CN115995282B - Expiratory flow data processing system based on knowledge graph - Google Patents
Expiratory flow data processing system based on knowledge graph Download PDFInfo
- Publication number
- CN115995282B CN115995282B CN202310287242.2A CN202310287242A CN115995282B CN 115995282 B CN115995282 B CN 115995282B CN 202310287242 A CN202310287242 A CN 202310287242A CN 115995282 B CN115995282 B CN 115995282B
- Authority
- CN
- China
- Prior art keywords
- expiration
- expiratory
- data
- short
- time
- 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.)
- Active
Links
- 238000012545 processing Methods 0.000 title claims abstract description 19
- 230000036461 convulsion Effects 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims description 25
- 230000001052 transient effect Effects 0.000 claims description 14
- 208000000059 Dyspnea Diseases 0.000 claims description 3
- 206010013975 Dyspnoeas Diseases 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 208000013220 shortness of breath Diseases 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 8
- 230000008859 change Effects 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 9
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 230000035790 physiological processes and functions Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 208000012661 Dyskinesia Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Image Processing (AREA)
Abstract
The invention relates to the technical field of data processing, and provides an expiratory flow data processing system based on a knowledge graph, which comprises the following steps: collecting age data and expiration flow data of a user to obtain an original expiration data sequence; acquiring a plurality of short-time data sequences of expiration, acquiring an expiration jerk coefficient of each short period, acquiring the expiration flow rate of each moment according to the difference of the expiration flow data of adjacent moments, and acquiring the expiration jerk coefficient of each short period according to the expiration flow rate and the expiration jerk coefficient; acquiring an expiration characteristic sequence according to the expiration drastic coefficient of each short period, converting to obtain an expiration characteristic image, clustering to acquire a first category and a second category, and obtaining an expiration characteristic index; and forming a triplet according to the age data of the user, the maximum peak value in the expiratory flow data and the expiratory characteristic index, constructing a knowledge graph and analyzing the expiratory state of the user. The present invention aims to improve the accuracy of the expiratory state analysis by the time correlation of expiratory flow data.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an expiratory flow data processing system based on a knowledge graph.
Background
The expiratory flow data is time series data, and the current expiratory state of the user can be evaluated through calculation and processing of the expiratory flow data; there are two general methods for expiratory flow data processing, one is a traditional large medical instrument, which is expensive and has poor flexibility, and meanwhile, related professional technicians are required to operate the instrument to obtain the expiratory state of the user; the other is a portable expiratory flow data acquisition instrument, but the instrument can only acquire the current expiratory flow data of a user and cannot effectively analyze the data; there is therefore a need for a system that can collect user's expiratory flow data and quickly process to get the user's expiratory state.
Referring to patent CN108186018B, curve fitting is performed on the expiratory flow data of the current user by using fourier transform, and because the distribution characteristics of expiratory flow data points in different states have strong randomness characteristics, the fitted curve is easy to deviate from the original data points, so that deviation occurs in judgment of the expiratory state of the current user; the expiratory flow data is changed according to time, and the characteristic of time correlation between the expiratory flow data is ignored in the scheme, so that the expiratory flow data processing reliability is poor.
