CN114947755A - NOX index calculation method and monitor - Google Patents
NOX index calculation method and monitor Download PDFInfo
- Publication number
- CN114947755A CN114947755A CN202210885773.7A CN202210885773A CN114947755A CN 114947755 A CN114947755 A CN 114947755A CN 202210885773 A CN202210885773 A CN 202210885773A CN 114947755 A CN114947755 A CN 114947755A
- Authority
- CN
- China
- Prior art keywords
- patient
- similarity
- patients
- category
- brain wave
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4821—Determining level or depth of anaesthesia
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
Abstract
The invention relates to the technical field of data processing, in particular to a NOX index calculation method and a monitor. The method comprises the following steps: acquiring a large amount of historical data, acquiring a brain wave sequence and a periodic sequence corresponding to each patient based on the historical data, and classifying all the patients according to the brain wave sequence and the periodic sequence corresponding to each patient to obtain a plurality of patient categories; acquiring a conventional brain wave sequence and a conventional periodic sequence of a clinical patient, acquiring a patient category of the clinical patient as a preferred category according to the conventional brain wave sequence and the conventional periodic sequence, and grouping all patients in the preferred category to obtain a plurality of groups; acquiring a group vector of each group, acquiring a preferred group of clinical patients according to the group vector, and acquiring the NOX index of the clinical patients according to the maximum anesthetic dose which can be borne by the patients in the preferred group; the subsequent auxiliary analysis to the anaesthetist is more reliable, and the problem of inaccurate control of the consumption of the anaesthesia is avoided.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a NOX index calculation method and a monitor.
Background
Anesthesia in anesthesia surgery is a reversible functional inhibition of the central and peripheral nervous systems by means of anesthetic drugs or other methods, which inhibition is characterized primarily by a loss of pain sensation, thereby resulting in surgical treatment in a pain-free state. However, any anesthesia has a certain risk, and the level of the risk is determined by various factors such as the physical condition of the patient, the type of operation, and the technical experience and conditions of the medical institution; the anesthesia operations at different parts have different risks in the anesthesia operation period. In addition, the narcotics for performing the narcotic operation are various, different narcotics have different drug effects, and some operations need to use multiple narcotics at the same time, and when multiple narcotics act on the same patient, the reaction phenomenon generated among the narcotics completely depends on the experience of an anesthesiologist, so that the risk of the narcotic operation is increased; experienced anesthesiologists can control the anesthetic and the anesthetic dosage in a spare way to achieve accurate anesthesia, but the experienced anesthesiologists are very nervous especially for high-risk patients in small and medium hospitals, and the percentage of the small and medium hospitals in China is almost 80%; thus, there is a need not only for an increased level of anesthesiologists, but also for advanced anesthesia monitoring equipment to more accurately monitor the index of depth of anesthesia.
The existing monitor mainly monitors electroencephalogram consciousness, is used for collecting the brain wave frequency of a patient in an operation and calculating the electroencephalogram consciousness, provides quantitative reference data of deep anesthesia and sedation for a clinician, and facilitates medical care personnel to judge the anesthesia state of the patient. However, the monitor only displays the electroencephalogram consciousness, different operations have different requirements on the depth of anesthesia, each patient has different tolerance on the dosage of anesthesia during the anesthesia operation, and the patient has different sensitivity to anesthesia, so that even if the patient is monitored by the existing monitor, an anesthetic pharmacist still cannot accurately control the dosage of anesthesia during the operation, and various potential hazards and complications caused by over-shallow anesthesia or over-deep anesthesia during the operation of the patient are difficult to avoid.
Disclosure of Invention
In order to solve the above technical problems, the present invention aims to provide a NOX index (Nociception index) calculation method and a monitor, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a NOX index calculation method, including the steps of:
acquiring historical data, wherein the historical data comprises brain wave curve data of a plurality of patients under different anesthetic doses;
acquiring a brain wave sequence and a periodic sequence corresponding to each patient based on the historical data, acquiring a first similarity between any two patients according to the brain wave sequence corresponding to each patient, and acquiring a second similarity between any two patients according to the periodic sequence corresponding to each patient; classifying all patients to obtain a plurality of patient categories based on the first similarity and the second similarity;
acquiring a conventional brain wave sequence and a conventional periodic sequence of a clinical patient, acquiring a first similarity between the clinical patient and each patient category according to the conventional brain wave sequence, acquiring a second similarity between the clinical patient and each patient category according to the conventional periodic sequence, acquiring the patient category of the clinical patient as a preferred category according to the first similarity and the second similarity, and grouping all patients in the preferred category to obtain a plurality of subgroups;
acquiring a group vector of each group, wherein each group vector corresponds to brain wave curve data, calculating the similarity between the brain wave curve data of the clinical patients and the group vector of each group, the corresponding group is the optimal group of the clinical patients when the similarity meets a preset condition, and obtaining the NOX index of the clinical patients according to the maximum anesthetic dose which can be borne by the patients in the optimal group.
Preferably, after the method for acquiring a large amount of historical data, the method further includes:
and acquiring the maximum anesthetic dose of each patient according to the historical data, wherein the maximum anesthetic dose is the NOX index of the corresponding patient.
Preferably, the method for acquiring a brain wave sequence and a periodic sequence corresponding to each patient based on the historical data includes:
each patient corresponds to one brain wave curve data under each anesthetic dose, the brain wave curve data corresponding to all anesthetic doses within the maximum anesthetic dose that the patient can bear are spliced to obtain complete brain wave curve data, and numerical values in the complete brain wave curve data are sequentially arranged to obtain a brain wave sequence of the patient;
the method comprises the steps of obtaining a dose electric wave sequence corresponding to each patient under each anesthetic dose, calculating the similarity between the dose electric wave sequences under two adjacent anesthetic doses, obtaining the similarity between every two adjacent anesthetic doses in the maximum anesthetic dose which can be borne by each patient, and sequentially arranging all the similarities to obtain a periodic sequence of the patient.
