Anesthesia department airway intubation image recognition method based on artificial intelligence

Abstract

The present application relates to the technical field of image processing, in particular to an anesthesia department airway intubation image recognition method based on artificial intelligence. A patient airway CT image is collected by double-view, a multi-curvature airway model is constructed and a corresponding image is generated; the model image is corrected in combination with the inherent distortion parameters of the lens to obtain a distortion correction image; then the influence of different curvatures on distortion is quantified under the multi-curvature distortion correction image to generate a mapping relationship of 'curvature-direction-distortion correction factor'; the current airway image is obtained in real time during intubation, the curvature feature is extracted by pixel-level matching with the distortion correction image, and is associated with the aforementioned mapping relationship to determine the distortion correction index corresponding to the current airway image; finally, the current airway image is double-corrected in combination with the lens distortion parameters and the distortion correction index to generate a high-credibility airway correction image for subsequent image recognition.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to an anesthesia airway intubation image recognition method based on artificial intelligence. BACKGROUND

[0002] In the clinical practice of anesthesiology, airway intubation is the core operation to ensure the smooth breathing of patients, and its success rate is directly related to patient safety. In recent years, the integration of artificial intelligence and medical images has provided a new way to solve this problem. By accurately identifying the airway image during airway intubation to plan the optimal intubation path, the catheter tip can be dynamically guided to avoid high-risk areas. This intelligent auxiliary system not only reduces the operation threshold, but also saves valuable time window for emergency rescue.

[0003] Current airway intubation mainly relies on endoscopic video guidance, and doctors judge the catheter position by visual observation. However, due to the design of the optical lens, it usually produces distortion, resulting in distorted airway images. The existing technology usually uses image distortion correction technology to correct the inherent distortion of the lens. However, the traditional distortion model usually assumes that the imaging environment is a straight pipe, while the patient's airway usually has different degrees of curvature. The curved section also causes serious image stretching / compression distortion. Therefore, if only the inherent distortion of the lens is corrected and the individual airway curvature characteristics of the patient are ignored, the accuracy and reliability of the final corrected airway image will be low, affecting the subsequent recognition effect. SUMMARY

[0004] In order to solve the technical problem that if only the inherent distortion of the lens is corrected and the individual airway curvature characteristics of the patient are ignored, the accuracy and reliability of the final corrected airway image will be low, affecting the subsequent recognition effect, the purpose of the present application is to provide an anesthesia airway intubation image recognition method based on artificial intelligence, and the technical solution adopted is as follows:

[0005] Obtain the airway CT image of the patient under double-view, obtain the airway model image of different curvatures, and obtain the airway image at the current time during airway intubation;

[0006] Based on the feature points in each airway model image, the airway model image is corrected using the lens inherent distortion correction parameters, so as to obtain the distortion correction image corresponding to each airway model image; compare the distortion correction images under different curvatures, so as to obtain the distortion correction factor of each curvature distortion correction image in each preset direction;

[0007] In each airway CT image, an airway wall edge line is extracted and a curvature value of each edge pixel point on the edge line is calculated; pixel points in the airway image are matched with the edge pixel points on the airway wall edge line to obtain matched edge points; based on the curvature values of the matched edge points, position conditions and distortion correction factors of a distortion correction image in each preset direction under each curvature, a distortion correction index of the airway image is determined;

[0008] Based on the distortion correction index of the airway image and the lens intrinsic distortion correction parameters, the airway image is subjected to distortion correction, so as to obtain an airway correction image for image recognition.

[0009] Further, the method for obtaining the distortion correction image comprises:

[0010] According to the lens intrinsic distortion correction parameters and the feature points in each airway model image, each airway model image is subjected to image correction by using a Brown-Conrady distortion model, so as to obtain a distortion correction image corresponding to each airway model image.

[0011] Further, the method for obtaining the distortion correction factor comprises:

[0012] Based on a center point and a diagonal line of each airway model image, the airway model image is regionally divided as a region block in four preset directions;

[0013] The distortion correction image with a curvature value of 0 is taken as a standard image, and other distortion correction images except the standard image are taken as to-be-analyzed images;

[0014] In the same preset direction, in the region blocks of the standard image and each to-be-analyzed image, according to the feature points and the Brown-Conrady distortion model, a distortion correction factor corresponding to the region block of each to-be-analyzed image in each preset direction is obtained.

