the cloud server through the wireless communication module
- the Bluetooth module may not communicate with the cloud server due to the limitation of its transmission distance. the best means of communication.
- the cloud server is provided with a cardiovascular classifier that has been trained from big data (eg, a labeled dataset of M users (patients)).
- big data eg, a labeled dataset of M users (patients)
- the cardiovascular classifier of the cloud server marks the data of each user, and the mark can represent the cardiovascular state of the corresponding user.
- the mark will be returned by the cloud server to the user terminal of the corresponding user, and presented to the user by the latter, for example, in a visual manner.
- the cloud server refers to big data and the user's own historical data, relatively accurate cardiovascular status feedback can be given in time.
- a user or a patient or wearable device wearer using the server, portable device, user terminal and cardiovascular assessment system described in this application, or executing the method for assessing cardiovascular status described in this application, a user or a patient or wearable device wearer
- the daily blood pressure, blood oxygen, pulse, body temperature and other vital signs data that can characterize the user's cardiovascular state, and their changes can form a "barometer" of cardiovascular health, so that it is convenient to recognize the user's cardiovascular health earlier. development status, and provide early warning of risky or deteriorating situations.
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
Description
Claims (20)
- 一种用于评估心血管状态的服务器,其包括:A server for assessing cardiovascular status, comprising:通信模块,其被配置为能与不同的用户终端通信,以接收与各用户终端关联的相应用户的生命体征数据,所述生命体征数据至少包括心血管参数;a communication module configured to be able to communicate with different user terminals to receive corresponding user vital sign data associated with each user terminal, the vital sign data including at least cardiovascular parameters;处理器,其被配置为将与各用户终端关联的相应用户的生命体征数据输入到心血管预测分类器中,由所述分类器至少基于所述生命体征数据中的心血管参数对相应用户的心血管状态进行标记;以及A processor configured to input the corresponding user's vital sign data associated with each user terminal into a cardiovascular prediction classifier, where the classification of the corresponding user based on at least the cardiovascular parameters in the vital sign data is performed by the classifier. Cardiovascular status markers; and所述通信模块进一步被配置为将对用户的心血管状态所进行的标记,发送给与该用户相关联的用户终端。The communication module is further configured to send the marking of the cardiovascular state of the user to a user terminal associated with the user.
- 根据权利要求1所述的服务器,其中,所述处理器还被配置为以与各用户终端关联的相应用户的生命体征数据,按照预定时间间隔,对所述心血管预测分类器进行训练并更新该分类器。The server of claim 1, wherein the processor is further configured to train and update the cardiovascular prediction classifier at predetermined time intervals based on the vital sign data of the corresponding user associated with each user terminal the classifier.
- 根据权利要求1或2所述的服务器,其中,所述处理器还被配置为在所述标记代表的值超出预设阈值时,生成警示信息并通过所述通信模块发出该警示信息。The server according to claim 1 or 2, wherein the processor is further configured to generate warning information and send the warning information through the communication module when the value represented by the flag exceeds a preset threshold.
- 根据权利要求1所述的服务器,其中,所述心血管预测分类器是基于XGBoost的心血管预测分类器。The server of claim 1, wherein the cardiovascular prediction classifier is an XGBoost based cardiovascular prediction classifier.
- 根据权利要求1所述的服务器,其中,所述对所述心血管预测分类器进行训练是按照如下损失函数进行t轮次的训练,t为正整数:The server according to claim 1, wherein the training of the cardiovascular prediction classifier is t rounds of training according to the following loss function, where t is a positive integer:其中,函数 表示了前(t-1)轮训练后的预测结果和给定的训练标签之间的损失函数,一般选取交叉熵损失函数;g i和h i是函数l关于x i二阶泰勒展开的系数;x为表示用户的变量,x i即为第i个用户的各项心血管参数;y是表示标记的变量,y i即为第i个用户的标记;f t(x i)代表了在第t轮的预测结果;Ω(f t)代表的是第t轮训练的模型正则项。 Among them, the function Represents the loss function between the prediction result after the first (t-1) rounds of training and the given training label, and the cross-entropy loss function is generally selected; g i and h i are the coefficients of the second-order Taylor expansion of the function l with respect to x i ; x is the variable representing the user, xi is the cardiovascular parameters of the ith user; y is the variable representing the mark, yi is the marker of the ith user; f t ( xi ) represents the The prediction result of the t-th round; Ω(f t ) represents the regularization term of the model trained in the t-th round.