Preferably, the method for acquiring a first similarity between any two patients according to the brain wave sequence corresponding to each patient and acquiring a second similarity between any two patients according to the periodic sequence corresponding to each patient comprises:
taking the brain wave sequence corresponding to each patient as a vector, and calculating cosine similarity between vectors corresponding to any two brain wave sequences of the patients, wherein the cosine similarity is a first similarity between the two patients;
if the lengths of the brain wave sequences of the two patients are not consistent, the shorter brain wave sequence is used as a truncation point to truncate the longer brain wave sequence, so that the lengths of the two brain wave sequences are consistent;
taking the periodic sequence corresponding to each patient as a vector, and calculating cosine similarity between vectors corresponding to any two periodic sequences of the patients, wherein the cosine similarity is a second similarity between the two patients;
and if the lengths of the periodic sequences of the two patients are not consistent, cutting the length of the shorter periodic sequence to the length of the longer periodic sequence so as to enable the lengths of the periodic sequences corresponding to the two patients to be consistent.
Preferably, the method for classifying all patients into a plurality of patient categories based on the first similarity and the second similarity comprises:
calculating first similarity and second similarity between all patients, and if the first similarity and the second similarity between two patients are greater than a preset threshold, dividing the two patients into the same patient category, thereby obtaining a plurality of categories; the patient with the first similarity or the second similarity smaller than a preset threshold is a patient to be classified;
acquiring the category center patient of each category and the brain wave sequence and the periodic sequence corresponding to the category center patient, calculating a first similarity between the brain wave sequence of the patient to be classified and the brain wave sequence of the category center patient, and a second similarity between the periodic sequence of the patient to be classified and the periodic sequence of the category center patient, and obtaining a comprehensive similarity according to the first similarity and the second similarity, wherein the category of the corresponding category center patient is the category of the patient to be classified when the comprehensive similarity is maximum;
and obtaining the belonged categories of all the patients to be classified, and updating all the categories to obtain a plurality of patient categories.
Preferably, the method for acquiring a class-centered patient of each of the classes comprises:
acquiring a first similarity and a second similarity between any patient and other patients in each category, and taking the first similarity and the second similarity as two dimensions respectively;
and calculating the sum of the first similarity and the sum of the second similarity, respectively arranging the sums of the first similarity and the second similarity corresponding to all the patients in the category in a descending order, and arranging the most advanced patient in two dimensions as the category center patient of the category.
Preferably, the method for obtaining the patient category of the clinical patient as a preferred category according to the first similarity and the second similarity, and grouping all patients in the preferred category to obtain a plurality of subgroups includes:
adding the first similarity and the second similarity to obtain a similarity sum, and obtaining the similarity sum of the clinical patient and each patient category, wherein the patient category corresponding to the maximum similarity sum is the preferred category of the clinical patient;
obtaining a NOX index of each patient in the preferred category, dividing the patients with the same NOX index into one group, and dividing the patients in the preferred category into a plurality of subgroups.
Preferably, the method for obtaining a group vector of each group includes:
acquiring a common factor matrix, an average weight coefficient vector and an average independent factor vector of each group, wherein the average weight coefficient vector is the average value of all weight coefficient vectors in the group, and the average independent factor vector is the average value of all independent factor vectors in the group; performing factor analysis according to the common factor matrix, the average weight coefficient vector and the average independent factor vector to obtain a group vector; each group vector corresponds to one brain wave curve data.
Preferably, the method for calculating the similarity between the brain wave curve data of the clinical patient and the group vector of each subgroup, wherein the corresponding subgroup is the preferred group of the clinical patient when the similarity satisfies a preset condition, further includes:
each group corresponds to a maximum anesthetic dose, similarity calculation is carried out on the anesthetic dose injected by the clinical patient and the corresponding group, and the corresponding group is the group with the maximum anesthetic dose identical to the anesthetic dose injected by the clinical patient;
the preset condition is that a turning point of similarity occurs, and a subgroup before the turning point of the similarity is the preferred group;
the method for judging the turning point of the similarity comprises the following steps: if the similarity between the clinical patient of the current anesthetic dose and the corresponding subgroup is less than the similarity between the clinical patient of the previous anesthetic dose and the corresponding subgroup, the similarity between the clinical patient of the current anesthetic dose and the corresponding subgroup is a similarity turning point.
In a second aspect, another embodiment of the present invention provides a monitor, including: a memory, a processor and a computer program stored in the memory and executable on the processor, the processor when executing the computer program implementing the steps of a NOX index calculation method as described above.
The invention has the following beneficial effects: according to the embodiment of the invention, a large amount of historical data is acquired for analysis, and the patients in the historical data are classified to obtain a plurality of patient categories, so that reliable data dependence is provided for the analysis of subsequent clinical patients; then, matching the clinical patients with each patient category based on brain wave curve data and periodically changing characteristics to obtain the preferred categories of the clinical patients, wherein the accuracy is high; furthermore, each patient in the optimal category is divided into a plurality of groups, similarity calculation is carried out on the clinical patient and each group to obtain a corresponding optimal group, so that the NOx index of the clinical patient is obtained based on the NOx index of the optimal group, the analysis on the NOx index of the clinical patient is more accurate, the follow-up auxiliary analysis on an anaesthetist is more reliable, and the problem of inaccurate control on the consumption of anaesthesia is avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a method for calculating a NOx index according to an embodiment of the invention.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description of the NOX index calculating method and monitor according to the present invention with reference to the accompanying drawings and preferred embodiments will be made below. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The method is suitable for auxiliary judgment of the anesthesia depth, and aims to solve the problem that an existing anesthesia physician is difficult to accurately control the anesthesia dosage.
The specific scheme of the NOX index calculation method and the monitor provided by the present invention is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flow chart of a NOX index calculation method according to an embodiment of the present invention is shown, the method including the steps of:
step S100, acquiring historical data, wherein the historical data comprises brain wave curve data of a plurality of patients under different anesthetic doses.