[0015] Further, the method for obtaining the airway wall edge line comprises:

[0016] Each airway CT image is taken as an input of a pre-trained neural network, so as to output an airway region in each airway CT image;

[0017] Based on a Canny operator, two airway wall edge lines of the airway region in each airway CT image are obtained.

[0018] Further, the method for obtaining the matched edge point comprises:

[0019] In any one airway CT image, the minimum distance of each pixel point in the airway to the airway wall edge line is obtained to generate a distance field, the maximum value point of the distance is obtained in the distance field, and the candidate points are connected through a minimum spanning tree algorithm to obtain an airway center line;

[0020] The DTW algorithm is used to match the points on the airway center point with the edge pixel points on the airway wall edge line to obtain the corresponding points of each point on the airway center point on each airway wall edge line.

[0021] The corresponding point of the point of the current time intubation progress on the airway center line is taken as a matching edge point.

[0022] Further, the distortion correction index acquisition method comprises:

[0023] The distortion includes radial distortion and tangential distortion.

[0024] Under each distortion, the distortion correction factor corresponding to each matching edge point is determined according to the curvature value of the matching edge pixel point and the distortion correction factor of the distortion correction image in each preset direction under each curvature.

[0025] The preset direction corresponding to each matching edge point is determined according to the viewing angle of the airway CT image to which each matching edge point belongs.

[0026] Based on the preset direction and the curvature value corresponding to each matching edge point, a curvature vector of each matching edge point is constructed, the curvature vectors of all matching edge points are vector summed to obtain a comprehensive curvature vector corresponding to the airway image at the current time.

[0027] The included angle between the comprehensive curvature vector and each preset direction is taken as the deviation angle value corresponding to each preset direction.

[0028] Under each preset direction, the distortion correction factor corresponding to each matching edge point under each distortion is weighted and fused with the deviation angle value to obtain the distortion correction index of the airway image at the current time under each distortion.

[0029] Further, the determination of the distortion correction factor corresponding to each matching edge point comprises:

[0030] Under each distortion, the distortion correction factor curves of each preset direction under each curvature are obtained by curve fitting.

[0031] Based on the curvature value of each matching edge point, the corresponding distortion correction factor is obtained in the corresponding distortion correction factor curve.

[0032] Further, the airway correction image acquisition method comprises:

[0033] The distortion correction index of the airway image is substituted into a Brown-Conrady distortion model, so that the airway image is corrected to obtain a primary correction image;

[0034] The lens inherent distortion correction parameter is substituted into the Brown-Conrady distortion model, so that the primary correction image is secondarily corrected to obtain the airway correction image.

[0035] Further, the feature point acquisition method comprises:

[0036] The corner points in each airway model image are acquired by corner detection and are taken as feature points.

[0037] Further, the neural network can adopt a CNN.

[0038] The present application has the following beneficial effects:

[0039] The airway CT image of the patient is acquired under double visual angles, and the airway model of different curvatures is constructed and the airway model image is acquired, and the airway image at the current moment in the airway intubation process is acquired. After the feature points in each airway model image are acquired, the airway model image is corrected in combination with the lens inherent distortion correction parameter, the initial interference of the lens inherent distortion is eliminated, and the "unbiased" distortion correction image is provided for subsequent curvature adaptation. Under the distortion correction image corresponding to the airway model of multiple curvatures, the influence of different bending degrees on the image distortion is quantified, the distortion correction factors in different preset directions are generated, and the accurate mapping of "curvature-direction-distortion correction factor" is realized. Further, the morphological characteristics of the airway in the airway CT image of the patient are analyzed, the key geometric parameters (curvatures of the edge pixel points on the airway wall edge) are provided for individualized correction, and the adaptation error caused by the dependence on the fixed model is avoided. The association between the current position in the real-time process and the morphological characteristics of the airway is established through pixel-level matching, the accurate correspondence between the subsequent correction parameters and the current intubation progress is ensured, and the reliability of dynamic adaptation is improved. Then, the distortion correction index of the real-time airway image is generated by fusing the individual airway curvature characteristics and the distortion correction factor. Finally, the high-credibility airway correction image at the current moment in the airway intubation process is generated through double correction (lens inherent distortion correction parameter + distortion correction index), reliable data basis is provided for subsequent intubation path planning, narrow positioning and other recognition tasks, and the misjudgment risk is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, below will briefly introduce the drawings required by the embodiments or prior art description. Obviously, the drawings described below only show some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0041] Figure 1 A method flow chart of an anesthesia airway intubation image recognition method based on artificial intelligence provided by an embodiment of the present application;