Disclosure of Invention
The invention provides an expiratory flow data processing system based on a knowledge graph, which aims to solve the problem of poor expiratory state analysis accuracy caused by strong expiratory flow data randomness, and adopts the following technical scheme:
one embodiment of the present invention provides an expiratory flow data processing system based on a knowledge-graph, the system comprising:
the expiration flow data acquisition module acquires age data and expiration flow data of a user, and obtains an original expiration data sequence according to the expiration flow data;
an expiratory flow data processing module: dividing an original expiration data sequence through preset short time periods to obtain a plurality of expiration short time data sequences, acquiring expiration jerk coefficients of each short time period according to expiration flow data in the expiration short time data sequences, acquiring expiration flow velocity of each time according to difference of expiration flow data of adjacent time points, and acquiring expiration jerk coefficients of each short time period according to the expiration flow velocity and the expiration jerk coefficients;
arranging the expiratory rapid coefficients of all short periods according to a time sequence relationship to obtain an expiratory feature sequence, acquiring an expiratory feature image according to the expiratory feature sequence, acquiring the sizes of windows according to the expiratory short-time data sequence, the original expiratory data sequence and the expiratory feature image, dividing the expiratory feature image through the windows to obtain a plurality of expiratory feature windows, acquiring clustering reference distances among the pixels according to coordinate information and gray values of the pixels in each expiratory feature window, and clustering the pixels in each expiratory feature window according to the clustering reference distances to obtain a first category and a second category in each expiratory feature window;
the sum of the pixel point numbers contained in all first types in all expiration feature windows is recorded as a first pixel quantity, the sum of the pixel point numbers contained in all second types in all expiration feature windows is recorded as a second pixel quantity, and the ratio of the first pixel quantity to the second pixel quantity is used as an expiration feature index of a user;
and the knowledge graph construction and retrieval module is used for forming an expiration state triplet according to the age data of the user, the maximum peak value in the expiration flow data and the expiration characteristic index, constructing a knowledge graph and retrieving and analyzing the expiration state of the user according to the knowledge graph.
Optionally, the obtaining the shortness of breath coefficient of each short period includes the following specific methods:
wherein ,
represent the first
A short period of the expiratory jerk coefficient,
represent the first
The number of data points in the short-period expiratory short-time data sequence where a jump occurs,
representing the number of elements in the short-time data sequence of expiration,
represent the first
The maximum value in the short-time data sequence of expiration for a short period of time,
represent the first
Minimum in the short-period expiratory short-time data sequence.
Optionally, the method for obtaining the expiratory flow rate at each moment according to the difference of the expiratory flow data at adjacent moments includes the following specific steps:
wherein ,
representing the first in the original expiration data sequence
The expiratory flow rate at each moment in time,
represent the first
Expiratory flow data at each moment in time,
represent the first
Expiratory flow data at each moment in time,
representing the time interval between adjacent moments in the original expiratory data sequence.
Optionally, the method for obtaining the expiratory transient coefficient of each short period according to the expiratory flow rate and the expiratory transient coefficient includes the following specific steps:
wherein ,
represent the first
A short period of the expiratory transient coefficient,
represent the first
A short period of the expiratory jerk coefficient,
representing expirationThe number of elements in the short-time data sequence,
represent the first
In the short time period
The expiratory flow rate at each moment in time,
represent the first
In the short time period
Expiratory flow rate at each moment.
Optionally, the method for acquiring the window size according to the short-time data sequence of expiration, the original expiration data sequence and the expiration feature image includes the following specific steps:
wherein ,
indicating the size of one edge of the window,
representing the number of elements in the short-time data sequence of expiration,
representing the number of elements in the original sequence of exhalation data,
the size of the image representing the characteristics of the exhalation,
representing a rounding down.
Optionally, the method for obtaining the clustering reference distance between the pixels according to the coordinate information and the gray value of the pixels in each exhalation feature window includes the following specific steps:
wherein ,
representing any of the exhalation signature windows
Dots and dots
The clustering of points refers to the distance of the points,
representation of
The coordinate information of the point(s),
representation of
The coordinate information of the point(s),
representation of
Dots and dots
The euclidean distance of the points,
representation of
The gray value of the dot is used,
representation of
Gray value of the dot.
The beneficial effects of the invention are as follows: according to the method, the original expiration data sequence of the user is divided into short time periods, and the expiration rapid change coefficient of each short time period is obtained through the expiration flow data change in the short time period and the expiration flow rate at each moment in a quantization mode, so that the expiration rapid change coefficient can approximately reflect the expiration state of each short time period; an expiration characteristic sequence is formed through expiration drastic coefficients, and an expiration characteristic image is obtained through conversion of a gram angle field, so that the expiration state judgment influence caused by time correlation among expiration flow data at different moments is more considered; through window segmentation and clustering of the expiration feature images, influence of discrete data point distribution on a clustering result is avoided, quantification of a severe expiration change part and a relatively stable expiration part obtained through clustering is more accurate, accuracy and reliability of expiration feature indexes are further improved, and a result of expiration state retrieval analysis of a user based on a knowledge graph is more accurate.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a block diagram of an expiratory flow data processing system based on a knowledge graph according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a block diagram of an expiratory flow data processing system based on a knowledge graph according to an embodiment of the present invention is shown, the system includes:
and the expiration flow data acquisition module S101 acquires age data and expiration flow data of the user and obtains an original expiration data sequence.
