CN114693715A - AI-based cross section change rate acquisition method for thoracic cavity simulator - Google Patents
AI-based cross section change rate acquisition method for thoracic cavity simulator Download PDFInfo
- Publication number
- CN114693715A CN114693715A CN202210419454.7A CN202210419454A CN114693715A CN 114693715 A CN114693715 A CN 114693715A CN 202210419454 A CN202210419454 A CN 202210419454A CN 114693715 A CN114693715 A CN 114693715A
- Authority
- CN
- China
- Prior art keywords
- cross
- simulator
- change rate
- endpoint
- section
- 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
- 230000008859 change Effects 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 33
- 210000000115 thoracic cavity Anatomy 0.000 title claims abstract description 13
- 210000000038 chest Anatomy 0.000 claims abstract description 77
- 238000012360 testing method Methods 0.000 claims abstract description 43
- 238000012417 linear regression Methods 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000011218 segmentation Effects 0.000 claims description 9
- 238000004040 coloring Methods 0.000 claims description 7
- 238000006073 displacement reaction Methods 0.000 claims description 5
- 230000010354 integration Effects 0.000 claims description 2
- 238000002680 cardiopulmonary resuscitation Methods 0.000 abstract description 17
- 238000003825 pressing Methods 0.000 description 11
- 230000006835 compression Effects 0.000 description 6
- 238000007906 compression Methods 0.000 description 6
- 230000009471 action Effects 0.000 description 5
- 230000002612 cardiopulmonary effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000012888 cubic function Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001125 extrusion Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 208000007204 Brain death Diseases 0.000 description 1
- 208000010496 Heart Arrest Diseases 0.000 description 1
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 206010049418 Sudden Cardiac Death Diseases 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008081 blood perfusion Effects 0.000 description 1
- 230000000747 cardiac effect Effects 0.000 description 1
- 238000012669 compression test Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000005399 mechanical ventilation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 208000014221 sudden cardiac arrest Diseases 0.000 description 1
- 230000003519 ventilatory effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
- 230000002747 voluntary effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
- G01B21/28—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring areas
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/005—Testing of complete machines, e.g. washing-machines or mobile phones
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/008—Subject matter not provided for in other groups of this subclass by doing functionality tests
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4023—Scaling of whole images or parts thereof, e.g. expanding or contracting based on decimating pixels or lines of pixels; based on inserting pixels or lines of pixels
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Geometry (AREA)
- Percussion Or Vibration Massage (AREA)
Abstract
The invention provides a cross section change rate acquisition method of a thoracic cavity simulator based on AI, which comprises the following steps: s1, testing the chest simulator for a plurality of times by using pressure and recording the testing result of each time, S2, calculating the change rate of the cross section area of each test; s3, correspondingly listing the test result and the cross-sectional area change rate of each time in a table as a training sample and a test sample; s4, training the multiple linear regression model by using the training samples to obtain a trained multiple linear regression model; and S5, inputting the test results of all the moving parts into the trained multiple linear regression model to obtain the cross section change rate of the chest simulator. According to the cross section change rate obtaining method of the thoracic cavity simulator based on the AI, the test result is input into the cross section change rate of the trained multiple linear regression model, the cross section area before and after the state change of the thoracic cavity simulator is not required to be calculated, and therefore the test efficiency of testing the cardiopulmonary resuscitation machine is improved.
Description
Technical Field
The invention relates to the technical field of cardio-pulmonary resuscitation medical equipment, in particular to a cross section change rate acquisition method of a thoracic cavity simulator based on AI.
Background
Cardiopulmonary resuscitation, CPR, is a life-saving technique taken on the heart and breath of sudden cardiac arrest, used to perform chest cardiac compressions on patients with acute cardiac arrest, with the aim of restoring the patient's voluntary behavior. Currently, most cardiopulmonary resuscitation uses chest compressions as a form of aid, usually involving both mechanical and freehand. Regardless of the compression mechanism, the ultimate goal is to change the cross-sectional area of the human thorax at a certain frequency to achieve effective blood perfusion and prevent brain death of the patient.
A cpr machine, also commonly referred to as a cpr apparatus, is a device that performs basic life support operations such as artificial respiration (mechanical ventilation) and chest compressions with mechanical instead of manual power. Such devices can provide high levels of uninterrupted manual circulation and ventilatory support, and certain portable, mobile cardiopulmonary resuscitators can be used in pre-hospital emergencies, without significantly affecting their operation even during transport of the patient.