[0042] Figure 2 An airway model schematic diagram under different curvatures provided by an embodiment of the present application;

[0043] Figure 3 A schematic diagram of a region block in four preset directions provided by an embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined purpose, the specific implementation, structure, features and effects of the anesthesia airway intubation image recognition method based on artificial intelligence according to the present application are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0046] The specific scheme of the anesthesia airway intubation image recognition method based on artificial intelligence provided by the present application is described in detail below with reference to the drawings.

[0047] Please refer to Figure 1 which shows a method flow chart of an anesthesia airway intubation image recognition method based on artificial intelligence provided by an embodiment of the present application, which includes the following steps:

[0048] Step S1: Obtain the airway CT image of the patient under double-view, obtain the airway model image of different curvatures, and obtain the airway image at the current time during the airway intubation process.

[0049] Since the different curvatures of the airway wall of the patient individual will cause different influences on the distortion of the image obtained during the airway intubation process, the airway CT image of the patient is first obtained in order to obtain the curvatures of different positions of the airway wall in the subsequent process.

[0050] Specifically, before the airway intubation, CT imaging equipment is used to collect CT images of the patient's airway from two perspectives, i.e., the front and the side of the face, so as to obtain the CT images of the patient's airway from two perspectives.

[0051] A magnetic sensor is installed on the medical endoscope, the current position of the endoscope in the airway is obtained in real time through an electromagnetic positioning system, and the airway image at the current time during the airway intubation is obtained through the medical endoscope.

[0052] In the process of performing the airway intubation operation, in order to expand the field of view, the medical endoscope is usually equipped with a short focal length wide-angle lens. The physical characteristics of the wide-angle lens can cause imaging distortion. In addition, due to the different curvatures of different positions in the airway, different curvatures can cause changes in the position and angle of the airway and the lens, thereby causing curvature distortion of the image. The specific principle is as follows: on a flat surface, the light ray follows the rule that the incident angle is equal to the reflection angle, the direction of the reflected light can be predicted, and the camera captures the original shape. On a curved surface, the tangent direction of each point is different, the light ray is reflected / refracted at different angles at different positions, the light ray path diverges or converges, at this time the incident angle is not equal to the reflection angle, the curved plane causes the edge image point to deviate from the optical axis direction, resulting in the actual imaging position deviating from the ideal plane. At this time, the magnification non-uniform effect of the distortion is more significant at the deviation position, and the distortion amount is geometrically magnified. Moreover, when the curvature is large, the tissue is more abrupt, resulting in an increase in the difference in the incident angle of the light ray, thereby causing the image to be more severely distorted (for example, the edge is stretched, the scale is distorted). Conversely, when the curvature is small, the tissue tends to be flat, the light ray is more uniform, and the degree of distortion is correspondingly reduced. Therefore, the greater the curvature of the airway, the greater the degree of image distortion in the image collected by the imaging equipment.