The embodiment of the invention aims to be used as an anesthesia reference of a doctor together with an electroencephalogram consciousness index through the assistance of the NOX index, and the NOX index is used for representing the probability of a patient reacting to harmful stimuli, so that the NOX index needs to be analyzed by utilizing a large amount of historical data, and reliable NOX data are obtained and used as the anesthesia reference.
Firstly, a large amount of historical data is obtained, the historical data comprises brain wave curve data of different patients under different anesthetic doses, and a large amount of clinical experience shows that different people have different constitutions and therefore have different responses to anesthetics, excessive anesthetic dose can cause great damage to human bodies, some people can go into deep sleep when the anesthetic dose is small, and some people still keep a shallow sleep state when the anesthetic dose is large, so that the probability of responses to harmful stimuli is high.
And then acquiring the NOx index corresponding to each patient, namely the maximum anesthetic dose which can be borne by each patient, and marking and retaining the NOx index corresponding to each patient, wherein the NOx index corresponding to each patient needs to be determined by a doctor according to the performance of the patient under different anesthetic doses.
Step S200, acquiring a brain wave sequence and a periodic sequence corresponding to each patient based on historical data, acquiring a first similarity between any two patients according to the brain wave sequence corresponding to each patient, and acquiring a second similarity between any two patients according to the periodic sequence corresponding to each patient; classifying all patients based on the first similarity and the second similarity to obtain a plurality of patient categories.
Specifically, since the NOX index corresponding to each patient is displayed in real time as a reference basis for a doctor, and each patient is an individual who cannot be repeatedly tested, when acquiring the NOX index of an actual clinical patient, it is necessary to refer to the NOX index corresponding to each patient in a large amount of historical data, and the clinical patient has the greatest similarity with which patient in the historical data, and the corresponding NOX indexes are consistent.
Considering that each patient in the historical data corresponds to one brain wave curve data when being subjected to external harmful stimulation under different anesthetic doses, the patients with similar brain wave curve data can be considered as the same type of patients to be analyzed, and therefore all the patients in the historical data can be preliminarily classified according to the brain wave curve data.
The specific method for classifying all patients to obtain a plurality of patient categories is as follows:
firstly, acquiring a brain wave sequence of each patient, and calculating first similarity between the brain wave sequences of different patients; the brain wave sequence corresponding to each patient is obtained by combining and splicing the brain wave sequences corresponding to each anesthetic dosage within the maximum anesthetic dosage that the patient can bear, namely the brain wave curve data corresponding to all anesthetic dosages within the maximum anesthetic dosage that the patient can bear are spliced to obtain complete brain wave curve data, and the numerical values in the complete brain wave curve data are sequentially arranged to obtain the brain wave sequences; because the maximum anesthetic dosages corresponding to different patients are different, and therefore the lengths of the brain wave sequences of different patients are different, in the embodiment of the invention, the brain wave sequences with smaller lengths are used as the truncation points, the brain wave sequences with the same length between two patients are obtained according to the truncation points, and then the first similarity calculation is carried out on the two brain wave sequences with the same length.
As an example, assuming that the first similarity between the patient B and the patient C needs to be calculated, the electroencephalogram sequence corresponding to the patient B and the electroencephalogram sequence corresponding to the patient C are acquired, if the maximum anesthetic dose corresponding to the patient B is 5 and the maximum anesthetic dose corresponding to the patient C is 3, the electroencephalogram sequences corresponding to the patient B and the patient C are different in length, the electroencephalogram sequence corresponding to the patient B under the anesthetic dose 3 is acquired through the truncation point with the shorter electroencephalogram sequence corresponding to the patient C as the truncation point, and the first similarity is calculated from the electroencephalogram sequence corresponding to the anesthetic dose 3 under the patient B and the electroencephalogram sequence corresponding to the anesthetic dose 3 under the patient C, in an embodiment of the present invention, the method of calculating the first similarity is: the brain wave sequences corresponding to two patients are regarded as two vectors, and the cosine similarity between the two vectors is the first similarity of the two patients; by analogy, the first similarity between the brain wave sequences of any two patients in the historical data is obtained.
Then, acquiring a periodic sequence corresponding to each patient, and calculating a second similarity between the periodic sequences of different patients; because each patient corresponds to different brain wave curve data under different anesthetic doses, when the anesthetic dose is 1, the corresponding brain wave curve data is recorded as a 1-dose electric wave sequence; when the anesthetic dose is 2, recording the corresponding brain wave curve data as a 2-dose electric wave sequence; calculating the similarity between the dose electric wave sequences under different anesthetic doses, and recording the similarity between the 1 dose electric wave sequence of the anesthetic dose 1 and the 2 dose electric wave sequence of the anesthetic dose 2 as s (12); the similarity between the 4-dose electrical wave sequence of the anesthetic dose 4 and the 5-dose electrical wave sequence of the anesthetic dose 5 is s (45), and by analogy, the similarity between the dose electrical wave sequences of two adjacent anesthetic doses in the anesthetic dose which can be borne by the patient is calculated, all the similarities sequentially form a periodic sequence corresponding to the patient, and if the maximum anesthetic dose corresponding to the patient B is 5, the periodic sequence is { s (12), s (23), s (34), s (45) } is obtained according to the plurality of dose electrical wave sequences of the patient B; the similarity in the embodiment of the invention is calculated by cosine similarity.
The periodic sequence corresponding to each patient in the historical data is obtained, and the maximum anesthetic dose corresponding to different patients is different, so that the length of the periodic sequence corresponding to each patient is different. The method for clipping the longer periodic sequence by the shorter periodic sequence comprises the following steps: assuming that the maximum anesthetic dose for patient B is 5 and the maximum anesthetic dose for patient C is 3, the periodic sequence length for patient B is 4 and the periodic sequence length for patient C is 2; when the second similarity calculation is performed on the periodic sequences of the patient B and the patient C, the periodic sequence of the patient B is cut by the periodic sequence corresponding to the patient C, namely the periodic sequence corresponding to the patient B is cut into the periodic sequence with the length of 2, and the second similarity between the cut periodic sequence of the patient B and the periodic sequence of the patient C is calculated.