Chinese patent publication No. CN114220326A discloses a human thorax simulator, comprising a moving part and an elastic part, wherein: the moving part is a structure which is used for bearing external pressing force and/or extrusion force and then moves, and a plurality of moving parts are arranged around the central area; when each moving part bears external pressing force and/or extrusion force, each moving part can approach or separate from the central area along the respective moving path, and the moving paths of all the moving parts radially surround the central area; each moving part is provided with a cut-off end point used for limiting the moving range of the moving part at one end of the moving path far away from the central area; the elastic part is a structure used for providing an elastic action pointing to a far away center area for the moving part; the elastic part elastically acts on the moving part, and in an initial state, the moving part abuts against the stop end point under the action of the elastic force of the elastic part. The simulator simulating the human thorax can be used for respectively testing the effects of the cardiopulmonary resuscitation machines of different manufacturers and models and manual compression, the pressure acting on the moving part is mainly used for deforming the whole simulator, and the change rate of the cross section area of the simulator reflects the effect of the cardiopulmonary resuscitation. Therefore, how to accurately and efficiently measure the change rate of the cross-sectional area is a key problem for improving the testing efficiency.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: in order to improve the efficiency of testing the performance of the cardiopulmonary resuscitation machine, the present invention provides an AI-based cross-sectional change rate acquisition method of a thorax simulator to solve the above-mentioned problems.
The technical scheme adopted by the invention for solving the technical problems is as follows: a cross section change rate obtaining method of an AI-based chest simulator comprises the following steps:
s1, testing the chest simulator for a plurality of times by using pressure and recording the testing result of each time, wherein the testing result comprises displacement of all moving parts, ranging data, pressure and/or pressure difference;
s2, calculating the cross-sectional area change rate of each test;
s3, correspondingly listing the test result and the cross-sectional area change rate of each time in a table as a training sample and a test sample;
s4, training the multiple linear regression model by using the training sample, and testing the multiple linear regression model by using the test sample to obtain the trained multiple linear regression model;
and S5, inputting the test results of all the moving parts into the trained multiple linear regression model to obtain the cross section change rate of the chest simulator.
Preferably, in step S2:
setting a plurality of reference points on the outline of one cross section of the chest simulator, measuring and calculating the coordinates of the reference points, fitting a closed curve of the cross section according to the coordinates of the reference points, and calculating the change rate of the cross section area:
E＝(M1-M2)/M1
wherein: m1 is the area of the cross section of the thorax simulator in the initial state; m2 is the cross-sectional area of the chest simulator at the final state; and E is the rate of change of the cross-sectional area of the thoracic cavity.
Preferably, the closed curve is fitted by a cubic spline interpolation algorithm, a linear interpolation algorithm, or a nearest neighbor interpolation algorithm.
Preferably, two reference points are taken as a first endpoint and a second endpoint, and the first endpoint and the second endpoint divide the outline of the cross section into a first curve and a second curve;
and fitting the first curve and the second curve through a cubic spline interpolation algorithm, wherein the first endpoint and the second endpoint meet natural boundary conditions.
Preferably, all pixel points in the front closed curve of the initial state of the chest simulator are colored, and the color value is A;
extracting all pixel points with color values of A or A1-A2 by adopting a threshold segmentation method, counting the number of the pixel points, and calculating the area S1 according to the number of the pixel points, wherein A is more than A1 and less than A2;
overlapping the front closed curve and the rear closed curve of the initial state and the final state of the chest simulator in a rectangular coordinate system overlapping mode; coloring all pixel points inside the front closed curve and outside the rear closed curve with a color value of B, and coloring all pixel points outside the front closed curve and inside the rear closed curve with a color value of C;
extracting all pixels with color values of B or B1-B2 by adopting a threshold segmentation method, counting the number of the pixels, and calculating the area S2 according to the number of the pixels; extracting all pixel points with color values of C or C1-C2 by adopting a threshold segmentation method, counting the number of the pixel points, and calculating an area S3 according to the number of the pixel points, wherein B is more than B1 and less than B2, and C is more than C1 and less than C2;
then E ═ S2-S3)/S1.
Preferably, the area M1 of the cross section of the initial state of the chest simulator and the area M2 of the cross section of the final state of the chest simulator are calculated using fixed integration.