[0053] In order to better measure the distortion effect of different curvatures on image imaging, the imaging equipment distortion under different curvatures can be measured through modeling and experiments, so as to obtain the mapping relationship between the curvature and the image distortion or correction parameter. Therefore, in the embodiment of the present application, a blood vessel model is calibrated with different curvatures, the curvatures cover the normal human airway curvature range ( ) and 0 curvature, and the step is , so as to obtain nine groups of airway models under different curvatures for subsequent quantification of the influence of image distortion. Please refer to Figure 2 , which shows the airway model under different curvatures in an embodiment of the present application. The width of the blood vessel model is set to 15 mm, and the thickness is 3 mm, which is consistent with the average size of the human airway. The blood vessel calibration model is 3D printed with a material similar to human tracheal tissue, i.e., transparent resin, and the inner surface of the blood vessel model is sprayed with a matte white paint, and the length and width are both The grid intersection is taken as a calibration point. Then, an airway model image is acquired by using a short focal length wide-angle lens commonly equipped in a medical endoscope, wherein the shooting angle should be consistent with the angle of the airway image in the airway intubation process, and in order to ensure that the curvature directions of the airway model image are the same, the lens direction should be adjusted when the airway model image is shot to obtain the airway model image under the same bending direction.

[0054] The collection and acquisition of personal information in the embodiments of the present application are all authorized by relevant users, the process does not violate relevant laws and regulations, and does not violate public order and good customs.

[0055] Step S2: based on the feature points in each airway model image, the airway model image is corrected by using the lens inherent distortion correction parameter, so as to obtain the distortion correction image corresponding to each airway model image; the distortion correction images under different curvatures are compared, so as to obtain the distortion correction factors of the distortion correction image under each preset direction.

[0056] Since the airway model image contains the inherent distortion of the lens, in order to avoid the inherent distortion from being superimposed during the subsequent multi-curvature correction fusion, the influence of the inherent distortion of the lens needs to be eliminated at this time. In the embodiment of the present application, the airway model image is mainly corrected by using the inherent distortion correction parameter of the lens and the feature points in the airway model image, so as to obtain the distortion correction image corresponding to each airway model image.

[0057] Firstly, a standard calibration board (such as a chessboard) is analyzed by an imaging device (specifically refers to a short focal length wide-angle lens commonly equipped in a medical endoscope in the present application), so as to obtain the lens inherent distortion correction parameter of the imaging device, including: the imaging device is spaced apart from the standard calibration board by a specific distance (3cm) under a vertical angle, and an image is collected to obtain a distortion image under the angle ; under the same conditions, a standard image of the calibration board is collected by using a device without lens distortion ; for reflecting the distortion effect of the actual imaging device; providing a distortion-free reference for quantifying the distortion degree of the lens.

[0058] the corner points in and are acquired by corner point detection, the position difference of the corner points in the two images is compared, the offset amount of the coordinates caused by the distortion can be quantified, and the corner point coordinates in are inversely calculated by the distortion correction formula of the Brown-Conrady distortion model to obtain the pixel coordinates without distortion.

[0059] The Brown-Conrady distortion model is a classical distortion correction model in the field of optical imaging, and is mainly used to solve the image deformation problem caused by the physical defects of the camera lens. The model divides the distortion into two parts, including a radial distortion component caused by the curvature of the lens, and a tangential component caused by the non-parallelism of the lens and the imaging plane. The complete coordinate transformation formula is:

[0060] ;

[0061] ;

[0062] In the formula, represents the coordinates of a pixel point in the distorted image; , represents the square of the distance of the pixel point to the center of the image; is a radial distortion coefficient, is a tangential distortion coefficient; represents the undistorted coordinates of the pixel point after correction.

[0063] Then, through an optimization algorithm, the lens inherent distortion correction parameters in the Brown-Conrady model are obtained according to the principle of minimizing the distance difference between the corner points in and the corner points in , wherein , wherein is the radial distortion coefficient of the image, is the tangential distortion coefficient of the image.

[0064] It should be noted that the corner point detection algorithm and the Brown-Conrady distortion model mentioned in the embodiments of the present application are all known technologies, and will not be described here.

[0065] Further, based on the feature points in each airway model image, the lens inherent distortion correction parameters obtained in the foregoing are used to correct the airway model image, to obtain a distortion correction image corresponding to each airway model image.

[0066] Preferably, in an embodiment of the present application, the method for obtaining a distortion correction image comprises:

[0067] First, the corner points in each airway model image are obtained through corner point detection, and are used as feature points.

[0068] Then, according to the lens inherent distortion correction parameters (obtained in the foregoing ) and the feature points in each airway model image, the Brown-Conrady distortion model is used to correct each airway model image, so as to obtain a distortion correction image corresponding to each airway model image.