The purpose of this embodiment is to process the expiratory flow data of the user, and then analyze and get the expiratory health status of the user, so this embodiment collects the expiratory flow data of the user first; the expiratory flow data of the user is collected by the sensor, and since the normal expiratory state can be maintained for a few seconds, in order to obtain the expiratory flow data of the user, the embodiment uses the sensor to sample once at intervals of 10ms, and the expiratory state is assumed to be continuous
Second, the total user expiration flow data acquired according to the expiration state at the moment is
I.e. sharing
Individual expiratory flow data; meanwhile, in order to quantify the expiratory health state of the user more accurately, the embodiment needs to collect the age data of the user, so as to avoid inaccurate expiratory health state judgment of the user caused by the difference of expiratory states of different age groups.
Further, since the normal exhalation state is maintained for several seconds, it is required to ensure that the total amount of the collected exhalation flow data is sufficiently large, and in this embodiment, the total amount of the exhalation flow data is removed and sampled again to obtain a plurality of pieces of exhalation flow data of the user, and the exhalation flow data is formed into a data sequence according to the collection time sequence and recorded as the original exhalation data sequence of the user.
Thus, the original expiration data sequence and the age data of the user are obtained.
Expiratory flow data processing module S102:
(1) And acquiring a plurality of short-time data sequences of expiration according to the original expiration data sequence, acquiring the expiration jerk coefficient of each short period according to the short-time data sequences of expiration, acquiring the expiration flow rate of each moment according to the difference of the expiration flow data of adjacent moments, and acquiring the expiration jerk coefficient of each short period according to the expiration flow rate and the expiration jerk coefficient.
It should be noted that, because the difference of the expiratory flow data of the user in the healthy state in a short time during expiration is small, that is, the expiratory flow data is relatively stable, the expiratory flow data of the user in a short time can be used to quantify whether the user exhales relatively rapidly in the short time, that is, the expiratory rapid coefficient of each moment of the user.
Specifically, in this embodiment, the original expiratory data sequence is divided by taking a preset short time period of 50ms, that is, from the first expiratory flow data, every 5 expiratory flow data is divided into an expiratory short time data sequence; it should be noted that, if the original expiration data sequence does not finally satisfy the 5 expiration flow data to form an expiration short-time data sequence, the embodiment adopts a quadratic linear interpolation method to make the element number of the last expiration short-time data sequence sufficient; since each short-term data sequence corresponds to a short period of time, then for the first
Short period of expiration jerk coefficient
The specific calculation method of (a) is as follows:
wherein ,
represent the first
The number of data points with jump in the short-period short-time expiration data sequence is obtained by performing bayesian variable point detection on the short-time expiration data sequence, and the bayesian variable point detection is a known technology and is not repeated in the embodiment;
representing the number of elements in the short-time data sequence of expiration,
represent the first
The maximum value in the short-time data sequence of expiration for a short period of time,
represent the first
A minimum value in the short-time data sequence of expiration of a short period of time; quantifying the shortness of breath coefficient by the ratio of the number of jump points to the number of elements in the short-time data sequence of expiration and the difference of the maximum and minimum values in the short-time data sequence of expiration; the larger the ratio is, the more data points are in jump in the short-time data sequence of expiration, the more unstable the expiration flow data is in a short period, and the more rapid the expiration state is in the short period; the greater the difference in maximum and minimum values, the greater the degree of fluctuation in the expiratory flow data over the short period of time, the more rapid the expiratory state may be; and acquiring the expiration jerk coefficient of each short period in the original expiration data sequence according to the method.
Further, since the value of the expiratory flow data only represents the expiratory flow at each time, and does not have the expiratory flow change that reflects each time in common with the adjacent time, it is necessary to obtain the expiratory flow rate at each time from the expiratory flow data at the adjacent time.
