Multipurpose medical monitoring devices and methods

Abstract

This invention discloses a multi-purpose medical monitoring device and method. The device includes: a vital signs monitoring module, a data processing module, an alarm analysis module, and a vital signs assessment module. The vital signs monitoring module collects a user's vital signs dataset and sends it to the data processing module. The data processing module processes the vital signs dataset to obtain a standard vital signs dataset. The alarm analysis module performs anomaly detection processing on the standard vital signs dataset to obtain alarm information. The vital signs assessment module evaluates the standard vital signs dataset to obtain the user's vital signs assessment value. This multi-purpose medical monitoring device and method utilizes collected data to analyze and detect the user's health and vital signs, helping medical personnel better understand the patient's condition, develop treatment plans, and evaluate treatment effectiveness.

Description

Technical Field

[0001] This invention relates to the field of medical devices and medical monitoring technology, and in particular to a multi-purpose medical monitoring device and method. Background Technology

[0002] Currently, general-purpose medical monitors are among the most commonly used medical devices. They monitor vital signs such as heart rate, respiration, body temperature, and blood oxygen saturation, transmitting the data to a central station or nurses' station for further processing and analysis. These devices typically use non-invasive sensors to collect physiological data and transmit it to the central station or nurses' station via wired or wireless means. In hospital clinical settings, general-purpose medical monitors are often used in conjunction with other medical equipment, such as ventilators, infusion pumps, and electrocardiographs. While existing general-purpose medical monitors can collect vital sign data, their ability to analyze and process this data is relatively weak. This means that medical staff need to spend more time and effort analyzing this data, potentially overlooking important information. Furthermore, existing monitors lack robust alarm mechanisms, failing to promptly detect abnormal patient conditions, which may lead to delays or errors in treatment. General-purpose medical monitors usually have fixed functions and configurations, which cannot adequately meet the diverse clinical needs of different individuals. This means that if the vital sign data to be monitored exceeds their original monitoring range, or if additional functional modules are required, existing monitors cannot meet the requirements. In addition, existing patient monitors have poor scalability, and cannot be flexibly modified and upgraded at the software or hardware level, which limits their flexibility and scalability in future development and application. Summary of the Invention

[0003] To address the problems of weak data analysis and processing capabilities and poor scalability of existing medical monitors, this invention discloses a multi-purpose medical monitoring device and method.

[0004] In a first aspect, the present invention discloses a multi-purpose medical monitoring device, comprising: a vital signs monitoring module, a data processing module, an alarm analysis module, and a vital signs assessment module;

[0005] The vital signs monitoring module is used to collect the user's vital signs dataset and send the vital signs dataset to the data processing module; the vital signs dataset includes heart rate sequence, respiratory rate sequence, body temperature sequence, blood oxygen saturation sequence and blood pressure sequence;

[0006] The vital signs monitoring module includes a heart rate sensor, a respiratory rate sensor, a body temperature sensor, a blood oxygen saturation sensor, and a blood pressure monitor; the heart rate sensor is used to acquire the user's heart rate sequence; the respiratory rate sensor is used to acquire the user's respiratory rate sequence; the body temperature sensor is used to acquire the user's body temperature sequence; the blood oxygen saturation sensor is used to acquire the user's blood oxygen saturation sequence; and the blood pressure monitor is used to acquire the user's blood pressure sequence.

[0007] The data processing module is used to process the vital signs dataset to obtain a standard vital signs dataset; the data processing module is connected to the vital signs monitoring module, the alarm analysis module, and the vital signs assessment module respectively.

[0008] The alarm analysis module is used to perform anomaly detection processing on the standard vital signs dataset to obtain alarm information;

[0009] The vital signs assessment module is connected to the data processing module and is used to assess and process the standard vital signs dataset to obtain the user's vital signs assessment value.

[0010] The data processing module includes a time alignment submodule, a normalization processing submodule, and a filtering processing submodule; the time alignment submodule is connected to the normalization processing submodule, and the normalization processing submodule is connected to the filtering processing submodule.

[0011] The time alignment submodule is used to perform time alignment on each data sequence in the vital signs dataset to obtain a time-aligned vital signs dataset.

[0012] The normalization processing submodule is used to normalize the time-aligned vital signs dataset to obtain a normalized vital signs dataset.

[0013] The filtering submodule is used to filter the normalized vital signs dataset to obtain a standard vital signs dataset.

[0014] The alarm analysis module includes a feature fitting analysis submodule, a deviation analysis submodule, and an alarm discrimination submodule; the feature fitting analysis submodule is connected to the deviation analysis submodule, and the deviation analysis submodule is connected to the alarm discrimination submodule.

[0015] The feature fitting analysis submodule is used to perform feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial.

[0016] The deviation analysis submodule is used to perform deviation analysis on each medical data sequence of the standard vital signs dataset to obtain a set of deviation sequence values.