[0069] At this time, the distortion correction image corresponding to each airway model image has eliminated the interference of the lens inherent distortion of the imaging device, and the influence of the curvature of the airway bending on the distortion can be further quantified, so as to extract the distortion correction factors in each direction under different curvatures, and construct the mapping relationship of "curvature-direction-distortion correction factor".

[0070] Preferably, in an embodiment of the present application, the method for obtaining the distortion correction factor comprises:

[0071] In the embodiment of the present application, the distortion influence can be divided into four preset directions, so as to facilitate more detailed analysis, and therefore the airway model image is divided into area blocks in the four preset directions based on the center point and the diagonal line of each airway model image, please refer to Figure 3 which shows the schematic diagram of the area blocks in the four preset directions in an embodiment of the present application, wherein the four preset directions are respectively denoted as up, down, left and right.

[0072] Then, the distortion correction image with a curvature value of 0 is taken as a standard image, and the other distortion correction images except the standard image are taken as to-be-analyzed images.

[0073] In the area blocks of the standard image and each to-be-analyzed image in the same preset direction, the distortion correction factors corresponding to the area blocks of each to-be-analyzed image in each preset direction are obtained according to the feature points (the corner points obtained in the foregoing) and the Brown-Conrady distortion model, wherein the distortion correction factors should include five, three radial distortions and two tangential distortions, the number of which corresponds to the number of the lens inherent distortion correction parameters.

[0074] At this time, the distortion correction image corresponding to each airway model image has eliminated the interference of the lens inherent distortion of the imaging device, and the influence of the curvature of the airway bending on the distortion can be further quantified, so as to extract the distortion correction factors in each direction under different curvatures, and construct the mapping relationship of "curvature-direction-distortion correction factor".

[0075] Step S3: In each airway CT image, the airway wall edge line is extracted and the curvature value of each edge pixel point on the edge line is calculated; the pixel points in the airway image are matched with the edge pixel points on the airway wall edge line to obtain matched edge points; based on the curvature value, position and distortion correction factor of each to-be-analyzed image in each preset direction, the distortion correction index of the airway image is determined.

[0076] In the real airway environment, the direction of the airway bending is often variable, and the analysis in step S2 is only carried out at a specific angle of curvature, which is bound to be different from the actual curvature direction of the airway of the patient, so when the image distortion correction is carried out on the airway image at the current time in the intubation process of the airway of the patient, the curvature information of the airway of the patient is obtained according to the CT image of the airway of the patient, so as to judge the influence of the curvature of the airway of the patient on the distortion, so as to achieve the best correction effect.

[0077] Firstly, the airway wall edge line can be extracted in each airway CT image, and the curvature value of each edge pixel point on the edge line is calculated.

[0078] Preferably, in an embodiment of the present application, the method for obtaining the airway wall edge line comprises:

[0079] Each airway CT image is taken as the input of the pre-trained neural network, so as to output the airway region in each airway CT image, wherein the neural network adopts a CNN neural network.

[0080] Two airway wall edge lines of the airway region in each airway CT image are obtained based on the Canny operator.

[0081] It should be noted that the training process of the neural network is a known technology, and the specific process is not described here; the Canny algorithm for obtaining the edge line is also a known technology, and is not described here.

[0082] Further, because the airway image obtained in the airway intubation process is a real-time dynamic image, and the airway CT image is a static image of the overall shape of the airway of the patient, in order to ensure that the two are consistent in space and shape, so as to accurately obtain the curvature characteristics of the airway in the airway image at the current time, the pixel points in the airway image at the current time can be matched with the edge pixel points on the airway wall edge line to obtain the matching edge points, which are used for subsequent analysis of the curvature characteristics in the airway image at the current time.

[0083] Preferably, in an embodiment of the present application, the method for obtaining the matching edge point comprises:

[0084] In any airway CT image, the minimum distance of each pixel point in the airway to the airway wall edge line is obtained and a distance field is generated, in the distance field, the maximum value points of the distance are obtained, these maximum value points correspond to the candidate points of the center line of the airway, and then the candidate points are connected through the minimum spanning tree algorithm, so as to obtain the center line of the airway.