Specifically, for the first in the original expiration data sequence
At various moments, its expiratory flow rate
The specific calculation method of (a) is as follows:
wherein ,
represent the first
Expiratory flow data at each moment in time,
represent the first
Expiratory flow data at each moment in time,
representing the time interval between adjacent moments in the original expiratory data sequence, the sampling interval in this embodiment is 10ms, then
The method comprises the steps of carrying out a first treatment on the surface of the Obtaining the expiratory flow rate of each moment through the numerical value difference of the expiratory flow data at adjacent moments, wherein the larger the difference is, the larger the expiratory flow data at the moment is changed, and the more rapid the expiratory flow is possible; and the smaller the difference, the smaller the change of the expiratory flow data at the moment, the smoother the expiration.
Further, according to the expiration jerk coefficient of each short period and the expiration flow rate of each moment in the short period, acquiring the expiration jerk coefficient of each short period to obtain the following point
For example, the expiratory transient coefficient is obtained for a short period of time
The specific calculation method of (a) is as follows:
wherein ,
represent the first
A short period of the expiratory jerk coefficient,
representing the number of elements in the short-time data sequence of expiration,
represent the first
In the short time period
The expiratory flow rate at each moment in time,
represent the first
In the short time period
Expiratory flow rate at each moment; obtaining the expiration rapid coefficient of the period through quantification by the expiration rapid coefficient and the difference of the expiration flow rates at adjacent moments in the expiration short-time data sequence, wherein the larger the expiration rapid coefficient is, the more unstable the expiration state of a user in a short period is, and the larger the corresponding expiration rapid coefficient is; the larger the difference of the expiratory flow velocity at the adjacent moment, the quicker the expiratory flow data in a short period changes, the larger the expiratory flow data fluctuation degree and the fluctuation frequency are, the more drastic changes of the expiratory flow data expression are, and the larger the expiratory drastic coefficient is; obtaining the expiratory transient coefficients of each short period according to the method; it should be noted that, for the last one in the original expiration data sequenceThe method comprises the steps of calculating the expiratory flow data by adopting quadratic linear interpolation when calculating the expiratory flow velocity of the expiratory flow data if the expiratory short-time data sequence to which the expiratory flow data belongs does not utilize quadratic linear interpolation filling sequence when being acquired; if the sequence of short-time data of the expiration, to which the short-time data of the expiration belongs, is filled by secondary linear interpolation when the short-time data of the expiration is acquired, only the data in the original expiration data sequence is calculated when the expiration jerkiness coefficient, the expiration flow rate and the expiration rapid change coefficient are acquired, and the data filled by the secondary linear interpolation does not participate in calculation.
Thus, the expiratory transient coefficient of each short period is obtained and used for reflecting the severe change degree of the expiratory flow data in each short period, and the expiratory state in each short period can be represented.
(2) And obtaining an expiration characteristic sequence according to the expiration drastic coefficient of each short period, converting to obtain an expiration characteristic image, obtaining a first category and a second category through clustering according to the expiration characteristic image, and further obtaining an expiration characteristic index of the original expiration data sequence.
After obtaining the expiratory transient coefficients of each short period, arranging the expiratory transient coefficients according to a time sequence relationship to obtain an expiratory feature sequence, and converting by using a gram angle field to obtain an expiratory feature image, wherein each pixel point in the expiratory feature image expresses the correlation relationship between the expiratory transient coefficients at different moments in the expiratory feature sequence, and the larger the gray value of the pixel point is, the larger the correlation degree of the expiratory transient coefficients calculated at corresponding two moments is; when a user generates a relatively rapid expiration abnormality in a state at a certain moment, expiration data corresponding to a plurality of moments before and after the moment also has certain abnormal change; compared with a one-dimensional expiration feature sequence, the expiration feature image obtained through conversion can better reflect the change relation caused by the time relevance of expiration flow data at different moments.