Finally, classifying the patient categories according to the first similarity and the second similarity between every two patients; calculating first similarity and second similarity between all patients, and if the first similarity and the second similarity between two patients are greater than a preset threshold, dividing the two patients into the same patient category, thereby obtaining a plurality of categories; the patient with the first similarity or the second similarity smaller than a preset threshold is a patient to be classified; acquiring the category center patient of each category and the brain wave sequence and the periodic sequence corresponding to the category center patient, calculating a first similarity between the brain wave sequence of the patient to be classified and the brain wave sequence of the category center patient, and a second similarity between the periodic sequence of the patient to be classified and the periodic sequence of the category center patient, and obtaining a comprehensive similarity according to the first similarity and the second similarity, wherein the category of the corresponding category center patient is the category of the patient to be classified when the comprehensive similarity is maximum; obtaining the belonged categories of all patients to be classified, and updating the patients in all categories to obtain a plurality of patient categories.
Specifically, the patients with the first similarity and the second similarity both greater than the preset threshold are classified into the same patient category, so that a plurality of categories can be obtained, and the first similarity and the second similarity between the patients in each category are both greater than the preset threshold.
Preferably, in the embodiment of the present invention, the preset threshold is set to 0.9, that is, when the first similarity and the second similarity between each two patients are greater than 0.9, the two patients are classified into the same category.
It should be noted that, because there are a large number of patients in the historical data, there may be a case where both the first similarity and the second similarity between some patients and any patient are less than 0.9, that is, there are patients that have not been successfully classified, and the patients that have not been successfully classified are marked as patients to be classified, and the category of the patients to be classified is identified again.
The method for identifying the category of the patient to be classified specifically comprises the following steps:
firstly, acquiring a classified central patient in each category successfully, wherein the acquisition method of the classified central patient in each category comprises the following steps: acquiring a first similarity and a second similarity between each patient in the category and other patients in the category, taking the first similarity as one dimension and the second similarity as the other dimension, and calculating the sum of the similarities of each patient in each dimension, namely calculating the sum of the first similarities and the sum of the second similarities between each patient in the category and other patients; and respectively sorting the sum of the first similarity and the sum of the second similarity corresponding to each patient in a descending order, wherein the patient with the highest total ranking in two dimensions is the category central patient of the category.
As an example, assume that patient a, patient b, and patient c are present in the category, where the first similarity of patient a to patient b is calculated to be 0.92, and the second similarity of patient a to patient b is calculated to be 0.93; the first similarity of patient a to patient c is 0.93, and the second similarity of patient a to patient c is 0.94; the first similarity of patient b to patient c is 0.92, and the second similarity of patient b to patient c is 0.95; the sum of the corresponding first similarities for patient a is therefore: 0.92+0.93=1.85, and the sum of the corresponding second similarities for patient a is: 0.93+0.94= 1.87; the sum of the corresponding first similarities for patient b is: 0.92+0.92=1.84, and the sum of the corresponding second similarities for patient b is: 0.93+0.95= 1.88; the sum of the corresponding first similarities for patient c is: 0.93+0.92=1.85, and the sum of the corresponding second similarities for patient c is: 0.94+0.95= 1.89; then the rank according to the sum of the first similarity in the dimension of the first similarity is respectively: patient a is first in parallel with patient c, and patient b is second; the dimension of the second similarity is sorted according to the sum of the second similarities, and the sorting is respectively as follows: patient c, patient b, patient a; thus the top of the total ordering in both dimensions is patient c, then the center patient in the category is patient c.
Further, calculating the similarity between each patient to be classified and the class center patient corresponding to each class; acquiring a first similarity and a second similarity between each patient to be classified and each category center patient, calculating a comprehensive similarity according to the first similarity and the second similarity, and selecting the category center patient with the highest comprehensive similarity with the patient to be classified, wherein the category to which the category center patient belongs is the patient category of the patient to be classified; the calculation formula of the comprehensive similarity is as follows:
wherein the content of the first and second substances,
representing a first similarity;
representing a second degree of similarity;
representing the integrated similarity.
By analogy, the classification of all patients in the historical data is completed by calculating the comprehensive similarity between each patient to be classified and each central patient in each category and identifying the category of the patient to be classified, so that a plurality of patient categories are obtained.
It should be noted that, when the above calculation is to classify each patient to be classified, the center patient of each category is updated in real time, that is, each time one patient to be classified is added to each category, the center patient of the category in the category is recalculated once, and subsequent calculation is performed on the center patient of the category updated in real time.
Step S300, acquiring a conventional brain wave sequence and a conventional periodic sequence of a clinical patient, acquiring a first similarity between the clinical patient and each patient category according to the conventional brain wave sequence, acquiring a second similarity between the clinical patient and each patient category according to the conventional periodic sequence, acquiring the patient category of the clinical patient as a preferred category according to the first similarity and the second similarity, and grouping all patients in the preferred category to obtain a plurality of groups.
In step S200, all patients are classified into a plurality of patient categories, and a category-centered patient of each patient category is obtained according to a method of obtaining the same category-centered patient of each category, and the brain wave sequence and the periodic sequence corresponding to the category-centered patient in each patient category are used as the category representative of the patient category, and then, in the NOX index analysis of real-time clinical patients, the matching analysis is performed on the clinical patients based on the category representative of each patient category.