Preferably, the coordinates of the first end point and the second end point in the initial state and the final state of the chest simulator are the same fixed value.
Preferably, the distance data of the first end point is detected by a distance measuring sensor and converted into the coordinate of the first end point; and detecting the distance data of the second end point through a distance measuring sensor and converting the distance data into the coordinate of the second end point.
Preferably, one reference point is selected as a first associated point of the first endpoint, and the coordinate of the first endpoint is obtained according to the coordinate of the first associated point; and selecting a reference point as a second associated point of the second endpoint, and obtaining the coordinate of the second endpoint according to the coordinate of the second associated point.
As a preference, the first and second liquid crystal compositions are,
0.95A≤A1≤0.99A，1.05A≤A2≤1.09A；
0.95B≤B1≤0.99B，1.05B≤B2≤1.09B；
0.95C≤C1≤0.99C，1.05C≤C2≤1.09C。
the cross section change rate obtaining method of the AI-based chest simulator has the advantages that the test result is input into the cross section change rate of the trained multiple linear regression model, the cross section area before and after the state change of the chest simulator does not need to be calculated, and therefore the test efficiency of testing the cardiopulmonary resuscitation machine is improved.
Drawings
The invention is further illustrated with reference to the following figures and examples.
FIG. 1 is a flow chart of an embodiment of the present invention.
FIG. 2 is a perspective view of a chest simulator in accordance with an embodiment of the present invention;
FIG. 3 is a front view of a chest simulator in accordance with an embodiment of the present invention;
FIG. 4 is a cross-sectional view of a chest simulator in accordance with an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of the chest simulator after being installed with a cardiopulmonary resuscitation machine according to the embodiment of the present invention;
FIG. 6 is a rectangular coordinate system, and an end point and reference point diagram according to an embodiment of the present invention;
FIG. 7 is a schematic diagram illustrating detection of end point position values in a rectangular coordinate system according to an embodiment of the present invention;
FIG. 8 is a first graphical illustration of an embodiment of the present invention constructed from a reference point and first and second endpoints on a first curve;
FIG. 9 is a second graphical illustration of an embodiment of the present invention constructed from a reference point and first and second endpoints on a second curve;
FIG. 10 is a schematic diagram of an embodiment of the present invention dividing a first curve into 8 intervals;
FIG. 11 is a front closure curve of the cross-sectional profile of the thorax simulator in an initial state in accordance with the embodiment of the present invention;
FIG. 12 is a posterior closed curve of a cross-sectional profile of a final state of a chest simulator in accordance with an embodiment of the present invention;
FIG. 13 is a schematic cross-sectional view of an initial state of a chest simulator in accordance with an embodiment of the present invention;
FIG. 14 is a schematic cross-sectional area view of a thorax simulator in both initial and final states in accordance with an embodiment of the present invention;
FIG. 15 is a schematic cross-sectional area view of a thorax simulator in both initial and final states in accordance with another embodiment of the present invention;
FIG. 16 is a cross-sectional area schematic view of a thorax simulator in both initial and final states in accordance with still another embodiment of the present invention.
In the figure, 1, a moving part; 2. an elastic portion; 3. a guide support portion; 4. a ranging sensor; 5. a center block; 6. a chest simulator; 7. a cardiopulmonary resuscitation machine; 8. a bandage; 9. a pressing head; 10. a guide bar; 11. a guide sleeve; 12. a nut; 13. a first endpoint; 14. a second endpoint; 15. a reference point; 16. a pressing part; 17. a first curve; 18. a second curve; 19. a front closure curve; 20. the curve is closed afterwards.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
As shown in fig. 2 to 5, the thorax simulator 6 is a device for simulating the change of the cross section of the thorax of a human body under the action of the cardiopulmonary resuscitation machine 7, and mainly comprises a moving part 1, an elastic part 2, a guide supporting part 3, a central block 5 and a distance measuring sensor 4.
The 6 moving parts 1 are arranged around the center block 5, and the elastic part 2 is a structure for providing an elastic force to the moving parts 1. The cardiopulmonary resuscitation machine 7 is mounted on the chest simulator 6 using a bandage 8. In the initial state of the thorax simulator 6, the pressing head 9 of the cardiopulmonary resuscitation machine 7 now faces the pressing part 16 of the uppermost mobile part 1, and the bandage 8 is attached to the other mobile parts. The bandage 8 has flexibility, and when the pressing head 9 applies a force to the moving portion 1, the bandage 8 can be always attached to the moving portion 1.