[0085] The DTW algorithm is used to match the points on the airway center line with the edge pixel points on the airway wall edge line, so that each point on the airway center line has a corresponding point on each airway wall edge line, and if there are multiple mapping points when mapping based on the DTW algorithm, one of them is randomly selected as the corresponding point.

[0086] Since the electromagnetic positioning system can obtain the current position of the endoscope in the airway in real time due to the magnetic sensor installed on the medical endoscope, the current intubation progress in the airway center line can be determined, and therefore the corresponding point of the point on the airway center line of the current airway image after DTW algorithm processing is used as the matching edge point in the current airway image.

[0087] At this time, the airway intubation progress corresponding to the current airway image has a matching edge point on each airway wall edge line.

[0088] It should be noted that the DTW algorithm is a known technology and will not be described here.

[0089] After obtaining the matching edge point corresponding to the intubation progress of the current airway image, the distortion correction index of the current airway image in each preset direction can be determined according to the curvature value, position, and distortion correction factor of each preset direction of the distortion correction image calculated in step S2, which is used to reflect the influence of the curvature feature in the current airway image on the image distortion.

[0090] Preferably, in an embodiment of the present application, the method for obtaining the distortion correction index comprises:

[0091] The distortion includes radial distortion and tangential distortion.

[0092] Since the curvature value of the airway model image may not completely cover every curvature that the airway may have, the distortion correction factors corresponding to each curvature in the same preset direction are curve-fitted using the least square method (with the curvature value as the horizontal coordinate and the distortion correction factor as the vertical coordinate) under each distortion, so as to obtain the distortion correction factor curve of each preset direction under each distortion. At this time, since the two distortions contain three radial distortions and two tangential distortions, five distortion correction factor curves can be obtained.

[0093] Then, the corresponding distortion correction factor (vertical coordinate) is obtained in the corresponding distortion correction factor curve based on the curvature value (horizontal coordinate) of each matching edge point.

[0094] Because the airway CT image is double-view, the two views in the embodiment of the present application are parallel to the front and side of the face, so the preset direction corresponding to each matching edge point can be determined according to the view of the airway CT image to which each matching edge point belongs. Specifically, the preset directions corresponding to the two airway wall edge lines in the airway CT image parallel to the front of the face should be left and right, and the preset directions corresponding to the two airway wall edge lines in the airway CT image parallel to the side of the face should be up and down. In view of the fact that the lens direction is adjusted to obtain the airway model image in the same bending direction, the preset directions corresponding to the two airway wall edge lines in the airway CT image can be determined according to the bending direction of the airway wall edge line. Thus, the preset direction corresponding to each matching edge point can be obtained.

[0095] Further, the curvature vector of each matching edge point is constructed based on the preset direction corresponding to each matching edge point and the curvature value, and the curvature vectors of all matching edge points are summed to obtain the comprehensive curvature vector corresponding to the airway image at the current time.

[0096] Because the distortion correction factor is the mapping between the curvature and the preset direction, the angle between the comprehensive curvature vector and the preset direction is obtained, and the angle and the distortion correction factor of the curvature in the preset direction are fused to obtain the distortion correction index of the airway image at the current time.

[0097] Therefore, the angles between the comprehensive curvature vector and each preset direction are calculated and taken as the deviation angle values corresponding to each preset direction. If the deviation angle value between the comprehensive curvature vector and one of the preset directions is denoted as , then the remaining deviation angle values are , and .

[0098] Finally, in each preset direction, the distortion correction factor corresponding to each matching edge point under each distortion and the deviation angle value are weighted and fused to obtain the distortion correction index of the airway image at the current time under each distortion.

[0099] The formula model of the distortion correction index is: , wherein represents the distortion correction index; represents the distortion correction factor of the matching edge point in the first preset direction; represents the distortion correction factor of the matching edge point in the second preset direction; represents the distortion correction factor of the matching edge point in the third preset direction; represents the distortion correction factor of the matching edge point in the fourth preset direction; represents the deviation angle value between the comprehensive curvature vector and the first preset direction.