- 一种可穿戴设备,其包括:A wearable device comprising:多个传感器,其被配置用于感测佩戴者的生命体征数据,所述生命体征数据至少包括心血管参数;a plurality of sensors configured to sense vital sign data of the wearer, the vital sign data including at least cardiovascular parameters;处理器,其被配置为至少将所感测的生命体征数据转换为适于传输的数据;a processor configured to convert at least the sensed vital sign data into data suitable for transmission;通信模块,其被配置为发送所述处理器处理过的生命体征数据。A communication module configured to transmit the processor-processed vital sign data.
- 根据权利要求6所述的可穿戴设备,其中,所述处理器进一步被配置为:The wearable device of claim 6, wherein the processor is further configured to:处理所感测的生命体征数据,以获得心血管参数随时间的变化情况;processing sensed vital sign data to obtain changes in cardiovascular parameters over time;处理所感测的生命体征数据,以获得心血管参数和上一周期的相应参数间的差值;processing the sensed vital sign data to obtain the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle;对所感测的生命体征数据进行标准化处理,并对标准化处理结果做one-hot编码以获得所述心血管参数的one-hot编码;Standardize the sensed vital sign data, and perform one-hot encoding on the standardized processing result to obtain the one-hot encoding of the cardiovascular parameter;按照所述生命体征数据的特定分布对所述心血管参数分类以获得分类信息。The cardiovascular parameters are classified according to a specific distribution of the vital sign data to obtain classification information.
- 根据权利要求7所述的可穿戴设备,其中,所述处理器进一步被配置为使所述心血管参数随时间的变化情况、所述心血管参数和上一周期的相应参数间的差值、各参数的one-hot编码、以及所述分类信息由所述通信模块传送给服务器。7. The wearable device of claim 7, wherein the processor is further configured to make the change of the cardiovascular parameter over time, the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle, The one-hot encoding of each parameter and the classification information are transmitted to the server by the communication module.
- 根据权利要求8所述的可穿戴设备,其中,所述通信模块接收来自所述服务器发送的标记,所述标记由所述服务器的心血管预测分类器至少基于所述可穿戴设备发送给其的数据而对所述可穿戴设备的佩戴者的心血管状态进行标记,并且,所述可穿戴设备的处理器被配置为:9. The wearable device of claim 8, wherein the communication module receives an indicia sent from the server, the indicia being sent to it by a cardiovascular prediction classifier of the server based at least on the information sent to it by the wearable device The cardiovascular state of the wearer of the wearable device is tagged with the data, and the processor of the wearable device is configured to:将所接收的标记示出给所述佩戴者;和/或showing the received indicia to the wearer; and/or将所接收的标记代表的数值与预设阈值比较,并在所述标记代表的数值超出所述预设阈值时,发出警示。The value represented by the received flag is compared with a preset threshold, and an alert is issued when the value represented by the flag exceeds the preset threshold.
- 根据权利要求6所述的可穿戴设备,其中,所述处理器被配置为将转换后适于传输的数据通过所述通信模块传送给与该可穿戴设备关联的用户终端,所述用户终端被配置为:The wearable device of claim 6, wherein the processor is configured to transmit the converted data suitable for transmission to a user terminal associated with the wearable device through the communication module, the user terminal being Configured as:处理所感测的生命体征数据,以获得心血管参数随时间的变化情况;processing sensed vital sign data to obtain changes in cardiovascular parameters over time;处理所感测的生命体征数据,以获得心血管参数和上一周期的相应参数间的差值;processing the sensed vital sign data to obtain the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle;对所感测的生命体征数据进行标准化处理,并对标准化处理结果做one-hot编码以获得所述心血管参数的one-hot编码;Standardize the sensed vital sign data, and perform one-hot encoding on the standardized processing result to obtain the one-hot encoding of the cardiovascular parameter;按照所述生命体征数据的特定分布对所述心血管参数分类以获得分类信息。The cardiovascular parameters are classified according to a specific distribution of the vital sign data to obtain classification information.
- 根据权利要求10所述的可穿戴设备,其中,所述用户终端被进一步配置为使所述心血管参数随时间的变化情况、所述心血管参数和上一周期的相应参数间的差值、所述心血管参数的one-hot编码、以及所述分类信息传送给服务器。The wearable device according to claim 10, wherein the user terminal is further configured to make the change of the cardiovascular parameter over time, the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle, The one-hot encoding of the cardiovascular parameters and the classification information are transmitted to the server.
- 根据权利要求6所述的可穿戴设备,其被实现为腕表、智能眼镜、腰部佩戴部件、胸部佩戴部件中的一种或其结合。The wearable device of claim 6, implemented as one of a wrist watch, smart glasses, a waist-worn part, a chest-worn part, or a combination thereof.
- 一种用户终端,其被配置为分别与可穿戴设备和远端服务器通信,其中,所述用户终端与所述可穿戴设备相关联使得该两者用于同一用户,所述用户终端被配置为:A user terminal configured to communicate with a wearable device and a remote server, respectively, wherein the user terminal is associated with the wearable device such that both are used for the same user, the user terminal is configured to :接收来自所述可穿戴设备传送的生命体征数据,所述生命体征数据至少包括心血管参数;receiving vital sign data transmitted from the wearable device, the vital sign data including at least cardiovascular parameters;处理所述生命体征数据,以获得心血管参数随时间的变化情况;processing the vital sign data to obtain changes in cardiovascular parameters over time;处理所述生命体征数据,以获得心血管参数和上一周期的相应参数间的差值;processing the vital sign data to obtain the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle;对所述生命体征数据进行标准化处理,并对标准化处理结果做one-hot编码以获得所述心血管参数的one-hot编码;Standardizing the vital sign data, and performing one-hot encoding on the standardized processing result to obtain the one-hot encoding of the cardiovascular parameter;按照所述生命体征数据的特定分布对所述心血管参数分类以获得分类信息;以及将classifying the cardiovascular parameters according to a specific distribution of the vital sign data to obtain classification information; and所获得心血管参数随时间的变化情况、所述心血管参数和上一周期的相应参数间的差值、所述心血管参数的one-hot编码、以及所述分类信息传送给服务器。The changes of the obtained cardiovascular parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot encoding of the cardiovascular parameters, and the classification information are transmitted to the server.
- 根据权利要求13所述的用户终端,其还被配置为接收由所述服务器发送的标记,所述标记由所述服务器的心血管预测分类器至少基于所述用户终端发送给所述服务器的所述心血管参数随时间的变化情况、所述心血管参数和上一周期的相应参数间的差值、所述心血管参数的one-hot编码、以及所述分类信息而对用户的心血管状态所进行的标记,并且,所述用户终端进一步被配置为:14. The user terminal of claim 13, further configured to receive an indicia sent by the server, the indicia by a cardiovascular prediction classifier of the server based at least on the information sent by the user terminal to the server The changes of the cardiovascular parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot coding of the cardiovascular parameters, and the classification information are related to the cardiovascular state of the user. marked, and the user terminal is further configured to:将所接收的标记示出给所述用户;和/或show the received indicia to the user; and/or将所接收的标记代表的值与预设阈值比较,并在所述标记代表的数值超出所述预设阈值时,发出警示。The value represented by the received flag is compared with a preset threshold, and an alert is issued when the value represented by the flag exceeds the preset threshold.
- 一种心血管状态评估系统,其包括可穿戴设备、用户终端以及服务器:A cardiovascular state assessment system, comprising a wearable device, a user terminal and a server:所述可穿戴设备包括:The wearable device includes:多个传感器,其被配置为用于感测佩戴者的生命体征数据,所述生命体征数据至少包括心血管参数;a plurality of sensors configured to sense vital sign data of the wearer, the vital sign data including at least cardiovascular parameters;处理器,其被配置为至少将所感测的生命体征数据转换为适于传输的数据;a processor configured to convert at least the sensed vital sign data into data suitable for transmission;通信模块,其被配置为发送所述处理器处理过的生命体征数据;a communication module configured to transmit the processor-processed vital sign data;所述用户终端包括:The user terminal includes:通信模块,其被配置为与所述可穿戴设备的通信模块建立通信,并接收其发送的生命体征数据;a communication module configured to establish communication with the communication module of the wearable device and receive vital sign data sent by it;处理器,其被配置为:A processor, which is configured to:处理所述生命体征数据,以获得心血管参数随时间的变化情况;processing the vital sign data to obtain changes in cardiovascular parameters over time;处理所感测的生命体征数据,以获得心血管参数和上一周期的相应参数间的差值;processing the sensed vital sign data to obtain the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle;对所述生命体征数据进行标准化处理,并对标准化处理结果做one-hot编码以获得各参数的one-hot编码;Standardize the vital sign data, and perform one-hot encoding on the standardized processing result to obtain the one-hot encoding of each parameter;按照所述生命体征数据的特定分布对各心血管参数分类以获得分类信息;以及classifying each cardiovascular parameter according to a particular distribution of said vital sign data to obtain classification information; and其中,所述通信模块被进一步配置为将所获得心血管参数随时间的变化情况、所获得心血管参数和上一周期的相应参数间的差值、所获得各参数的one-hot编码以及所获得的得分类信息发送给所述服务器;Wherein, the communication module is further configured to record the variation of the obtained cardiovascular parameters with time, the difference between the obtained cardiovascular parameters and the corresponding parameters of the previous cycle, the obtained one-hot encoding of each parameter, and the obtained one-hot encoding of each parameter. The obtained score category information is sent to the server;所述服务器包括:The server includes:通信模块,其用于接收与所述用户终端的通信模块所发送的经其处理器处理的生命体征数据;a communication module, configured to receive vital sign data sent by the communication module of the user terminal and processed by its processor;处理器,其被配置为将所接收的生命体征数据输入到心血管预测分类器中,由所述分类器至少基于所述体征数据中的心血管参数对用户的心血管状态进行标记;a processor configured to input the received vital sign data into a cardiovascular prediction classifier that labels the cardiovascular state of the user based at least on cardiovascular parameters in the vital sign data;所述通信模块进一步被配置为将所述标记发送给所述用户终端;以及the communication module is further configured to transmit the indicia to the user terminal; and其中,所述用户终端则进一步被配置为以可视化方式和/或语音方式向用户示出所述标记。Wherein, the user terminal is further configured to show the mark to the user in a visual manner and/or a voice manner.
- 根据据权利要求15所述的心血管状态评估系统,其中,所述用户终端被配置为在所述标记代表的数值超出预设阈值时,生成警示信息并通过所述通信模块发出该警示信息。The cardiovascular state assessment system according to claim 15, wherein the user terminal is configured to generate warning information and send the warning information through the communication module when the value represented by the mark exceeds a preset threshold.
- 根据权利要求16所述的心血管状态评估系统,其中,所述心血管预测分类器是基于XGBoost的心血管预测分类器。The cardiovascular state assessment system of claim 16, wherein the cardiovascular prediction classifier is an XGBoost based cardiovascular prediction classifier.
- 一种用于在服务器端执行的评估心血管状态的方法,其包括:A method for evaluating cardiovascular status performed on a server side, comprising:接收来自用户终端传送的与各用户终端关联的相应用户的生命体征数据,所述生命体征数据至少包括心血管参数;receiving vital sign data of a corresponding user associated with each user terminal transmitted from the user terminal, where the vital sign data at least includes cardiovascular parameters;将与各用户终端关联的相应用户的生命体征数据输入到心血管预测分类器中,由所述分类器至少基于所述体征数据中的心血管参数对相应用户的心血管状态进行标记;inputting the vital sign data of the corresponding user associated with each user terminal into a cardiovascular prediction classifier, and the classifier marking the cardiovascular state of the corresponding user based on at least the cardiovascular parameters in the physical sign data;将所述标记发送给相应的用户终端以便其示出该标记。The indicia is sent to the corresponding user terminal so that it displays the indicia.
- 一种评估心血管状态的方法,其包括:A method of assessing cardiovascular status, comprising:通过用户佩戴的可穿戴设备感测用户的心血管参数;Sensing the user's cardiovascular parameters through a wearable device worn by the user;所述可穿戴设备将所感测的所述用户的心血管参数传送给用户终端;The wearable device transmits the sensed cardiovascular parameters of the user to the user terminal;所述用户终端处理所感测的生命体征数据以获得心血管参数随时间的变化情况及心血管参数和上一周期的相应参数间的差值;所述用户终端对所感测的生命体征数据进行标准化处理,并对标准化处理结果做one-hot编码以获得各参数的one-hot编码;以及所述用户终端按照所述生命体征数据的特定分布对各心血管参数分类以获得分类信息;The user terminal processes the sensed vital sign data to obtain the variation of the cardiovascular parameter with time and the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle; the user terminal standardizes the sensed vital sign data processing, and one-hot coding is done to the standardized processing result to obtain the one-hot coding of each parameter; and the user terminal classifies each cardiovascular parameter according to the specific distribution of the vital sign data to obtain classification information;所述用户终端将所获得的心血管参数随时间的变化情况、心血管参数和上一周期的相应参数间的差值、各参数的one-hot编码以及分类信息发送给云端服务器;The user terminal sends the obtained changes of the cardiovascular parameters over time, the difference between the cardiovascular parameters and the corresponding parameters of the previous cycle, the one-hot encoding of each parameter and the classification information to the cloud server;所述云端服务器将所接收的心血管参数随时间的变化情况、心血管参数和上一周期的相应参数间的差值、各参数的one-hot编码以及分类信息输入到心血管预测分类器中,由其对用户的心血管状态进行标记;The cloud server inputs the received cardiovascular parameter changes over time, the difference between the cardiovascular parameter and the corresponding parameter of the previous cycle, the one-hot encoding of each parameter, and the classification information into the cardiovascular prediction classifier. , which marks the user's cardiovascular status;所述云端服务器将所述标记发送给所述用户终端;the cloud server sends the mark to the user terminal;所述用户终端向用户示出所述标记。The user terminal shows the indicia to the user.
- 根据权利要求19所述的评估心血管状态的方法,其还包括将所接收的标记代表的值与预设阈值比较,并在所述标记代表的值超出所述预设阈值的情况下,发出警示。20. The method of assessing a cardiovascular state of claim 19, further comprising comparing the value represented by the received indicia to a preset threshold, and in the event that the value represented by the indicia exceeds the preset threshold, issuing Warning.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2020/112638 WO2022041225A1 (en) | 2020-08-31 | 2020-08-31 | Server for use in assessing cardiovascular state, wearable device, and method for cardiovascular state assessment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2020/112638 WO2022041225A1 (en) | 2020-08-31 | 2020-08-31 | Server for use in assessing cardiovascular state, wearable device, and method for cardiovascular state assessment |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022041225A1 true WO2022041225A1 (en) | 2022-03-03 |
Family
ID=80354350
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/112638 WO2022041225A1 (en) | 2020-08-31 | 2020-08-31 | Server for use in assessing cardiovascular state, wearable device, and method for cardiovascular state assessment |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2022041225A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118383729A (en) * | 2024-06-26 | 2024-07-26 | 无锡智臻医疗有限公司 | Patient sign early warning method and system based on Internet of things |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9811818B1 (en) * | 2002-10-01 | 2017-11-07 | World Award Academy, World Award Foundation, Amobilepay, Inc. | Wearable personal digital device for facilitating mobile device payments and personal use |
CN109157202A (en) * | 2018-09-18 | 2019-01-08 | 东北大学 | A kind of cardiovascular disease early warning system based on more physiological signal depth integrations |
CN110766140A (en) * | 2019-10-22 | 2020-02-07 | 电子科技大学 | XGboost and LSTM-based multi-sensor real-time threat level classification method |
CN110868404A (en) * | 2019-11-05 | 2020-03-06 | 北京航空航天大学 | Industrial control equipment automatic identification method based on TCP/IP fingerprint |