The elastic portion 2 is specifically a spring. The guiding and supporting part 3 is composed of a guide rod 10, a guide sleeve 11 and a nut 12, wherein the guide sleeve 11 is in a sleeve structure and is assembled on the guide rod 10 in a sliding mode, and the moving part 1 is fixed relative to the guide sleeve 11, so that the moving part 1 can slide along the guide rod 10. In the assembled state, the spring 2 acts in the longitudinal direction of the guide bar 10. Specifically, a spring is sleeved on a guide rod 10, one end of the spring is positioned oppositely, and the other end of the spring acts on a moving part 1; in order to define the starting position of the moving part 1, a nut 12 is mounted on the guide bar 10. The central block 5 is a base body of the thorax simulator 6, the central block 5 is a block-shaped body structure, and each guide rod 10 is fixedly supported on the central block 5. The distance measuring sensor 4 is mounted on the center block 5, and detects distance data of the distance measuring sensor 4 and the moving part 1.
As shown in fig. 1. The invention provides an embodiment of a cross section change rate obtaining method of a thoracic cavity simulator based on AI, which comprises the following steps:
the chest simulator 6 is subjected to a plurality of compression tests using the cardiopulmonary resuscitator 7, and the distance data of each moving part 1 and the distance sensor 4 before and after the state change of the chest simulator 6 in each test is acquired by the distance sensor 4, so that the data is converted into the displacement. The initial state refers to a state of the thoracic simulator 6 when the pressing head 9 is not driven to be further extended outward, and the final state refers to a state of the thoracic simulator 6 when the pressing head 9 is fully extended.
As shown in fig. 6, at least one point on each of the moving parts 1 is selected as a reference point 15, and all the reference points 15 are located in the same plane, which forms a cross section by cutting the chest simulator 6 and the bandage 8, and all the reference points 15 fall on the outline of the cross section. Two reference points are selected on the cross section outline, namely a first end point 13 and a second end point 14. To this end, the profile of the cross-section is divided by a first end point 13 and a second end point 14 into a first curve and a second curve.
A rectangular coordinate system is established in this plane. As shown in fig. 7, the distance measuring sensor 4 may be additionally provided to acquire the distances between the first and second end points 13 and 14 and the distance measuring sensor, and since the coordinates of the distance measuring sensor 4 in the coordinate system are determined, the coordinates of the first and second end points 13 and 14 before and after the state change of the chest simulator 6 may be converted according to the distances. In the present embodiment, in order to fit the first curve and the second curve by the cubic spline interpolation algorithm, the first endpoint 13 and the second endpoint 14 satisfy the natural boundary condition, and the outline of the cross section is close to the regular ellipse, so that the first endpoint 13 and the second endpoint 14 both fall on the X-axis of the coordinate system. In order to simplify the operation, in other examples, the first endpoint 13 and the second endpoint 14 may be considered as fixed points whose coordinates do not change before and after the state of the chest simulator 6 changes, and it is not necessary to additionally provide a distance measuring sensor for the first endpoint 13 and the second endpoint 14, and the fixed coordinates may be directly used when the curve is fitted.
As shown in fig. 6, a total of 14 reference points 15 are provided on the 6 moving parts 1 in this example. Wherein, 3 reference points 15 are respectively arranged on the upper and lower moving parts 1, and 2 reference points 15 are respectively arranged on other moving parts 1, the reason for the arrangement is that the thoracic cavity simulator 6 is under the action of the pressing head 9 of the cardiopulmonary resuscitator 7The deformation of the upper part and the lower part of the cross section outline of the simulator 6 is large, the number of the reference points 15 is large, the real situation can be reflected more accurately, and the simulation accuracy is further improved. In addition to the reference point 15, the coordinates of the target point directly detected by the distance measuring sensor 4 are also determined on the moving part 1, and the coordinates of the reference point 15 before and after the state change of the chest simulator 6 can be further converted from the measured distance regardless of whether the target point is the reference point 15 or another point. In other embodiments, the first endpoint 13 and the second endpoint 14 are still points whose coordinates may change, but it is not necessary to additionally set a distance measurement sensor for the first endpoint 13 and the second endpoint 14, but one reference point 15 is selected as a first associated point of the first endpoint, and the coordinates of the first endpoint are obtained according to the coordinates of the first associated point; and selecting one reference point 15 as a second associated point of the second endpoint, and obtaining the coordinate of the second endpoint according to the coordinate of the second associated point. For example, the first endpoint has coordinates of (x)e，ye) Coordinates (x) of the first associated pointr，yr) Then xe＝xr+ empirical value, ye＝yr+ empirical value.