[0100] At this point, the distortion correction index of the airway image at the current time can be obtained, and similarly, the distortion correction index should have five, including three radial distortions and two tangential distortions.

[0101] Step S4: based on the distortion correction index of the airway image and the lens inherent distortion correction parameter, the airway image is corrected for distortion, so as to obtain the airway correction image for image recognition.

[0102] Based on the foregoing steps, the distortion correction index of the airway image at the current time can be obtained, which characterizes the influence of airway curvature on distortion. Therefore, in order to obtain the real image in the airway at the current time during the airway intubation process, the airway image captured by the medical endoscope can be corrected twice based on the distortion correction index of the airway image and the lens inherent distortion correction parameter to obtain the airway correction image without image distortion.

[0103] Preferably, in an embodiment of the present application, the method for obtaining the airway correction image comprises:

[0104] The distortion correction index of the airway image is substituted into the Brown-Conrady distortion model, so as to correct the airway image and obtain a primary correction image; the purpose of image distortion correction here is to eliminate the influence of airway wall bending on image distortion.

[0105] Then, the lens inherent distortion correction parameter is substituted into the Brown-Conrady distortion model, so as to correct the primary correction image twice and obtain the airway correction image; the second image distortion correction here is to eliminate the inherent distortion of the camera.

[0106] At this point, the airway correction image can more truly reflect the actual situation in the airway at the current time, so that the image recognition and other processing based on this can more accurately guide the doctor to perform airway intubation and effectively avoid possible accidents during the airway intubation process.

[0107] In summary, the airway CT image of the patient is acquired under the double view angle, and the airway model of different curvatures is constructed and the airway model image is acquired, and the airway image at the current moment in the airway intubation process is acquired. After the feature points in each airway model image are acquired, the airway model image is corrected in combination with the lens inherent distortion correction parameters, the initial interference of the lens inherent distortion is eliminated, and the "unbiased" distortion correction image is provided for subsequent curvature adaptation. Under the distortion correction image corresponding to the airway model of multiple curvatures, the influence of different bending degrees on the image distortion is quantified, the distortion correction factors in different preset directions are generated, and the accurate mapping of "curvature-direction-distortion correction factor" is realized. Further, the morphological characteristics of the airway in the airway CT image of the patient are analyzed, the key geometric parameters (curvatures of edge pixel points on the airway wall edge) for individual correction are provided, and the adaptation error caused by the dependence on the fixed model is avoided. The association between the current position in the real-time process and the morphological characteristics of the airway is established through pixel-level matching, the accurate correspondence between the subsequent correction parameters and the current intubation progress is ensured, and the reliability of dynamic adaptation is improved. Then, the individual airway curvature characteristics and the distortion correction factor are fused to generate the distortion correction index of the real-time airway image. Finally, the high-credibility airway correction image at the current moment in the airway intubation process is generated through double correction (lens inherent distortion correction parameters + distortion correction index), which provides a reliable data basis for subsequent intubation path planning, narrow positioning and other identification tasks, and reduces the risk of misjudgment.

[0108] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0109] Each embodiment in the specification is described in a progressive manner, and the same and similar parts between each embodiment can be referred to each other, and each embodiment mainly describes the difference from other embodiments.

Claims

1. An artificial intelligence-based anesthesiology airway intubation image recognition method, characterized in that, The method comprises: acquiring airway CT images of a patient under double-view angles, acquiring airway model images of different curvatures, and acquiring an airway image at a current time during airway intubation; based on feature points in each airway model image, correcting the airway model image using lens intrinsic distortion correction parameters to obtain a distortion correction image corresponding to each airway model image; comparing distortion correction images under different curvatures to obtain distortion correction factors of the distortion correction image under each curvature in each preset direction; in each airway CT image, extracting an airway wall edge line and calculating curvature values of each edge pixel point on the edge line; matching the pixel points in the airway image with the edge pixel points on the airway wall edge line to obtain matching edge points; based on the curvature values, position conditions of the matching edge points, and the distortion correction factors of the distortion correction image under each curvature in each preset direction, determining a distortion correction index of the airway image; based on the distortion correction index of the airway image and the lens intrinsic distortion correction parameters, correcting the distortion of the airway image to obtain an airway correction image for image recognition; the method for obtaining the distortion correction factors comprises: based on a center point and a diagonal line of each airway model image, dividing the airway model image into region blocks as region blocks in four preset directions; taking a distortion correction image with a curvature value of 0 as a standard image, and taking other distortion correction images except the standard image as to-be-analyzed images; in the same preset direction, in the region blocks of the standard image and each to-be-analyzed image, according to the feature points and a Brown-Conrady distortion model, obtaining a distortion correction factor corresponding to each region block in each preset direction of each to-be-analyzed image.