[0017] The alarm discrimination submodule is used to calculate and discriminate the deviation sequence value set and the best consistent approximation polynomial to obtain alarm information.

[0018] A second aspect of this invention discloses a multi-purpose medical monitoring method, implemented using the aforementioned multi-purpose medical monitoring device, comprising:

[0019] S1, using the vital signs monitoring module, collect the user's vital signs dataset and send the vital signs dataset to the data processing module;

[0020] S2, using the data processing module, perform time alignment on each data sequence in the vital signs dataset to obtain a time-aligned vital signs dataset;

[0021] S3, normalize the time-aligned vital signs dataset to obtain a normalized vital signs dataset;

[0022] S4, the normalized vital signs dataset is filtered to obtain a standard vital signs dataset;

[0023] S5, using the alarm analysis module, perform feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial;

[0024] S6, perform deviation analysis on each medical data sequence of the standard vital signs dataset to obtain a set of deviation sequence values;

[0025] S7, perform calculation and discrimination processing on the set of deviation index values ​​and the best consistent approximation polynomial to obtain alarm information;

[0026] S8. Using the vital signs assessment module, the standard vital signs dataset is evaluated and processed to obtain the user's vital signs assessment value.

[0027] The normalization process of the time-aligned vital signs dataset to obtain a normalized vital signs dataset includes:

[0028] S31, for each medical data sequence q in the time-aligned vital signs dataset i To obtain the reasonable range of values ​​for the corresponding medical parameters [e] i1 ,e i2 ];q i Let q represent the i-th medical data sequence in the time-aligned vital signs dataset. i =[qi1 ,q i2 ,…,q iN ], q ij This represents the j-th data point in the i-th medical data sequence, where i = 1, 2, 3, 4, 5, and represents the heart rate data sequence, respiratory rate data sequence, body temperature data sequence, blood oxygen saturation data sequence, and blood pressure data sequence, respectively. N represents the number of data points contained in the medical data sequence of the time-aligned vital signs dataset; e i1 Represents the i-th medical data sequence q i The lower limit of the reasonable range of values ​​for the corresponding medical parameter, e i2 Represents the i-th medical data sequence q i The upper limit of the reasonable range of values ​​for the corresponding medical parameters;

[0029] S32, for each data point in each medical data sequence of the time-aligned vital signs dataset, normalization calculation is performed using the reasonable value range to obtain the normalized value of the data;

[0030] The normalization calculation process is expressed as follows:

[0031]

[0032] in, For q ij The normalized value obtained after normalization calculation;

[0033] S33, after normalization calculation processing is completed for all data of each medical data sequence in the time-aligned vital signs dataset, a normalized vital signs dataset is obtained.

[0034] The filtering process performed on the normalized vital signs dataset to obtain a standard vital signs dataset includes:

[0035] S41, For each medical data sequence in the normalized vital signs dataset, a filtering transformation model is used to perform transformation processing to obtain the transformation matrix of the medical data sequence;

[0036] The processing expression of the filtering transformation model is:

[0037]

[0038] Where v() represents a medical data sequence in the normalized vital signs dataset, v(k) is the kth element of the medical data sequence, Y(m,n) is the element in the mth row and nth column of the transformation matrix of the medical data sequence, h1() is the transformation function corresponding to the filtering transformation model, h1(kT1-mT1) is the value of h1() at kT1-mT1, and T1 and F1 are the time-domain transformation length and frequency-domain transformation length of the filtering transformation model, respectively.

[0039] S42, for each element of the transformation matrix of the medical data sequence, determine whether its element value is within the preset filtering interval [z1, z2]. If it is within the filtering interval, do not process the element. If it is not within the filtering interval, when the element value is less than z1, set the element value to z1, and when the element value is greater than z2, set the element value to z2.

[0040] S43, after the judgment and setting process described in S42 has been completed for each element of the transformation matrix of the medical data sequence, an updated transformation matrix is ​​obtained; the updated transformation matrix is ​​confirmed to be a filtered medical matrix.

[0041] S44, The filtered medical matrix is ​​transformed using the inverse filter transformation model to obtain the filtered data sequence of the medical data sequence;

[0042] The processing expression of the inverse filter transform model is as follows:

[0043]

[0044] Where v′() represents the filtered data sequence of the medical data sequence, v′(k) is the k-th element of the filtered data sequence of the medical data sequence, Y′(m,n) is the element in the m-th row and n-th column of the filtered medical matrix, and YM and YN are the number of rows and columns of the filtered medical matrix, respectively. The conjugate function of the transform function corresponding to the inverse filter transform model is . for The values ​​at kT1-mT1 are T1 and F1, respectively, representing the time-domain transform length and frequency-domain transform length of the inverse filter transform model.

[0045] S45, using the filtered data sequences corresponding to all medical data sequences contained in the normalized vital signs dataset, a standard vital signs dataset is constructed.