As shown in FIGS. 8-10, the first curve 17 and the second curve 18 are each constructed using a cubic spline interpolation algorithm. The following describes how to construct the first curve 17 by using a cubic spline interpolation algorithm, taking the first curve 17 as an example. Cubic spline interpolation algorithm, also called cubic interpolation method, is a polynomial interpolation method, which successively uses cubic curve phi (t) as a0+a1t+a2t2+a3t3The minimal point approximation method of (a) is a method for finding the minimal point of the function f (t), which can be found in encyclopedia of hundred degrees and related content introduction, and the algorithm per se belongs to the prior art.
In the application, a piecewise interpolation mode is adopted, and coordinates corresponding to the known first endpoint 13, the reference point on each first curve 17 and the second endpoint 14 are assumed to be (x)0,y0),(x1,y1)，………，(xn,yn) I.e. with the first end point 13 and the second end point 14 as the two end points of the curve, the first curve 17 is referred to the reference pointIs divided into n intervals on the X axis, and the X axis coordinate of the n intervals is (X)0,x1),(x1,x2),………,(xn-1,xn) Total n +1 points, where x0,xnThe X-axis coordinates of the two endpoints. In the present embodiment, the first curve 17 is divided into 8 sections, and n is 8. Cubic spline is to define the curve between each cell as a cubic equation in the interval (x)i,xi+1) The corresponding cubic function in (1) is:
y＝ai+bix+cix2+dix3is like
In the formula I: a is ai、bi、ci、diRespectively representing four unknown constants in a cubic function, wherein i represents a point location code number of an interval, and the value range of i is 0-n; a isi、bi、ci、diThe solving formula of (2) is as follows:
ai＝yi
wherein h isiStep size of the interval is represented by the formula hi＝xi+1-xi(ii) a And m isiRepresents an unknown constant;
at natural boundary conditions, m0＝0，mn＝0；
The second expression is m as unknown numberWherein: y is0、y1、y2…ynY-axis coordinates representing the first end point (13), each upper reference point, and the second end point (14); passing through type two can solve m0、m1、m2…mnFurther, a is obtainedi、bi、ci、diFurther solving the expression of each cell;
as shown in fig. 11 and 12, since the curve equation between each cell of the first curve 17 and the second curve 18 is determined, the front closure curve 19 of the cross section of the initial state of the chest simulator 6 can be fitted accordingly. A rear closed curve 20 of the cross section of the final state of the chest simulator 6 is fitted in the same way. In other embodiments, a linear interpolation algorithm or a nearest neighbor interpolation algorithm may also be used to fit the closed curve.
Calculating the rate of change of the cross-sectional area:
E＝(M1-M2)/M1
wherein: m1 is the cross-sectional area of the thorax simulator 6 in the initial state; m2 is the cross-sectional area of the chest simulator 6 at the end state; and E is the rate of change of the cross-sectional area of the thoracic cavity.
As shown in FIG. 13, for the area calculation, in this example, all the pixels in the anterior closure curve 19 of the initial state of the thorax simulator are colored with a color value of A, which is a pixel value, and is usually in the range of 0-255. And extracting all pixel points with the color value of A by adopting a threshold segmentation method, counting the number of the pixel points, and calculating the area S1 according to the number of the pixel points. In order to get the pixels at the border of the graph colored as a and make the data more accurate, all the pixels with color values between a1 and a2 may be extracted, where a1 is slightly smaller than a and a2 is slightly larger than a. For example, a value of 190 for a1 is 180 and a value of 200 for a 2.