Specifically, when clinical patients are anesthetized, the anesthetic dose is a common anesthetic dose that most people can bear, and the common anesthetic dose is obtained by analyzing a large amount of historical data; the method comprises the steps of obtaining a brain wave sequence of a clinical patient according to all brain wave curve data of the clinical patient when the clinical patient is subjected to external harmful stimulation under a common anesthetic dose, further calculating to obtain a periodic sequence corresponding to the clinical patient, then calculating a first similarity of a category in different patient categories representing the brain wave sequence of the clinical patient and a second similarity of the periodic sequence of the clinical patient, adding the first similarity and the second similarity to obtain a similarity sum, wherein the maximum similarity and the corresponding category represent that the patient category to which the clinical patient belongs is the patient category of the clinical patient, accordingly obtaining the patient category to which the clinical patient belongs, and marking the patient category to which the clinical patient belongs as a preferred category.
Since the NOX index varies from patient to patient within the same patient category, which is the maximum anesthetic dose that each patient can tolerate, the NOX index for that clinical patient is not available solely on the basis of the preferred category, which requires analysis of each patient in the preferred category. To improve the efficiency of clinical patient analysis, all patients in the preferred category are grouped, and the grouping in the present example is done by grouping patients with the same NOX index into groups, thereby resulting in multiple subgroups in the preferred category.
Step S400, acquiring a group vector of each group, wherein each group vector corresponds to brain wave curve data, calculating the similarity between the brain wave curve data of clinical patients and the group vector of each group, the corresponding group is a preferred group of clinical patients when the similarity meets a preset condition, and obtaining the NOX index of the clinical patients according to the maximum anesthetic dose which can be borne by the patients in the preferred group.
The preferred categories are divided into a plurality of subgroups in step S300, and since the sleep degree becomes deeper as the anesthetic dose becomes larger, the periodicity of the brain wave curve data becomes stronger while the sleep is deep, and thus the brain wave curve data becomes less and less changed when the external harmful stimulus is applied; the reason why the periodic variations of different patients are similar is that the brain wave curve data of different patients have common factors, the factors refer to different characteristics of the human body, for example, the factor x represents constitutional characteristics, the factor y represents narcotic resistance characteristics, and the factor z represents brain wave activity degree characteristics, and when the factors of different patients are similar, the brain wave curve data of the patients are similar, so that the patients are classified into the same patient category.
But the corresponding weight vectors differ between different common factors, leading to different actual maximum anesthetic doses between patients; for example: the common factors between the patient A and the patient B are a factor x and a factor y, and the periodic change of the brain wave curve data between the patient A and the patient B is similar due to the factor x and the factor y, but the weights of the common factors of the patient A and the patient B are different, the weights of the factor x and the factor y in the patient A are [0.7,0.3], the weights of the factor x and the factor y in the patient B are [0.2,0.8], and therefore, the maximum anesthetic dose between the patient A and the patient B is different.
In the embodiment of the invention, a factor analysis method is adopted to calculate the weight vector of the common factors of each patient, and the mathematical formula of the factor analysis is as follows:
wherein the content of the first and second substances,
representing a vector;
representing a weight coefficient;
representing a common factor matrix;
representing independent factor vectors.
The vector X in the embodiment of the invention is brain wave curve data corresponding to the patients in each patient category, and the common factor matrix is a common factor among the extracted patients in the patient category and represents a common characteristic among the brain wave curve data of the patients in the patient category; the brain wave curve data of the patients in each patient category is analyzed, the brain wave curve data of each patient is regarded as a column vector data, and the column vector data is the vector X, so that a common factor matrix of all the column vector data in the patient categories can be obtained through the formula of the factor analysis.
Based on the factor analysis of the patients in each subgroup, the common factors of each subgroup and the independent factor vectors of each patient in the subgroup are obtained, so that the brain wave curve data of the clinical patients are matched with each subgroup, and the specific method is as follows:
(1) because the factor analysis method is to extract common features among data by analyzing a plurality of data, and for any patient, the corresponding brain wave curve data does not meet the conditions of the factor analysis method, calculating an average weight coefficient vector and an average independent factor vector in each group, wherein the average weight coefficient vector is the average value of all weight coefficient vectors in the group, and the average independent factor vector is the average value of all independent factor vectors in the group; according to a formula of factor analysis, a corresponding vector X can be calculated based on a common factor matrix, an average weight coefficient vector and an average independent factor vector, the obtained vector X is marked as a group vector of the group, and by analogy, a group vector corresponding to each group is obtained, and each group vector is actually brain wave curve data.
(2) Acquiring similarity between brain wave curve data of clinical patients and a group vector corresponding to each group, wherein the maximum anesthetic dose corresponding to the patients in each group is the same because the NOX index in each group is the same, calculating the similarity between the anesthetic dose of the actual clinical patient and the corresponding group until a similarity turning point appears, wherein the group corresponding to the anesthetic dose before the similarity turning point is a preferred group of the clinical patient, and the NOX index of the patients in the preferred group is the NOX index of the clinical patient.
As an example, assume that an analysis is performed on a clinical patient W, when the clinical patient W is injected with a general anesthetic dose, assume that the general anesthetic dose is anesthetic dose 3; a total of 4 subgroups are a1, a2, a3, a 4; wherein, the maximum anesthetic dosage corresponding to each group is anesthetic dosage 3, anesthetic dosage 5, anesthetic dosage 4 and anesthetic dosage 6; therefore, the similarity between the electroencephalogram curve data corresponding to the clinical patient W and the group vector corresponding to the a1 subgroup is calculated to be 0.7 at this time; the injection of this clinical patient W was then continued to be changed to anesthetic dose 4, at which time the similarity between the clinical patient W and the a3 panel was calculated to be 0.9; and then continuing to inject the clinical patient W to change the injection into anesthetic dosage 5, calculating the similarity between the clinical patient W and the a2 subgroup to be 0.6, wherein the similarity turning point appears at the moment, namely the similarity is changed from big to small, the subgroup of the anesthetic dosage 4 before the similarity turning point is the preferred group of the clinical patient W, and the anesthetic dosage 4 of the preferred group is the maximum anesthetic dosage of the clinical patient W, so that the NOx index of the patient is obtained.