Specifically, the expiratory hypervariability coefficients of all short periods are arranged according to a time sequence relationship to obtain an expiratory feature sequence, and the expiratory feature sequence is converted into an expiratory feature image by using a gram angle field; the specific calculation process of the gram angle field is a known technology, and the embodiment is not repeated; according to the characteristics of the conversion calculation of the gram angle field, the obtained expiration characteristic image has the same size in the horizontal direction and the vertical direction, namely the expiration characteristic image is a square image; at this time, the gray values of different pixels in the exhalation characteristic image and the exhalation rapid change coefficient show a certain relation, if the exhalation data in a certain short period of time shows rapid abnormality, namely the exhalation rapid change coefficient is larger, at the position of the corresponding pixel obtained after the calculation of the gram angle field, the gray value of the pixel is larger, and the gray value of the pixel shows a certain difference with the data in the normal exhalation state; the expiratory feature sequence is converted into the expiratory feature image through the gram angle field, so that expiratory drastic coefficients of different periods of a user can be combined with expiratory drastic coefficients of adjacent moments to be analyzed, time relevance among expiratory flow data is considered more, and the respiratory state analysis result of the user is more accurate.
Further, because the expiratory flow data at different moments has the characteristic of stronger randomness, fluctuation variation in the calculated expiratory feature sequence has the characteristic of uncertainty, so that the distribution state of pixels in the obtained expiratory feature image has the characteristic of certain uncertainty and the distribution of data points is discrete; therefore, different types of pixel points in the exhalation characteristic image can be analyzed and calculated through a clustering algorithm, so that short time periods with similar exhalation drastic change coefficients can be gathered into one type to quantify the exhalation state; meanwhile, in order to reduce the influence of data point distribution dispersion on a clustering result, the embodiment divides the exhalation feature image by setting a window for clustering analysis.
Specifically, first use
Segmentation of the exhalation signature image by a window of size, wherein
，
Representing the number of elements in the short-time data sequence of expiration,
representing the number of elements in the original sequence of exhalation data,
the size of the expiration feature image, i.e. the number of pixels comprised by one edge of the expiration feature image,
representing a downward rounding; acquiring the window size through the ratio of the element number of the expiration short-time data sequence in the original expiration data sequence; dividing the exhalation characteristic image according to the acquired window size to acquire a plurality of exhalation characteristic windows; when the exhalation feature image is divided by the window, the window with insufficient pixels is filled by the secondary linear interpolation.
Further, clustering each exhalation characteristic window, clustering pixel points in the exhalation characteristic window by a K-means method, setting the K value to be 2, randomly selecting the pixel points in the exhalation characteristic window as initial clustering centers, and clustering the reference distances among the pixel points
The specific calculation method of (a) is as follows:
wherein ,
representing any of the exhalation signature windows
Dots and dots
The clustering of points refers to the distance of the points,
representation of
The coordinate information of the point(s),
representation of
The coordinate information of the point(s),
representation of
Dots and dots
The euclidean distance of the points,
representation of
The gray value of the dot is used,
representation of
Gray value of the dot; the Euclidean distance between pixel point coordinates in the expiration feature window and the gray value distance are used for quantifying the clustering reference distance for clustering, and the smaller the Euclidean distance of the pixel points is, the closer the pixel points are in the window, the greater the probability that the expiration states are similar is, and the more likely the clustering reference distances are in a class; the smaller the gray value distance is, the influence of the discrete distribution of the data points in the exhalation characteristic window on the clustering result is avoided, so that the pixel points with similar gray values and far Euclidean distance can be clustered into one type, and the accuracy and the reliability of the clustering result are improved.
Further, according to the K value and the clustering reference distanceK-means clustering is carried out on each expiration feature window from the initial clustering center, and each expiration feature window is respectively provided with two categories; respectively calculating the gray average value of pixel points in any two classes in the expiration feature window, marking the class with the largest gray average value as the first class in the expiration feature window, and marking the class with the smallest gray average value as the second class in the expiration feature window; obtaining a first category and a second category in each expiration feature window according to the method, marking the sum of the pixel numbers contained in all the first categories in all the expiration feature windows as a first pixel quantity, marking the sum of the pixel numbers contained in all the second categories in all the expiration feature windows as a second pixel quantity, taking the ratio of the first pixel quantity to the second pixel quantity as an expiration feature index of an original expiration data sequence, namely, the expiration feature index of a user, and expressing the ratio as
The method comprises the steps of carrying out a first treatment on the surface of the At this time, since the short period of time in which the expiratory transient coefficient is large corresponds to the pixel point in the expiratory feature image, the gray value thereof is large, so the first category reflects the portion in which the expiratory state is drastically changed, the second category reflects the portion in which the expiratory state is relatively stable, and the expiratory state of the user is quantified by the ratio of the sum of the two.