Overlapping a front closed curve 19 and a rear closed curve 20 of the initial state and the final state of the chest simulator in a manner of superposing rectangular coordinate systems; coloring all the pixel points inside the front closed curve 19 and outside the rear closed curve 20 with a color value of B, and coloring all the pixel points outside the front closed curve 19 and inside the rear closed curve 20 with a color value of C; extracting all pixels with color values of B or B1-B2 by adopting a threshold segmentation method, counting the number of the pixels, and calculating the area S2 according to the number of the pixels; extracting all pixels with the color value of C or C1-C2 by adopting a threshold segmentation method, counting the number of the pixels, and calculating the area S3 according to the number of the pixels, wherein B is more than B1 and less than B2, B1 is more than or equal to 0.95B and less than or equal to 0.99B, and B2 is more than or equal to 1.05B and less than or equal to 1.09B; c is more than C1 and less than C2, C is more than or equal to 0.95 and less than or equal to 0.99, C is more than or equal to 1.05 and less than or equal to 1.09; then E ═ S2-S3)/S1. As shown in fig. 14, in this example, since the two ends of the bandage 8 are connected to the cardiopulmonary resuscitation device 7, the body is in a movable state, so that when the pressing head 9 is driven to further extend, the body moves upward while pulling the bandage 8, so that the bandage 8 is attached to the moving part 1 and applies pressure to the moving part 1, thereby the whole chest simulator is contracted, and there is no area outside the front closed curve 19 and inside the rear closed curve 20, S3 is 0. The overall shrinkage can also be in the form shown in fig. 15. In other embodiments, the bandage is wrapped around the chest simulator for one turn and over all moving parts, and pressure is applied to the chest simulator by the CPR machine or manually, while the body of the CPR machine 7 needs to be immobilized and the compression head is ready to compress against one of the moving parts. The nuts on the partial moving parts are removed, and the moving parts are not limited by the nuts and are limited by bandages. When the chest simulator is in the final state, the moving parts are expanded outwards under the action of the spring due to the deformation of the bandage, and the chest simulator is in a state that partial areas shrink and partial areas expand, and then areas outside the front closed curve 19 and inside the rear closed curve 20 exist, as shown in fig. 16, S3 is not equal to 0; in this example, the cardiopulmonary resuscitator used is not required to use a bandage when actually acting on a human body, and is directly pressed.
In other embodiments, the areas corresponding to the front closed curve 19 and the rear closed curve 20 may be calculated by integrating them. The specific method for calculating the area of the closed curve is as follows:
obtaining a curve formula through a cubic spline interpolation algorithm, and respectively carrying out fixed integral calculation on the formula before and after change to obtain corresponding closed curve areas;
further, the domain of the closed curve before change is [ x ]0，xn]The definition domain of the closed curve after change is [ x ]0，xn]。
The initial state area was found to be M1, and the final state area was found to be M2.
Next, the displacement amount and the cross-sectional area change rate of each moving part 1 in each test are correspondingly listed in a table, 95% of data is used as a training sample, and 5% of data is used as a test sample; for example, the following table:
in other embodiments, a pressure sensor may be disposed on the mobile portion 1 to detect the pressure applied to the mobile portion 1, and the displacement of the training sample and the test sample may be partially or completely replaced by the distance measurement data, the applied pressure, and/or the applied pressure difference.
Training the multiple linear regression model by using a training sample, and testing the multiple linear regression model by using a test sample to obtain a trained multiple linear regression model;
finally, when the multiple linear regression model is used, the test results of all the moving parts 1 in the final state of the chest simulator 6 are input into the trained multiple linear regression model, and the cross section change rate of the chest simulator 6 is obtained. The input data can be measured once, or the batch input can be measured for multiple times, or the generated measured data can be transmitted to the multiple linear regression model in real time, and the change rate of the cross section can be output in real time.
The following provides a specific tool for training a multiple linear regression model:
the required related libraries are imported in the pycharm software (realized by commands), and the method comprises the following steps: pandas, numpy, sklern, visualization library matplotlib, and the like;
read _ csv ('training sample data path');
defining the column corresponding to the test result of each moving part 1 as an independent variable and the cross-sectional area change rate as a dependent variable through an iloc function;
applying a multiple linear regression model: linear _ model. linear regression () is trained;
and testing the multiple regression model by using the test sample to verify the accuracy of the model.
In light of the foregoing description of the preferred embodiment of the present invention, many modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention. The technical scope of the present invention is not limited to the content of the specification, and must be determined according to the scope of the claims.
Claims (10)
1. A cross section change rate acquisition method of an AI-based chest simulator is characterized by comprising the following steps of:
s1, testing the chest simulator for a plurality of times by using pressure and recording the testing result of each time, wherein the testing result comprises displacement of all moving parts, ranging data, pressure and/or pressure difference;
s2, calculating the cross-sectional area change rate of each test;
s3, correspondingly listing the test result and the cross-sectional area change rate of each time in a table as a training sample and a test sample;
s4, training the multiple linear regression model by using the training sample, and testing the multiple linear regression model by using the test sample to obtain the trained multiple linear regression model;
and S5, inputting the test results of all the moving parts into the trained multiple linear regression model to obtain the cross section change rate of the chest simulator.