2. The artificial intelligence-based anesthesiology airway intubation image recognition method of claim 1, wherein the method for obtaining the distortion correction image comprises: based on the lens intrinsic distortion correction parameters and the feature points in each airway model image, correcting each airway model image using the Brown-Conrady distortion model to obtain a distortion correction image corresponding to each airway model image. 3.The anesthesia airway intubation image recognition method based on artificial intelligence according to claim 1, characterized in that, the method for obtaining the airway wall edge line comprises: taking each airway CT image as an input of a pre-trained neural network to output an airway region in each airway CT image; based on a Canny operator, obtaining two airway wall edge lines of the airway region in each airway CT image. 4.The artificial intelligence-based anesthesiology airway intubation image recognition method of claim 1, wherein the method for obtaining the matching edge points comprises: in any airway CT image, obtaining a minimum distance of each pixel point in the airway to the airway wall edge line and generating a distance field, obtaining a maximum value point of the distance in the distance field, and connecting candidate points through a minimum spanning tree algorithm to obtain an airway center line; using a DTW algorithm to match points on the airway center line with edge pixel points on the airway wall edge line to obtain a corresponding point of each point on the airway center line on each airway wall edge line; taking a corresponding point of a point of the intubation progress at the current time on the airway center line as a matching edge point.

5. The artificial intelligence-based anesthesiology airway intubation image recognition method of claim 1, wherein the method for obtaining the distortion correction index comprises: the distortion includes radial distortion and tangential distortion; According to the curvature value of the matched edge pixel point and the distortion correction factor of the distortion correction image in each preset direction under each distortion, a distortion correction factor corresponding to each matched edge point is determined; According to the viewing angle of the airway CT image to which each matched edge point belongs, a preset direction corresponding to each matched edge point is determined; Based on the preset direction and the curvature value corresponding to each matched edge point, a curvature vector of each matched edge point is constructed, and the curvature vectors of all the matched edge points are summed to obtain a comprehensive curvature vector corresponding to the airway image at the current moment; An angle between the comprehensive curvature vector and each preset direction is taken as a deviation angle value corresponding to each preset direction; Under each preset direction, the distortion correction factor corresponding to each matched edge point under each distortion and the deviation angle value are weighted and fused to obtain a distortion correction index of the airway image at the current moment under each distortion.

6. The artificial intelligence-based anesthesiology airway intubation image recognition method of claim 5, wherein, The determination of the distortion correction factor corresponding to each matched edge point comprises: Under each distortion, the distortion correction factors corresponding to each curvature in each preset direction are curve fitted to obtain a distortion correction factor curve of each distortion in each preset direction; Based on the curvature value of each matched edge point, a corresponding distortion correction factor is obtained in the corresponding distortion correction factor curve.

7. The artificial intelligence-based anesthesiology airway intubation image recognition method of claim 1, wherein, The airway correction image acquisition method comprises: The distortion correction index of the airway image is substituted into a Brown-Conrady distortion model to correct the airway image and obtain a primary correction image; The lens intrinsic distortion correction parameter is substituted into the Brown-Conrady distortion model to perform secondary correction on the primary correction image and obtain the airway correction image. 8.The anesthesia airway intubation image recognition method based on artificial intelligence according to claim 1, wherein, The feature point acquisition method comprises: Corner points in each airway model image are acquired by corner point detection and taken as feature points. 9.The anesthesiology airway intubation image recognition method based on artificial intelligence of claim 3, wherein, The neural network is a CNN.

Images