The method comprises the steps of obtaining the expiration characteristic index of a user, reflecting the expiration state in expiration flow data acquired by the user, optimizing a clustering algorithm through window segmentation and gray value distance, reducing the influence of data point discrete distribution on a clustering result, and simultaneously considering the influence of time relevance among expiration flow data on expiration state judgment.
And the knowledge graph construction and retrieval module S103 is used for forming an expiration state triplet according to the age data of the user, the maximum peak value in the expiration flow data and the expiration characteristic index, constructing a knowledge graph and retrieving and analyzing the expiration state of the user according to the knowledge graph.
It should be noted that, the module S102 has already analyzed the original expiration data sequence of the user to obtain the expiration characteristic index of the user, and by combining the age data of the user and the maximum peak value of expiration flow data, the current health state can be evaluated and judged for the user in each physiological state of each age group, where the age data is used in each age group, the maximum peak value of expiration flow data is used to determine the physiological state of the user, for example, the greater the maximum peak value of expiration flow data of the user with large vital capacity will be; the three indexes form a triplet, and a knowledge graph is constructed for a large number of people in a hospital, so that the expiration state of the user can be analyzed according to the knowledge graph.
Specifically, the maximum peak in the original sequence of exhalation data, i.e., the maximum in all of the exhalation flow data, is extracted as
According to age data of the user
Index of exhalation characteristics
Maximum peak of expiratory flow data
Forming an expiration state triplet for characterizing the current expiration state of the user, the expiration state triplet being expressed as
The method comprises the steps of carrying out a first treatment on the surface of the Collecting expiration flow data and age data of a large number of people in a hospital, representing expiration state triplets of each person by the method, constructing a knowledge graph according to the expiration state triplets of the large number of people in the hospital, and taking the knowledge graph as data in a standard library; the knowledge graph is constructed as the prior art, and this embodiment is not described in detail.
Further, searching the user's expiration state triples in the constructed knowledge graph, and outputting the searched user's expiration state triples to obtain the current expiration state of the user.
The processing and the expiration state analysis of the expiration flow data of the user are completed through the expiration state triplets of the user and the constructed knowledge graph, and the expiration state analysis of the user is more accurate by combining the age data and the expiration flow data maximum peak value, so that the expiration state analysis of the user is more accurately combined with the age stage and the physiological state of the user.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (2)
1. An expiratory flow data processing system based on a knowledge graph, the system comprising:
the expiration flow data acquisition module acquires age data and expiration flow data of a user, and obtains an original expiration data sequence according to the expiration flow data;
an expiratory flow data processing module: dividing an original expiration data sequence through preset short time periods to obtain a plurality of expiration short time data sequences, acquiring expiration jerk coefficients of each short time period according to expiration flow data in the expiration short time data sequences, acquiring expiration flow velocity of each time according to difference of expiration flow data of adjacent time points, and acquiring expiration jerk coefficients of each short time period according to the expiration flow velocity and the expiration jerk coefficients;
arranging the expiratory rapid coefficients of all short periods according to a time sequence relationship to obtain an expiratory feature sequence, acquiring an expiratory feature image according to the expiratory feature sequence, acquiring the sizes of windows according to the expiratory short-time data sequence, the original expiratory data sequence and the expiratory feature image, dividing the expiratory feature image through the windows to obtain a plurality of expiratory feature windows, acquiring clustering reference distances among the pixels according to coordinate information and gray values of the pixels in each expiratory feature window, and clustering the pixels in each expiratory feature window according to the clustering reference distances to obtain a first category and a second category in each expiratory feature window;