2. The AI-based thorax simulator cross-sectional change rate acquisition method according to claim 1, wherein in the step S2:
setting a plurality of reference points on the outline of one cross section of the chest simulator, measuring and calculating the coordinates of the reference points, fitting a closed curve of the cross section according to the coordinates of the reference points, and calculating the change rate of the cross section area:
E＝(M1-M2)/M1
wherein: m1 is the area of the cross section of the thorax simulator in the initial state; m2 is the cross-sectional area of the thorax simulator in the final state; and E is the rate of change of the cross-sectional area of the thoracic cavity.
3. The AI-based thorax simulator cross-sectional change rate acquisition method of claim 2, wherein:
and fitting a closed curve by a cubic spline interpolation algorithm, a linear interpolation algorithm or a nearest neighbor interpolation algorithm.
4. The AI-based thorax simulator cross-sectional change rate acquisition method of claim 3, wherein:
taking two reference points as a first endpoint and a second endpoint, wherein the first endpoint and the second endpoint divide the outline of the cross section into a first curve and a second curve;
and fitting the first curve and the second curve through a cubic spline interpolation algorithm, wherein the first endpoint and the second endpoint meet natural boundary conditions.
5. The AI-based thorax simulator of claim 4, wherein:
coloring all pixel points in a front closed curve of the initial state of the chest simulator, wherein the color value is A;
extracting all pixel points with color values of A or A1-A2 by adopting a threshold segmentation method, counting the number of the pixel points, and calculating the area S1 according to the number of the pixel points, wherein A is more than A1 and less than A2;
overlapping the front closed curve and the rear closed curve of the initial state and the final state of the chest simulator in a rectangular coordinate system overlapping mode; coloring all pixel points inside the front closed curve and outside the rear closed curve with a color value of B, and coloring all pixel points outside the front closed curve and inside the rear closed curve with a color value of C;
extracting all pixels with color values of B or B1-B2 by adopting a threshold segmentation method, counting the number of the pixels, and calculating the area S2 according to the number of the pixels; extracting all pixel points with color values of C or C1-C2 by adopting a threshold segmentation method, counting the number of the pixel points, and calculating an area S3 according to the number of the pixel points, wherein B is more than B1 and less than B2, and C is more than C1 and less than C2;
then E ═ S2-S3)/S1.
6. The AI-based thorax simulator of claim 4, wherein:
the area M1 of the cross section of the initial state of the chest simulator and the area M2 of the cross section of the final state of the chest simulator were calculated using fixed integration.
7. The AI-based thorax simulator cross-sectional change rate acquisition method as recited in claim 5 or 6, wherein: the coordinates of the first endpoint and the second endpoint in the initial state and the final state of the chest simulator are the same fixed value.
8. The AI-based thorax simulator cross-sectional change rate acquisition method as recited in claim 5 or 6, wherein: detecting distance data of the first endpoint through a distance measuring sensor and converting the distance data into coordinates of the first endpoint; and detecting the distance data of the second end point through a distance measuring sensor and converting the distance data into the coordinate of the second end point.
9. The AI-based thorax simulator cross-sectional change rate acquisition method as recited in claim 5 or 6, wherein: selecting a reference point as a first associated point of the first endpoint, and obtaining the coordinate of the first endpoint according to the coordinate of the first associated point; and selecting a reference point as a second associated point of the second endpoint, and obtaining the coordinate of the second endpoint according to the coordinate of the second associated point.