It should be noted that, in the embodiment of the present invention, the method for calculating the similarity is obtained by using the cosine similarity, and the calculation of the cosine similarity is known in the prior art and is not described again.
Furthermore, an anaesthetist can be judged in an auxiliary mode according to the NOX index of a clinical patient, and the anaesthesia depth of the clinical patient in an anaesthesia operation is analyzed together with the electroencephalogram consciousness index, so that the anaesthetist is helped to control the anaesthesia dosage in the operation process.
In summary, in the embodiment of the present invention, a large amount of historical data is obtained, and analysis is performed based on brain wave curve data of patients under different anesthetic doses in the historical data, so that all patients in the historical data are divided into a plurality of patient categories, and a category-center patient is selected from each patient category; then obtaining the conventional anesthetic dosage based on historical data, injecting actual clinical patients by using the conventional anesthetic dosage to obtain corresponding brain wave curve data, matching the brain wave curve data of the clinical patients with all patient categories to obtain the preferred categories of the clinical patients, further analyzing the patients in the preferred categories, dividing all the patients in the preferred categories into a plurality of groups according to the corresponding NOX index of each patient in the preferred categories, calculating the similarity by adjusting the anesthetic dosage of the clinical patients and each group, obtaining the preferred group of the clinical patients according to the turning point of the similarity, wherein the NOX index in the preferred group is the NOX index of the clinical patients, performing auxiliary judgment according to the NOX index of the clinical patients under the corresponding maximum anesthetic dosage, analyzing the anesthetic depth by combining the NOX index with the electroencephalogram consciousness index, the accuracy of analysis is improved, so that the anaesthetist can more accurately control the anaesthesia dosage.
Based on the same inventive concept as the method embodiment, the embodiment of the present invention further provides a monitor, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor, when executing the computer program, implements the steps of one of the above-described NOX index calculation method embodiments, such as the steps shown in fig. 1. The NOX index calculation method has been described in detail in the above embodiments, and is not described again.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that are within the spirit of the present invention are intended to be included therein.
Claims (10)
1. A NOX index calculation method, characterized by comprising the steps of:
acquiring historical data, wherein the historical data comprises brain wave curve data of a plurality of patients under different anesthetic doses;
acquiring a brain wave sequence and a periodic sequence corresponding to each patient based on the historical data, acquiring a first similarity between any two patients according to the brain wave sequence corresponding to each patient, and acquiring a second similarity between any two patients according to the periodic sequence corresponding to each patient; classifying all patients to obtain a plurality of patient categories based on the first similarity and the second similarity;
acquiring a conventional brain wave sequence and a conventional periodic sequence of a clinical patient, acquiring a first similarity between the clinical patient and each patient category according to the conventional brain wave sequence, acquiring a second similarity between the clinical patient and each patient category according to the conventional periodic sequence, acquiring the patient category of the clinical patient as a preferred category according to the first similarity and the second similarity, and grouping all patients in the preferred category to obtain a plurality of subgroups;
acquiring a group vector of each group, wherein each group vector corresponds to brain wave curve data, calculating the similarity between the brain wave curve data of the clinical patients and the group vector of each group, the corresponding group is the optimal group of the clinical patients when the similarity meets a preset condition, and obtaining the NOX index of the clinical patients according to the maximum anesthetic dose which can be borne by the patients in the optimal group.
2. The NOX index calculation method of claim 1, wherein the method of obtaining a large amount of historical data is followed by further comprising:
and acquiring the maximum anesthetic dose of each patient according to the historical data, wherein the maximum anesthetic dose is the NOX index of the corresponding patient.
3. The NOX index calculation method according to claim 1, wherein the method of obtaining the brain wave sequence and the periodic sequence corresponding to each patient based on the historical data includes:
each patient corresponds to one brain wave curve data under each anesthetic dose, the brain wave curve data corresponding to all anesthetic doses within the maximum anesthetic dose that the patient can bear are spliced to obtain complete brain wave curve data, and numerical values in the complete brain wave curve data are sequentially arranged to obtain a brain wave sequence of the patient;
acquiring a dose radio wave sequence corresponding to each patient under each anesthetic dose, calculating the similarity between the dose radio wave sequences under two adjacent anesthetic doses, acquiring the similarity of each two adjacent anesthetic doses in the maximum anesthetic dose which can be borne by each patient, and sequentially arranging all the similarities to obtain a periodic sequence of the patient.
4. The NOX index calculation method according to claim 1, wherein the method of obtaining a first similarity between any two patients from the brain wave sequence corresponding to each patient and obtaining a second similarity between any two patients from the periodic sequence corresponding to each patient includes:
taking the brain wave sequence corresponding to each patient as a vector, and calculating cosine similarity between vectors corresponding to any two brain wave sequences of the patients, wherein the cosine similarity is a first similarity between the two patients;
if the lengths of the brain wave sequences of the two patients are not consistent, the shorter brain wave sequence is used as a truncation point to truncate the longer brain wave sequence, so that the lengths of the two brain wave sequences are consistent;
taking the periodic sequence corresponding to each patient as a vector, and calculating cosine similarity between vectors corresponding to any two periodic sequences of the patients, wherein the cosine similarity is a second similarity between the two patients;
and if the lengths of the periodic sequences of the two patients are not consistent, cutting the length of the shorter periodic sequence to the length of the longer periodic sequence so as to enable the lengths of the periodic sequences corresponding to the two patients to be consistent.