CN111356401A (en) * | 2017-11-29 | 2020-06-30 | 威里利生命科学有限责任公司 | Continuous detection and monitoring of cardiac arrhythmias using wearable sensors and cloud residency analysis |
-
2020
- 2020-08-31 WO PCT/CN2020/112638 patent/WO2022041225A1/en active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9811818B1 (en) * | 2002-10-01 | 2017-11-07 | World Award Academy, World Award Foundation, Amobilepay, Inc. | Wearable personal digital device for facilitating mobile device payments and personal use |
CN111356401A (en) * | 2017-11-29 | 2020-06-30 | 威里利生命科学有限责任公司 | Continuous detection and monitoring of cardiac arrhythmias using wearable sensors and cloud residency analysis |
CN109157202A (en) * | 2018-09-18 | 2019-01-08 | 东北大学 | A kind of cardiovascular disease early warning system based on more physiological signal depth integrations |
CN110766140A (en) * | 2019-10-22 | 2020-02-07 | 电子科技大学 | XGboost and LSTM-based multi-sensor real-time threat level classification method |
CN110868404A (en) * | 2019-11-05 | 2020-03-06 | 北京航空航天大学 | Industrial control equipment automatic identification method based on TCP/IP fingerprint |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118383729A (en) * | 2024-06-26 | 2024-07-26 | 无锡智臻医疗有限公司 | Patient sign early warning method and system based on Internet of things |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106580282A (en) | Human body health monitoring device, system and method | |
JP5669787B2 (en) | Residue-based management of human health | |
Zhang et al. | A context-aware mhealth system for online physiological monitoring in remote healthcare | |
KR20200005987A (en) | System and method for diagnosing cognitive impairment using touch input | |
WO2016167923A1 (en) | Prediction of health status from physiological data | |
CN117438024B (en) | Intelligent acquisition and analysis system and method for acute diagnosis patient sign data | |
EP3238611B1 (en) | A method and device for estimating a condition of a person | |
CN117598700B (en) | Intelligent blood oxygen saturation detection system and method | |
US11670421B2 (en) | Method and system enabling digital biomarker data integration and analysis for clinical treatment impact | |
CN111524093A (en) | Intelligent screening method and system for abnormal tongue picture | |
CN111227789A (en) | Human health monitoring method and device | |
CN113990500A (en) | Vital sign parameter monitoring method and device and storage medium | |
CN117643461A (en) | Heart rate intelligent monitoring system and method based on artificial intelligence | |
CN117438087A (en) | Health management method and system based on health monitoring data analysis | |
CN118280606B (en) | Intelligent accompanying equipment control method and system | |
WO2022041225A1 (en) | Server for use in assessing cardiovascular state, wearable device, and method for cardiovascular state assessment | |
CN111403031A (en) | Physique test data acquisition and detection system and method | |
CN113096792B (en) | Intelligent health monitoring and early warning method and system based on machine learning | |
CN117831743A (en) | Household care health monitoring management system and method based on Internet of things technology | |
CN109842682A (en) | A kind of study of distributed environment safety and method for early warning based on Internet of Things | |
CN116098595B (en) | System and method for monitoring and preventing sudden cardiac death and sudden cerebral death | |
CN112057053A (en) | Abnormal body temperature sequence detection device and method based on SVM | |
Orpa et al. | An IoT based healthcare solution with ESP32 using machine learning model | |
CN117322852A (en) | Intelligent clothing health monitoring system and method based on deep learning | |
KR102645192B1 (en) | Electronic device for managing bedsores based on artificial intelligence model and operating method thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20950894 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20950894 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 27.06.2023) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20950894 Country of ref document: EP Kind code of ref document: A1 |