[0046] The step of performing feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial includes:

[0047] S51, using the standard vital signs dataset, a standard vital signs matrix is ​​constructed; the row vectors of the standard vital signs matrix are medical data sequences in the standard vital signs dataset.

[0048] S52, decompose the standard vital signs matrix to obtain the left decomposition matrix, feature matrix and right decomposition matrix of the standard vital signs matrix;

[0049] The decomposition process is calculated using the following expression:

[0050] Y = UAV,

[0051] Where U is the left decomposition matrix, A is the characteristic matrix, V is the right decomposition matrix, U and V are both orthogonal matrices, and A is a diagonal matrix;

[0052] S53, extract the diagonal elements of the feature matrix to obtain the feature vector; the feature vector is represented as I. a I a =[λ1,λ2,…,λ N1 N1 is the number of elements contained in the feature vector;

[0053] S54, perform linear fitting on the elements and element index values ​​of the feature vector to obtain the best uniform approximation polynomial;

[0054] The linear fitting process involves using the characteristic vector element index value Ix as the known independent variable and the characteristic vector element value as the known dependent variable. The curve to be approximated is constructed using the known independent and dependent variables. The curve to be approximated is then fitted using the function approximation method to obtain the best uniform approximation polynomial f(Ix).

[0055] The deviation analysis performed on each medical data sequence of the standard vital signs dataset yields a set of deviation index values, including:

[0056] S61, For each medical data sequence in the standard vital signs dataset, calculate the average value u0 of the medical data sequence;

[0057] S62, for each element in the medical data sequence, calculate the absolute value of the difference between it and the average value u0, and determine the absolute value of the difference as the offset of the element;

[0058] S63, for the medical data sequence, find the element with the largest offset, and determine the index value of the element in the medical data sequence as the offset index value of the medical data sequence;

[0059] S64, using the deviation index values ​​of all medical data sequences in the standard vital signs dataset, construct a set of deviation index values.

[0060] The step of using the vital signs assessment module to evaluate and process the standard vital signs dataset to obtain the user's vital signs assessment values ​​includes:

[0061] S81, using the standard vital signs dataset, a standard vital signs matrix is ​​constructed; the row vectors of the standard vital signs matrix are medical data sequences in the standard vital signs dataset.

[0062] S82, calculate the cross-correlation value of the standard vital signs matrix to obtain the cross-correlation matrix;

[0063] S83, Perform feature extraction processing on the cross-correlation matrix to obtain a weighted vector;

[0064] S84, using the weighted vector, perform a weighted summation on the set of deviation index values ​​to obtain the user's vital sign assessment value.

[0065] The calculation and discrimination process of the deviation index value set and the best consistent approximation polynomial to obtain alarm information includes:

[0066] S71, take each deviation index value in the set of deviation index values ​​as an independent variable, and use the best uniform approximation polynomial to calculate and process the independent variable to obtain the function value of the independent variable; determine the function value of the independent variable as the weight value of the deviation index value;

[0067] S72, after obtaining the weight values ​​of all deviation index values ​​in the deviation index value set, the corresponding deviation index values ​​are weighted and summed using the weight value of each deviation index value to obtain the alarm judgment value.

[0068] S73, determine whether the alarm discrimination value is greater than the alarm threshold. If it is greater than the alarm threshold, confirm that the alarm information is that the user's vital signs are abnormal; if it is less than or equal to the alarm threshold, confirm that the alarm information is that the user's vital signs are normal.

[0069] The beneficial effects of this invention are as follows:

[0070] The multi-purpose medical monitoring device and method of the present invention can collect, store and process more data, and use the collected data to analyze and detect the user's health and vital signs, helping medical personnel to better understand the condition, formulate treatment plans and evaluate treatment effects.

[0071] The multi-purpose medical monitoring device of this invention has a higher degree of integration and can achieve data sharing and collaboration through network and cloud computing technologies. It can also be seamlessly integrated with other medical devices and information systems. This makes medical services more intelligent and efficient. Attached Figure Description

[0072] Figure 1 This is a schematic diagram of the multi-purpose medical monitoring device of the present invention;

[0073] Figure 2 This is a schematic diagram of the multi-purpose medical monitoring method of the present invention. Detailed Implementation

[0074] To better understand the content of this invention, an embodiment is provided here.

[0075] Figure 1 This is a schematic diagram of the multi-purpose medical monitoring device of the present invention; Figure 2 This is a schematic diagram of the multi-purpose medical monitoring method of the present invention.

[0076] The first aspect of the present invention discloses a multi-purpose medical monitoring device, comprising: a vital signs monitoring module, a data processing module, an alarm analysis module, and a vital signs assessment module;

[0077] The vital signs monitoring module is used to collect the user's vital signs dataset and send the dataset to the data processing module. The vital signs dataset includes heart rate, respiratory rate, body temperature, blood oxygen saturation, and blood pressure sequences. The vital signs monitoring module includes a heart rate sensor, a respiratory rate sensor, a body temperature sensor, a blood oxygen saturation sensor, and a blood pressure monitor. The heart rate sensor is used to collect the user's heart rate sequence; the respiratory rate sensor is used to collect the user's respiratory rate sequence; the body temperature sensor is used to collect the user's body temperature sequence; the blood oxygen saturation sensor is used to collect the user's blood oxygen saturation sequence; and the blood pressure monitor is used to collect the user's blood pressure sequence.

[0078] The data processing module is used to process the vital signs dataset to obtain a standard vital signs dataset; the data processing module is connected to the vital signs monitoring module, the alarm analysis module, and the vital signs assessment module respectively.

[0079] The alarm analysis module is used to perform anomaly detection processing on the standard vital signs dataset to obtain alarm information;

[0080] The vital signs assessment module is used to assess and process the standard vital signs dataset to obtain the user's vital signs assessment value.

[0081] The data processing module includes a time alignment submodule, a normalization processing submodule, and a filtering processing submodule; the time alignment submodule, the normalization processing submodule, and the filtering processing submodule are connected in sequence.

[0082] The time alignment submodule is used to perform time alignment on each data sequence in the vital signs dataset to obtain a time-aligned vital signs dataset.

[0083] The normalization processing submodule is used to normalize the time-aligned vital signs dataset to obtain a normalized vital signs dataset.

[0084] The filtering submodule is used to filter the normalized vital signs dataset to obtain a standard vital signs dataset.

[0085] The time alignment is based on the sampling start time and sampling time interval of each data sequence, so that the starting data of each data sequence is collected at the same time, ensuring that data with the same sequence number in different data sequences are collected at the same time.

[0086] The alarm analysis module includes a feature fitting analysis submodule, a deviation analysis submodule, and an alarm discrimination submodule; the feature fitting analysis submodule, the deviation analysis submodule, and the alarm discrimination submodule are connected in sequence.

[0087] The feature fitting analysis submodule is used to perform feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial.

[0088] The deviation analysis submodule is used to perform deviation analysis on each medical data sequence of the standard vital signs dataset to obtain a set of deviation sequence values.

[0089] The alarm discrimination submodule is used to perform calculation and discrimination processing on the set of deviation index values ​​and the best consistent approximation polynomial to obtain alarm information.

[0090] A second aspect of this invention discloses a multi-purpose medical monitoring method, implemented using the aforementioned multi-purpose medical monitoring device, comprising:

[0091] S1, using the vital signs monitoring module, the user's vital signs dataset is collected and sent to the data processing module; the vital signs dataset includes heart rate sequence, respiratory rate sequence, body temperature sequence, blood oxygen saturation sequence and blood pressure sequence;

[0092] S2, using the data processing module, perform time alignment on each data sequence in the vital signs dataset to obtain a time-aligned vital signs dataset;

[0093] S3, normalize the time-aligned vital signs dataset to obtain a normalized vital signs dataset;

[0094] S4. Filter the normalized vital signs dataset to obtain a standard vital signs dataset.

[0095] S5, using the alarm analysis module, perform feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial;

[0096] S6, perform deviation analysis on each medical data sequence of the standard vital signs dataset to obtain a set of deviation sequence values;

[0097] S7, perform calculation and discrimination processing on the set of deviation index values ​​and the best consistent approximation polynomial to obtain alarm information;

[0098] S8. Using the vital signs assessment module, the standard vital signs dataset is evaluated and processed to obtain the user's vital signs assessment value.

[0099] The normalization process of the time-aligned vital signs dataset to obtain a normalized vital signs dataset includes:

[0100] S31, for each medical data sequence q in the time-aligned vital signs dataset i To obtain the reasonable range of values ​​for the corresponding medical parameters [e] i1 ,e i2 ];q i Let q represent the i-th medical data sequence in the time-aligned vital signs dataset. i =[q i1 ,q i2 ,…,q iN ], q ij This represents the j-th data point in the i-th medical data sequence, where i = 1, 2, 3, 4, 5, and represents the heart rate data sequence, respiratory rate data sequence, body temperature data sequence, blood oxygen saturation data sequence, and blood pressure data sequence, respectively. N represents the number of data points contained in the medical data sequence of the time-aligned vital signs dataset; e i1 Represents the i-th medical data sequence q i The lower limit of the reasonable range of values ​​for the corresponding medical parameter, e i2 Represents the i-th medical data sequence q i The upper limit of the reasonable range of values ​​for the corresponding medical parameters;

[0101] S32, for each data point in each medical data sequence of the time-aligned vital signs dataset, normalization calculation is performed using the reasonable value range to obtain the normalized value of the data;

[0102] The normalization calculation process is expressed as follows:

[0103]

[0104] in, For q ij The normalized value obtained after normalization calculation;

[0105] S33, after all data of each medical data sequence in the time-aligned vital signs dataset have been normalized, a normalized vital signs dataset is obtained.

[0106] The reasonable range of values ​​for the medical parameters is preset or obtained based on relevant medical monitoring standards or human health standards.

[0107] The filtering process performed on the normalized vital signs dataset to obtain a standard vital signs dataset includes:

[0108] S41, For each medical data sequence in the normalized vital signs dataset, a filtering transformation model is used to perform transformation processing to obtain the transformation matrix of the medical data sequence;

[0109] The processing expression of the filtering transformation model is:

[0110]

[0111] Where v() represents a medical data sequence in the normalized vital signs dataset, v(k) is the kth element of the medical data sequence, Y(m,n) is the element in the mth row and nth column of the transformation matrix of the medical data sequence, h1() is the transformation function corresponding to the filtering transformation model, h1(kT1-mT1) is the value of h1() at kT1-mT1, and T1 and F1 are the time-domain transformation length and frequency-domain transformation length of the filtering transformation model, respectively.

[0112] The transformation function corresponding to the filtering transformation model can be a Gaussian function or a Gabor function;

[0113] S42, for each element of the transformation matrix of the medical data sequence, determine whether its element value is within the preset filtering interval [z1, z2]. If it is within the filtering interval, do not process the element. If it is not within the filtering interval, when the element value is less than z1, set the element value to z1, and when the element value is greater than z2, set the element value to z2.

[0114] S43, after the judgment and setting process described in S42 has been completed for each element of the transformation matrix of the medical data sequence, an updated transformation matrix is ​​obtained; the updated transformation matrix is ​​confirmed to be a filtered medical matrix.

[0115] S44, The filtered medical matrix is ​​transformed using the inverse filter transformation model to obtain the filtered data sequence of the medical data sequence;

[0116] The processing expression of the inverse filter transform model is as follows:

[0117]

[0118] Where v′() represents the filtered data sequence of the medical data sequence, v′(k) is the k-th element of the filtered data sequence of the medical data sequence, Y′(m,n) is the element in the m-th row and n-th column of the filtered medical matrix, and YM and YN are the number of rows and columns of the filtered medical matrix, respectively. The conjugate function of the transform function corresponding to the inverse filter transform model is . for The values ​​at kT1-mT1 are T1 and F1, respectively, representing the time-domain transform length and frequency-domain transform length of the inverse filter transform model.

[0119] S45, using the filtered data sequences corresponding to all medical data sequences contained in the normalized vital signs dataset, a standard vital signs dataset is constructed.

[0120] The step of performing feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial includes:

[0121] S51, using the standard vital signs dataset, a standard vital signs matrix is ​​constructed; the row vectors of the standard vital signs matrix are medical data sequences in the standard vital signs dataset; the standard vital signs matrix includes five row vectors, which are medical data sequences of five types of medical parameters contained in the standard vital signs dataset.

[0122] S52, decompose the standard vital signs matrix to obtain the left decomposition matrix, feature matrix and right decomposition matrix of the standard vital signs matrix;

[0123] The decomposition process is calculated using the following expression:

[0124] Y = UAV,

[0125] Where U is the left decomposition matrix, A is the characteristic matrix, V is the right decomposition matrix, U and V are both orthogonal matrices, and A is a diagonal matrix;

[0126] The decomposition process can be implemented using the singular value decomposition algorithm.

[0127] S53, extract the diagonal elements of the feature matrix to obtain the feature vector; the feature vector is represented as I. a I a =[λ1,λ2,…,λ N1 N1 is the number of elements contained in the feature vector;

[0128] S54, perform linear fitting on the elements and element index values ​​of the feature vector to obtain the best uniform approximation polynomial;

[0129] The linear fitting process involves using the characteristic vector element index value Ix as the known independent variable and the characteristic vector element value as the known dependent variable. The curve to be approximated is constructed using the known independent and dependent variables. The curve to be approximated is then fitted using the function approximation method to obtain the best uniform approximation polynomial f(Ix).

[0130] The curve fitting of the curve to be approximated using the function approximation method can employ the best uniform linear approximation method. The best uniform approximation polynomial f(Ix) is expressed as:

[0131] f(Ix)=α P1 (Ix) P1 +α P1-1 (Ix) P1-1 +…+α2(Ix) 2 +α1(Ix)+α0,

[0132] Where P1 is the order of the best uniform approximation polynomial f(Ix), α0, α1, α2, ..., α P1 The coefficients of the best uniform approximation polynomial f(Ix);

[0133] The deviation analysis performed on each medical data sequence of the standard vital signs dataset yields a set of deviation index values, including:

[0134] S61, For each medical data sequence in the standard vital signs dataset, calculate the average value u0 of the medical data sequence;

[0135] S62, for each element in the medical data sequence, calculate the absolute value of the difference between it and the average value u0, and determine the absolute value of the difference as the offset of the element;

[0136] S63, for the medical data sequence, find the element with the largest offset, and determine the index value of the element in the medical data sequence as the offset index value of the medical data sequence;

[0137] For example, in the medical data sequence p i In the middle, element p ij For the element with the largest offset, the index value is determined to be j.

[0138] S64, using the deviation index values ​​of all medical data sequences in the standard vital signs dataset, construct a set of deviation index values.

[0139] The calculation and discrimination process of the deviation index value set and the best consistent approximation polynomial to obtain alarm information includes:

[0140] S71, take each deviation index value in the set of deviation index values ​​as an independent variable, and use the best uniform approximation polynomial to calculate and process the independent variable to obtain the function value of the independent variable; determine the function value of the independent variable as the weight value of the deviation index value;

[0141] S72, after obtaining the weight values ​​of all deviation index values ​​in the deviation index value set, the corresponding deviation index values ​​are weighted and summed using the weight value of each deviation index value to obtain the alarm judgment value.

[0142] S73, determine whether the alarm discrimination value is greater than the alarm threshold. If it is greater than the alarm threshold, confirm that the alarm information is that the user's vital signs are abnormal; if it is less than or equal to the alarm threshold, confirm that the alarm information is that the user's vital signs are normal.

[0143] The alarm threshold can be preset or obtained by calculating the mean value based on the corresponding index value of the element closest to the mean value of each medical data sequence in the standard vital signs dataset.

[0144] The step of using the vital signs assessment module to evaluate and process the standard vital signs dataset to obtain the user's vital signs assessment values ​​includes:

[0145] S81, using the standard vital signs dataset, a standard vital signs matrix is ​​constructed; the row vectors of the standard vital signs matrix are medical data sequences in the standard vital signs dataset.

[0146] S82, calculate the cross-correlation value of the standard vital signs matrix to obtain the cross-correlation matrix;

[0147] S83, Perform feature extraction processing on the cross-correlation matrix to obtain a weighted vector;

[0148] S84, using the weighted vector, perform a weighted summation on the set of deviation index values ​​to obtain the user's vital sign assessment value.

[0149] The step of using the weighted vector to perform a weighted summation of the deviation index value set to obtain the user's vital sign assessment value includes:

[0150] Using the i-th element in the weighted vector as the weight of the i-th deviation value in the deviation value set, the deviation value set is weighted and summed to obtain the user's vital sign assessment value.

[0151] The cross-correlation calculation of the standard vital signs matrix yields a cross-correlation matrix, including:

[0152] Cross-correlation calculations are performed on the row vectors of the standard vital signs matrix to obtain cross-correlation values, and a cross-correlation matrix is ​​constructed using the cross-correlation values; the elements of the i-th row and j-th column of the cross-correlation matrix are obtained by performing cross-correlation operations on the i-th row vector and j-th row vector of the standard vital signs matrix;

[0153] The step of performing feature extraction processing on the cross-correlation matrix to obtain a weighted vector includes:

[0154] The eigenvalue sequence and corresponding eigenvector of the cross-correlation matrix are calculated; the cross-correlation matrix is ​​expressed as follows:

[0155]

[0156] Where, r ij Let λi represent the cross-correlation coefficient between the i-th row vector and the j-th row vector of the standard vital signs matrix, where i,j = 1, 2, ..., z, z = 5; and let λi represent the eigenvalue sequence, expressed as λ1, λ2, ..., λj. z , where λ i f represents the eigenvalue of the i-th row vector, and the eigenvector corresponding to the i-th row vector is f. i ;

[0157] The eigenvector is normalized and accumulated using eigenvalues ​​to obtain a normalized and accumulated eigenvector; the normalization and accumulation process includes:

[0158] Normalized summation is performed on each element of the feature vector to obtain the corresponding normalized summation value; using all the normalized summation values, a normalized summation feature vector is constructed.

[0159] The normalized cumulative calculation process is as follows:

[0160]

[0161] Among them, f ij Represents the eigenvector f i The j-th element, λ u Let l represent the eigenvalue of the u-th row vector of the standard vital signs matrix. ij Represents the i-th normalized cumulative eigenvector l i The j-th element.

[0162] Using the eigenvalue sequence, the corresponding normalized cumulative eigenvectors are weighted and summed to obtain a weighted vector; the weighted vector is represented as [y1, y2, ..., y]. 10 ], where the j-th element y of the weighted vector j The calculation expression is:

[0163]

[0164] Among them, y j It represents the j-th element of the weighted vector, and also the weight value corresponding to the j-th row vector.

[0165] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A multi-purpose medical monitoring method, characterized by, This is achieved using a multi-purpose medical monitoring device, which includes: a vital signs monitoring module, a data processing module, an alarm analysis module, and a vital signs assessment module. The vital signs monitoring module is used to collect the user's vital signs dataset and send the vital signs dataset to the data processing module; the vital signs dataset includes heart rate sequence, respiratory rate sequence, body temperature sequence, blood oxygen saturation sequence and blood pressure sequence; The vital signs monitoring module includes a heart rate sensor, a respiratory rate sensor, a body temperature sensor, a blood oxygen saturation sensor, and a blood pressure monitor; the heart rate sensor is used to acquire the user's heart rate sequence; the respiratory rate sensor is used to acquire the user's respiratory rate sequence; the body temperature sensor is used to acquire the user's body temperature sequence; the blood oxygen saturation sensor is used to acquire the user's blood oxygen saturation sequence; and the blood pressure monitor is used to acquire the user's blood pressure sequence. The data processing module is used to process the vital signs dataset to obtain a standard vital signs dataset; the data processing module is connected to the vital signs monitoring module, the alarm analysis module, and the vital signs assessment module respectively. The alarm analysis module is used to perform anomaly detection processing on the standard vital signs dataset to obtain alarm information; The vital signs assessment module is connected to the data processing module and is used to assess and process the standard vital signs dataset to obtain the user's vital signs assessment value. The alarm analysis module includes a feature fitting analysis submodule, a deviation analysis submodule, and an alarm discrimination submodule; the feature fitting analysis submodule is connected to the deviation analysis submodule, and the deviation analysis submodule is connected to the alarm discrimination submodule. The feature fitting analysis submodule is used to perform feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial. The deviation analysis submodule is used to perform deviation analysis on each medical data sequence of the standard vital signs dataset to obtain a set of deviation sequence values. The alarm discrimination submodule is used to perform calculation and discrimination processing on the set of deviation index values ​​and the best consistent approximation polynomial to obtain alarm information. The method includes: S1, using the vital signs monitoring module, collect the user's vital signs dataset and send the vital signs dataset to the data processing module; S2, using the data processing module, perform time alignment on each data sequence in the vital signs dataset to obtain a time-aligned vital signs dataset; S3, normalize the time-aligned vital signs dataset to obtain a normalized vital signs dataset; S4, the normalized vital signs dataset is filtered to obtain a standard vital signs dataset; S5, using the alarm analysis module, perform feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial; S6, perform deviation analysis on each medical data sequence of the standard vital signs dataset to obtain a set of deviation sequence values; S7, perform calculation and discrimination processing on the set of deviation index values ​​and the best consistent approximation polynomial to obtain alarm information; S8. Using the vital signs assessment module, the standard vital signs dataset is evaluated and processed to obtain the user's vital signs assessment value. The filtering process performed on the normalized vital signs dataset to obtain a standard vital signs dataset includes: S41, For each medical data sequence in the normalized vital signs dataset, a filtering transformation model is used to perform transformation processing to obtain the transformation matrix of the medical data sequence; The processing expression of the filtering transformation model is: ， wherein, denotes a certain medical data sequence of the normalized vital sign dataset, is the k-th element of the medical data sequence, is the element of the m-th row and n-th column of the transformation matrix of the medical data sequence, is a transformation function corresponding to the filter transformation model, is is the value of the element at and are the time domain transformation length and the frequency domain transformation length of the filter transformation model, respectively.​ S42, judging whether the element value of each element of the transformation matrix of the medical data sequence is in a preset filtering interval If yes, the element is not processed, and if not, when the element value is less than , the element value is set as , and when the element value is greater than , the element value is set as ; S43, after the judgment and setting process described in S42 has been completed for each element of the transformation matrix of the medical data sequence, an updated transformation matrix is ​​obtained; the updated transformation matrix is ​​confirmed to be a filtered medical matrix. S44, The filtered medical matrix is ​​transformed using the inverse filter transformation model to obtain the filtered data sequence of the medical data sequence; The processing expression of the inverse filter transform model is as follows: ， wherein, denotes a filtered data sequence of the medical data sequence, is a k-th element of the filtered data sequence of the medical data sequence, is an element of the m-th row and n-th column of the filtered medical matrix, and denote a number of rows and a number of columns of the filtered medical matrix, respectively, is a conjugate function of a transform function corresponding to the filtered inverse transform model, is a value at , and denote a time domain transform length and a frequency domain transform length of the filtered inverse transform model, respectively. S45, using the filtered data sequences corresponding to all medical data sequences contained in the normalized vital signs dataset, a standard vital signs dataset is constructed.

2. The multipurpose medical monitoring method of claim 1, wherein, The data processing module includes a time alignment submodule, a normalization processing submodule, and a filtering processing submodule; the time alignment submodule is connected to the normalization processing submodule, and the normalization processing submodule is connected to the filtering processing submodule. The time alignment submodule is used to perform time alignment on each data sequence in the vital signs dataset to obtain a time-aligned vital signs dataset. The normalization processing submodule is used to normalize the time-aligned vital signs dataset to obtain a normalized vital signs dataset. The filtering submodule is used to filter the normalized vital signs dataset to obtain a standard vital signs dataset.

3. The multipurpose medical monitoring method of claim 2, wherein, The normalization process of the time-aligned vital signs dataset to obtain a normalized vital signs dataset includes: S31, obtaining a reasonable value range of a corresponding medical parameter of each medical data sequence of the time-aligned vital sign data set denotes the i-th medical data sequence of the time-aligned vital sign data set, , denotes the j-th data of the i-th medical data sequence, , respectively represent a heart rate data sequence, a respiration rate data sequence, a body temperature data sequence, a blood oxygen saturation data sequence and a blood pressure data sequence, denotes the number of data contained in the medical data sequence of the time-aligned vital sign data set; denotes the lower limit of the reasonable value range of the corresponding medical parameter of the i-th medical data sequence denotes the upper limit of the reasonable value range of the corresponding medical parameter of the i-th medical data sequence ​​​​ S32, for each data point in each medical data sequence of the time-aligned vital signs dataset, normalization calculation is performed using the reasonable value range to obtain the normalized value of the data; The normalization calculation process is expressed as follows: ， wherein, is a normalized value obtained by performing a normalization calculation process on the value of S33, after normalization calculation processing is completed for all data of each medical data sequence in the time-aligned vital signs dataset, a normalized vital signs dataset is obtained.

4. The multipurpose medical monitoring method of claim 1, wherein, The step of performing feature fitting processing on the standard vital signs dataset to obtain the best consistent approximation polynomial includes: S51, using the standard vital signs dataset, a standard vital signs matrix is ​​constructed; the row vectors of the standard vital signs matrix are medical data sequences in the standard vital signs dataset. S52, decompose the standard vital signs matrix to obtain the left decomposition matrix, feature matrix and right decomposition matrix of the standard vital signs matrix; The decomposition process is calculated using the following expression: ， wherein, is a left decomposition matrix, is a characteristic matrix, is a right decomposition matrix, and are orthogonal matrices, is a diagonal matrix; S53, extracting diagonal elements of the feature matrix to obtain a feature vector; the feature vector is expressed as , , is the number of elements contained in the feature vector; S54, perform linear fitting on the elements and element index values ​​of the feature vector to obtain the best uniform approximation polynomial; The linear fitting process is based on the indices of the characteristic vector elements. Given the independent variable and the eigenvector elements as the dependent variable, a curve to be approximated is constructed using the known independent and dependent variables. The curve is then fitted using a function approximation method to obtain the best uniform approximation polynomial. .

5. The multipurpose medical monitoring method of claim 1, wherein, The deviation analysis performed on each medical data sequence of the standard vital signs dataset yields a set of deviation index values, including: S61, for each medical data sequence of the standard vital sign dataset, calculating a mean value of the medical data sequence ; S62, for each element in the medical data sequence, calculate its value relative to the average value. The absolute value of the difference is used to determine the offset of the element; S63, for the medical data sequence, find the element with the largest offset, and determine the index value of the element in the medical data sequence as the offset index value of the medical data sequence; S64, using the deviation index values ​​of all medical data sequences in the standard vital signs dataset, construct a set of deviation index values.

6. The multipurpose medical monitoring method of claim 1, wherein, The step of using the vital signs assessment module to evaluate and process the standard vital signs dataset to obtain the user's vital signs assessment values ​​includes: S81, using the standard vital signs dataset, a standard vital signs matrix is ​​constructed; the row vectors of the standard vital signs matrix are medical data sequences in the standard vital signs dataset. S82, calculate the cross-correlation value of the standard vital signs matrix to obtain the cross-correlation matrix; S83, Perform feature extraction processing on the cross-correlation matrix to obtain a weighted vector; S84, using the weighted vector, perform a weighted summation on the set of deviation index values ​​to obtain the user's vital sign assessment value.

7. The multipurpose medical monitoring method of claim 1, wherein, The calculation and discrimination process of the deviation index value set and the best consistent approximation polynomial to obtain alarm information includes: S71, take each deviation index value in the set of deviation index values ​​as an independent variable, and use the best uniform approximation polynomial to calculate and process the independent variable to obtain the function value of the independent variable; determine the function value of the independent variable as the weight value of the deviation index value; S72, after obtaining the weight values ​​of all deviation index values ​​in the deviation index value set, the corresponding deviation index values ​​are weighted and summed using the weight value of each deviation index value to obtain the alarm judgment value. S73, determine whether the alarm discrimination value is greater than the alarm threshold. If it is greater than the alarm threshold, confirm that the alarm information is that the user's vital signs are abnormal; if it is less than or equal to the alarm threshold, confirm that the alarm information is that the user's vital signs are normal.

Images