5. The NOX index calculation method of claim 1, wherein the method of classifying all patients into a plurality of patient categories based on the first and second similarities comprises:
calculating first similarity and second similarity between all patients, and if the first similarity and the second similarity between two patients are greater than a preset threshold, dividing the two patients into the same patient category, thereby obtaining a plurality of categories; the patient with the first similarity or the second similarity smaller than a preset threshold is a patient to be classified;
acquiring the category center patient of each category and the brain wave sequence and the periodic sequence corresponding to the category center patient, calculating a first similarity between the brain wave sequence of the patient to be classified and the brain wave sequence of the category center patient, and a second similarity between the periodic sequence of the patient to be classified and the periodic sequence of the category center patient, and obtaining a comprehensive similarity according to the first similarity and the second similarity, wherein the category of the corresponding category center patient is the category of the patient to be classified when the comprehensive similarity is maximum;
and obtaining the belonged categories of all the patients to be classified, and updating all the categories to obtain a plurality of patient categories.
6. The NOX index calculation method of claim 5 wherein the method of obtaining a category-centered patient for each of the categories comprises:
acquiring a first similarity and a second similarity between any patient and other patients in each category, and taking the first similarity and the second similarity as two dimensions respectively;
and calculating the sum of the first similarity and the sum of the second similarity, respectively arranging the sums of the first similarity and the second similarity corresponding to all the patients in the category in a descending order, and arranging the most advanced patient in two dimensions as the category center patient of the category.
7. The NOX index calculation method of claim 2, wherein the method of obtaining the patient category of the clinical patient as a preferred category according to the first similarity and the second similarity, and grouping all patients in the preferred category into a plurality of subgroups comprises:
adding the first similarity and the second similarity to obtain a similarity sum, and obtaining the similarity sum of the clinical patient and each patient category, wherein the patient category corresponding to the maximum similarity sum is the preferred category of the clinical patient;
obtaining a NOX index of each patient in the preferred category, dividing the patients with the same NOX index into one group, and dividing the patients in the preferred category into a plurality of subgroups.
8. The NOX index calculation method of claim 1, wherein the method of obtaining a group vector for each subgroup comprises:
acquiring a common factor matrix, an average weight coefficient vector and an average independent factor vector of each group, wherein the average weight coefficient vector is the average value of all weight coefficient vectors in the group, and the average independent factor vector is the average value of all independent factor vectors in the group; and performing factor analysis according to the common factor matrix, the average weight coefficient vector and the average independent factor vector to obtain a group vector.
9. The NOX index calculation method according to claim 1, wherein the method of calculating the similarity between the brain wave curve data of the clinical patient and the group vector of each subgroup whose corresponding subgroup is a preferred group of the clinical patient when the similarity satisfies a preset condition further includes:
each group corresponds to a maximum anesthetic dose, similarity calculation is carried out on the anesthetic dose injected by the clinical patient and the corresponding group, and the corresponding group is the group with the maximum anesthetic dose identical to the anesthetic dose injected by the clinical patient;
the preset condition is that a turning point of similarity occurs, and a subgroup before the turning point of the similarity is the preferred group;
the method for judging the turning point of the similarity comprises the following steps: if the similarity between the clinical patient of the current anesthetic dose and the corresponding subgroup is less than the similarity between the clinical patient of the previous anesthetic dose and the corresponding subgroup, the similarity between the clinical patient of the current anesthetic dose and the corresponding subgroup is a similarity turning point.
10. A monitor comprising a memory, a processor and a computer program stored in said memory and executable on said processor, wherein said processor when executing said computer program performs the steps of a method of NOX index calculation as claimed in any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210885773.7A CN114947755B (en) | 2022-07-26 | 2022-07-26 | NOX index calculation method and monitor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210885773.7A CN114947755B (en) | 2022-07-26 | 2022-07-26 | NOX index calculation method and monitor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114947755A true CN114947755A (en) | 2022-08-30 |
| CN114947755B CN114947755B (en) | 2022-11-08 |
Family
ID=82969104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210885773.7A Active CN114947755B (en) | 2022-07-26 | 2022-07-26 | NOX index calculation method and monitor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114947755B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116434975A (en) * | 2023-06-14 | 2023-07-14 | 西安交通大学医学院第一附属医院 | Breathe internal medicine device intelligent management system that clears away lung-heat and discharges phlegm |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150164412A1 (en) * | 2013-12-16 | 2015-06-18 | General Electric Company | Point of care measurement of patient sensitivity to anesthetics |
| US20180085055A1 (en) * | 2016-09-27 | 2018-03-29 | Boston Scientific Neuromodulation Corporation | Method and apparatus for pain management using objective pain measure |
| US20190069841A1 (en) * | 2017-09-05 | 2019-03-07 | Korea University Research And Business Foundation | Method and apparatus of monitoring anaesthesia and consciousness depth through brain network analysis |
| CN109662691A (en) * | 2018-12-24 | 2019-04-23 | 深圳市美格尔医疗设备股份有限公司 | A kind of patient monitor of band NOX index |
| CN111883244A (en) * | 2020-07-27 | 2020-11-03 | 深圳市美格尔医疗设备股份有限公司 | Method and device for constructing display interface of trend information of anesthesia monitor |