the sum of the pixel point numbers contained in all first types in all expiration feature windows is recorded as a first pixel quantity, the sum of the pixel point numbers contained in all second types in all expiration feature windows is recorded as a second pixel quantity, and the ratio of the first pixel quantity to the second pixel quantity is used as an expiration feature index of a user;
the knowledge graph construction and retrieval module is used for forming an expiration state triplet according to the age data of the user, the maximum peak value in the expiration flow data and the expiration characteristic index, constructing a knowledge graph and retrieving and analyzing the expiration state of the user according to the knowledge graph;
the method for acquiring the shortness of breath coefficient of each short period comprises the following specific steps:
wherein ,
indicate->
A short period of expiratory jerk coefficient, +.>
Indicate->
The number of data points in the short-period expiration short-time data sequence at which a jump occurs, < >>
Representing the number of elements in the short-term data sequence of expiration, < >>
Indicate->
Maximum value in the short-term data sequence of expiration for a short period,/-for a short period of time>
Indicate->
A minimum value in the short-time data sequence of expiration of a short period of time;
the method for acquiring the expiratory flow rate at each moment according to the difference of the expiratory flow data at the adjacent moment comprises the following specific steps:
wherein ,
representing the +.f in the original expiration data sequence>
Expiratory flow at various moments,/>
Indicate->
Expiratory flow data at each instant +.>
Indicate->
Expiratory flow data at each instant +.>
Representing time intervals of adjacent moments in the original expiratory data sequence;
the method for acquiring the expiratory transient coefficients of each short period according to the expiratory flow rate and the expiratory transient coefficients comprises the following specific steps:
wherein ,
indicate->
Expiratory transient coefficient for a short period, +.>
Indicate->
A short period of the expiratory jerk coefficient,
representing the number of elements in the short-term data sequence of expiration, < >>
Indicate->
First->
Expiratory flow at various moments,/>
Indicate->
First->
Expiratory flow rate at each moment;
the method for acquiring the clustering reference distance between the pixel points according to the coordinate information and the gray value of the pixel points in each expiration feature window comprises the following specific steps:
wherein ,
representing arbitrary +.>
Point and->
Clustering reference distance of points +_>
Representation->
Coordinate information of point,/->
Representation->
Coordinate information of point,/->
Representation->
Point and->
European distance of point, ++>
Representation->
The gray value of the dot is used,
representation->
Gray value of the dot.
2. The knowledge-based expiratory flow data processing system according to claim 1, wherein the method for obtaining the window size according to the expiratory short-time data sequence, the original expiratory data sequence and the expiratory feature image comprises the following specific steps:
wherein ,
represents the size of one edge of the window, +.>
Representing the number of elements in the short-term data sequence of expiration, < >>
Representing the number of elements in the original expiratory data sequence, +.>
Size of the characteristic image representing expiration, +.>
Representing a rounding down. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310287242.2A CN115995282B (en) | 2023-03-23 | 2023-03-23 | Expiratory flow data processing system based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310287242.2A CN115995282B (en) | 2023-03-23 | 2023-03-23 | Expiratory flow data processing system based on knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115995282A CN115995282A (en) | 2023-04-21 |
| CN115995282B true CN115995282B (en) | 2023-06-02 |
Family
ID=85995315
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310287242.2A Active CN115995282B (en) | 2023-03-23 | 2023-03-23 | Expiratory flow data processing system based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115995282B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111529847A (en) * | 2020-05-07 | 2020-08-14 | 山东凯鑫宏业生物科技有限公司 | Breathing machine system with breathing phase identification control applied to intelligent medical system and control method |
| CN111657950A (en) * | 2020-06-12 | 2020-09-15 | 山东凯鑫宏业生物科技有限公司 | Breathing machine system applied to intelligent medical treatment and capable of predicting breathing state of patient and control method thereof |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103169476B (en) * | 2013-03-14 | 2015-03-25 | 中山大学 | Method and device for identification and early warning of respiration wave form image |
| EP2989979B8 (en) * | 2014-08-27 | 2022-02-16 | Wellell Inc. | Respiratory waveform recognition method and system |
| CN108113678A (en) * | 2017-12-04 | 2018-06-05 | 广州和普乐健康科技有限公司 | A kind of respiratory air flow detection method |
| CN110731778B (en) * | 2019-07-22 | 2022-04-29 | 华南师范大学 | Method and system for recognizing breathing sound signal based on visualization |
| CN111134682A (en) * | 2020-01-16 | 2020-05-12 | 广州昊熹医疗科技有限公司 | Intelligent respiratory disease state recording and detecting method and device based on peak current meter |
| CN113591701A (en) * | 2021-07-30 | 2021-11-02 | 深圳市商汤科技有限公司 | Respiration detection area determination method and device, storage medium and electronic equipment |
| CN113506626B (en) * | 2021-09-03 | 2021-12-10 | 南通嘉蒂体育用品有限公司 | Sleep characteristic data evaluation processing method and system based on wearable device |
| CN114224316A (en) * | 2021-11-16 | 2022-03-25 | 中科广润(中山)科技有限公司 | Lung function parameter calculation method based on turbo type lung function instrument |
| CN114236525A (en) * | 2021-12-27 | 2022-03-25 | 浙江大学 | Multi-target tracking and respiration detection method and device based on millimeter wave radar |
| CN114422267B (en) * | 2022-03-03 | 2024-02-06 | 北京天融信网络安全技术有限公司 | Flow detection method, device, equipment and medium |
| CN115040109B (en) * | 2022-06-20 | 2024-03-22 | 徐州工程学院 | Breathing pattern classification method and system |
-
2023
- 2023-03-23 CN CN202310287242.2A patent/CN115995282B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111529847A (en) * | 2020-05-07 | 2020-08-14 | 山东凯鑫宏业生物科技有限公司 | Breathing machine system with breathing phase identification control applied to intelligent medical system and control method |
| CN111657950A (en) * | 2020-06-12 | 2020-09-15 | 山东凯鑫宏业生物科技有限公司 | Breathing machine system applied to intelligent medical treatment and capable of predicting breathing state of patient and control method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115995282A (en) | 2023-04-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101779800B1 (en) | System and method for evaluating multifaceted growth based on machine learning | |
| CN111493828B (en) | Sequence-to-sequence sleep disorder detection method based on full convolution network | |
| US20040199482A1 (en) | Systems and methods for automatic and incremental learning of patient states from biomedical signals | |
| CN108959004B (en) | Disk failure prediction method, device, equipment and computer readable storage medium | |
| CN109009102B (en) | Electroencephalogram deep learning-based auxiliary diagnosis method and system | |
| CN105595990A (en) | Intelligent terminal device for evaluating and distinguishing quality of electrocardiosignal | |
| CN116418882B (en) | Memory data compression method based on HPLC dual-mode carrier communication | |
| CN111370124A (en) | Health analysis system and method based on facial recognition and big data | |
| CN108847284B (en) | Human body biological age measuring and calculating device and system | |
| CN105574322A (en) | Physiological parameter index operation system and method | |
| CN117809124B (en) | Medical image association calling method and system based on multi-feature fusion | |
| CN114224343B (en) | Cognitive disorder detection method, device, equipment and storage medium | |
| CN117473351B (en) | Power supply information remote transmission system based on Internet of things | |
| CN113724878B (en) | Medical risk information pushing method and device based on machine learning | |
| CN114550932A (en) | Sleep apnea risk assessment method, device, equipment and storage medium | |
| CN115995282B (en) | Expiratory flow data processing system based on knowledge graph | |
| CN111481193B (en) | Fall risk assessment and early warning method and system | |
| US20240054360A1 (en) | Similar patients identification method and system based on patient representation image | |
| CN107194918B (en) | Data analysis method and device | |
| KR101345216B1 (en) | Mehtod and apparatus for analyzing multi-subject electroencephalograph | |
| CN110222459B (en) | Individual height analysis method, system, readable storage medium and terminal | |
| CN116186561A (en) | Running gesture recognition and correction method and system based on high-dimensional time sequence diagram network | |
| CN110647870B (en) | Method for calculating approximate entropy of resting state fMRI data based on sliding window | |
| CN113903470A (en) | Intelligent life reminding method and system for patient after hemodialysis | |
| CN109907730B (en) | Data processing method and device, storage medium and computer equipment |
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 |