10. The AI-based thorax simulator cross-sectional change rate acquisition method as recited in claim 5 or 6, wherein:
0.95A≤A1≤0.99A，1.05A≤A2≤1.09A；
0.95B≤B1≤0.99B，1.05B≤B2≤1.09B；
0.95C≤C1≤0.99C，1.05C≤C2≤1.09C。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210419454.7A CN114693715B (en) | 2022-04-20 | 2022-04-20 | Method for acquiring cross section change rate of AI-based chest simulator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210419454.7A CN114693715B (en) | 2022-04-20 | 2022-04-20 | Method for acquiring cross section change rate of AI-based chest simulator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114693715A true CN114693715A (en) | 2022-07-01 |
| CN114693715B CN114693715B (en) | 2023-04-28 |
Family
ID=82144591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210419454.7A Active CN114693715B (en) | 2022-04-20 | 2022-04-20 | Method for acquiring cross section change rate of AI-based chest simulator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114693715B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108875113A (en) * | 2017-09-07 | 2018-11-23 | 湖南大学 | Fusion penetration prediction and control method and the system of the cross sectional plates that thicken laser welding |
| CN111178756A (en) * | 2019-12-29 | 2020-05-19 | 杭州拓深科技有限公司 | Multiple linear regression fire risk assessment method based on environmental big data |
| CN113113150A (en) * | 2021-04-15 | 2021-07-13 | 上海交通大学医学院附属第九人民医院 | Lymph node metastasis prediction model construction and training method, device, equipment and medium |
| CN114373372A (en) * | 2021-11-29 | 2022-04-19 | 苏州尚领医疗科技有限公司 | Human chest cardio-pulmonary resuscitation simulation method |
-
2022
- 2022-04-20 CN CN202210419454.7A patent/CN114693715B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108875113A (en) * | 2017-09-07 | 2018-11-23 | 湖南大学 | Fusion penetration prediction and control method and the system of the cross sectional plates that thicken laser welding |
| CN111178756A (en) * | 2019-12-29 | 2020-05-19 | 杭州拓深科技有限公司 | Multiple linear regression fire risk assessment method based on environmental big data |
| CN113113150A (en) * | 2021-04-15 | 2021-07-13 | 上海交通大学医学院附属第九人民医院 | Lymph node metastasis prediction model construction and training method, device, equipment and medium |
| CN114373372A (en) * | 2021-11-29 | 2022-04-19 | 苏州尚领医疗科技有限公司 | Human chest cardio-pulmonary resuscitation simulation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114693715B (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bachofen et al. | Area analysis of pressure-volume hysteresis in mammalian lungs. | |
| Vigmostad et al. | Fluid–structure interaction methods in biological flows with special emphasis on heart valve dynamics | |
| KR20200006447A (en) | Method and program for stroke volume acquisition using arterial pressure waveform | |
| CN104274172A (en) | Monitoring method and system for external chest compression depth and defibrillator | |
| Finsberg et al. | Estimating cardiac contraction through high resolution data assimilation of a personalized mechanical model | |
| US11020020B2 (en) | Variable ventilation as a diagnostic tool for assessing lung mechanical function | |
| Niu et al. | Study on air flow dynamic characteristic of mechanical ventilation of a lung simulator | |
| Houssein et al. | Estimation of respiratory variables from thoracoabdominal breathing distance: a review of different techniques and calibration methods | |
| CN114693715B (en) | Method for acquiring cross section change rate of AI-based chest simulator | |
| CN114664417A (en) | Virtual first-aid intelligent training method based on Gaussian model | |
| CN104714200A (en) | Magnetic resonance super undersampled K data imaging method based on studying generalized double-layer Bergman non-convex-type dictionary | |
| Cecchini et al. | An open-loop controlled active lung simulator for preterm infants | |
| Tomlinson et al. | Computer simulation of the human breathing process | |
| Kretschmer et al. | Differences in respiratory mechanics estimation with respect to manoeuvres and mathematical models | |
| CN114041758B (en) | Radial artery palpation positioning method and device, electronic equipment and storage medium | |
| Sauer et al. | Blind source separation of inspiration and expiration in respiratory sEMG signals | |
| Verbraak et al. | Serial lung model for simulation and parameter estimation in body plethysmography | |
| KR101606012B1 (en) | Method for correcting data of Center Of Pressure using image processing technique | |
| Baudet et al. | Towards Accurate Tumour Tracking in Lungs. | |
| WO2007129908A1 (en) | System and method for validating a qcpr device | |
| Fry | The physics of air flow in emphysema | |
| CN114187582A (en) | Artificial heart-lung machine data recording method based on artificial neural network technology | |
| CN113456959A (en) | Method and device for setting positive end expiratory pressure of respirator and storage medium | |
| JP2022526569A (en) | A method for designing and verifying the shape and positioning of a cannula for a patient by simulating insertion into the trachea. | |
| Nakamura et al. | A new method of analyzing the distribution of mechanical time constants in the lungs. |
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 |