| BR112020013317A2 (en) * | 2018-03-22 | 2020-12-01 | Universidad De La Sabana | process to classify anesthetic depth in interventions with total intravenous anesthesia |
| CN112006658A (en) * | 2020-08-21 | 2020-12-01 | 思澜科技（成都）有限公司 | Anesthesia state monitoring method and device |
| CN112587154A (en) * | 2020-12-29 | 2021-04-02 | 深圳美格尔生物医疗集团有限公司 | Electroencephalogram dual-spectrum monitoring signal acquisition device and system |
| WO2021110879A1 (en) * | 2019-12-06 | 2021-06-10 | École Normale Superieure Paris-Saclay | System and method for predicting depth of anesthesia |
| CN113729730A (en) * | 2021-09-03 | 2021-12-03 | 河南省安信科技发展有限公司 | Anesthesia state monitoring system based on electroencephalogram analysis technology |
| CN113886407A (en) * | 2021-09-29 | 2022-01-04 | 中国人民解放军中部战区总医院 | Intelligent updating method, device, system and medium for anesthetic medication database |
| CN114389437A (en) * | 2021-12-30 | 2022-04-22 | 深圳市美格尔医疗设备股份有限公司 | Control circuit and brain electricity dual-frequency index simulator |
| US20220240834A1 (en) * | 2019-10-24 | 2022-08-04 | The Trustees Of Columbia University In The City Of New York | System, method, and computer-accessible medium for visualization and analysis of electroencephalogram oscillations in the alpha band |
-
2022
- 2022-07-26 CN CN202210885773.7A patent/CN114947755B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150164412A1 (en) * | 2013-12-16 | 2015-06-18 | General Electric Company | Point of care measurement of patient sensitivity to anesthetics |
| US20180085055A1 (en) * | 2016-09-27 | 2018-03-29 | Boston Scientific Neuromodulation Corporation | Method and apparatus for pain management using objective pain measure |
| US20190069841A1 (en) * | 2017-09-05 | 2019-03-07 | Korea University Research And Business Foundation | Method and apparatus of monitoring anaesthesia and consciousness depth through brain network analysis |
| BR112020013317A2 (en) * | 2018-03-22 | 2020-12-01 | Universidad De La Sabana | process to classify anesthetic depth in interventions with total intravenous anesthesia |
| CN109662691A (en) * | 2018-12-24 | 2019-04-23 | 深圳市美格尔医疗设备股份有限公司 | A kind of patient monitor of band NOX index |
| US20220240834A1 (en) * | 2019-10-24 | 2022-08-04 | The Trustees Of Columbia University In The City Of New York | System, method, and computer-accessible medium for visualization and analysis of electroencephalogram oscillations in the alpha band |
| WO2021110879A1 (en) * | 2019-12-06 | 2021-06-10 | École Normale Superieure Paris-Saclay | System and method for predicting depth of anesthesia |
| CN111883244A (en) * | 2020-07-27 | 2020-11-03 | 深圳市美格尔医疗设备股份有限公司 | Method and device for constructing display interface of trend information of anesthesia monitor |
| CN112006658A (en) * | 2020-08-21 | 2020-12-01 | 思澜科技（成都）有限公司 | Anesthesia state monitoring method and device |
| CN112587154A (en) * | 2020-12-29 | 2021-04-02 | 深圳美格尔生物医疗集团有限公司 | Electroencephalogram dual-spectrum monitoring signal acquisition device and system |
| CN113729730A (en) * | 2021-09-03 | 2021-12-03 | 河南省安信科技发展有限公司 | Anesthesia state monitoring system based on electroencephalogram analysis technology |
| CN113886407A (en) * | 2021-09-29 | 2022-01-04 | 中国人民解放军中部战区总医院 | Intelligent updating method, device, system and medium for anesthetic medication database |
| CN114389437A (en) * | 2021-12-30 | 2022-04-22 | 深圳市美格尔医疗设备股份有限公司 | Control circuit and brain electricity dual-frequency index simulator |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116434975A (en) * | 2023-06-14 | 2023-07-14 | 西安交通大学医学院第一附属医院 | Breathe internal medicine device intelligent management system that clears away lung-heat and discharges phlegm |
| CN116434975B (en) * | 2023-06-14 | 2023-09-01 | 西安交通大学医学院第一附属医院 | Breathe internal medicine device intelligent management system that clears away lung-heat and discharges phlegm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114947755B (en) | 2022-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Leguia et al. | Seizure cycles in focal epilepsy | |
| EP1414343B1 (en) | Method for predicting outcome of treatments | |
| US8364254B2 (en) | Method and device for probabilistic objective assessment of brain function | |
| Katz | Fractals and the analysis of waveforms | |
| EP1253853B1 (en) | Method for classifying physiologic brain imbalances using quantitative eeg | |
| Yang et al. | Selection of features for patient-independent detection of seizure events using scalp EEG signals | |
| US20110087125A1 (en) | System and method for pain monitoring at the point-of-care | |
| Yan et al. | Multi-modality of polysomnography signals’ fusion for automatic sleep scoring | |
| AU2002329590A1 (en) | Electroencephalography based systems and methods for selecting therapies and predicting outcomes | |
| AU2001236794A1 (en) | Method for classifying and treating physiologic brain imbalances using quantitative EEG | |
| Anheim et al. | Improvement in Parkinson disease by subthalamic nucleus stimulation based on electrode placement: effects of reimplantation | |
| Li et al. | Efficacy of and risk factors for percutaneous balloon compression for trigeminal neuralgia in elderly patients | |
| CN114947755B (en) | NOX index calculation method and monitor | |
| Robu et al. | Medical data mining with extended WEKA | |
| Haring et al. | Assessment of intraoperative nerve monitoring parameters associated with facial nerve outcome in parotidectomy for benign disease | |
| Wong et al. | Computationally efficient epileptic seizure prediction based on extremely randomised trees | |
| CN114356911B (en) | Data missing processing method and system based on set division information quantity maximization | |
| Zeng et al. | Evaluation of relational and NoSQL approaches for patient cohort identification from heterogeneous data sources | |
| Foroozandeh et al. | Effectiveness of Mindfulness-based Cognitive Therapy on Resilience and Difficulties in Emotion Regulation among Divorced Women in Tehran | |
| CN114255887A (en) | Information display method and device, electronic equipment and storage medium | |
| CN116884605B (en) | Intelligent management system for operation anesthesia information | |
| Delgado | A Data-Driven Intervention Framework for Improving Adherence to Growth Hormone Therapy Based on Clustering Analysis and Traffic Light Alerting Systems | |
| Subramanian et al. | Behavioral and autonomic dynamics during propofol-induced unconsciousness | |
| Sandhu et al. | Machine learning for cognitive treatment planning in patients with neurodisorder and trauma injuries | |
| Garcia et al. | Using Cartesian Coordinate Systems to Create, Classify, and Retrieve Biomedical Time-Series: Applications to 24-hour Ambulatory Blood Pressure Monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |