Gastric content estimation method based on intelligent image recognition

Abstract

The invention discloses a stomach content amount estimation method based on intelligent image recognition, and belongs to the technical field of stomach content amount estimation, and the method comprises the following steps: S1, collecting and processing a stomach content perception image of a patient, and determining a stomach content feature image of the patient; s2, determining an optimal stomach content amount estimation model according to the stomach content amount estimation demand; s3, predicting and analyzing the real-time image of the stomach content of the patient, estimating the stomach content amount of the patient, and determining an estimated value of the stomach content amount of the patient; and S4, carrying out intelligent management and control on the estimation of the content amount of the stomach of the patient. According to the method, the problems that the stomach content amount cannot be effectively estimated based on intelligent image recognition and timely management and control treatment cannot be provided for the patient in the prior art, so that the diagnosis and treatment effects of the patient are poor are solved. According to the method, the content of the stomach can be effectively estimated based on intelligent image recognition, timely management and control treatment can be provided for a patient, and the diagnosis and treatment effects of the patient can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of gastric content volume estimation, and in particular to a gastric content volume estimation method based on intelligent image recognition. Background Art

[0002] As the aging process of my country's population accelerates, the incidence of gastrointestinal diseases is increasing year by year. Accurate and rapid detection and estimation of gastric content volume is of great significance for the diagnosis and treatment of gastrointestinal diseases. Among them, gastric content volume estimation has important clinical significance for the diagnosis, treatment and rehabilitation evaluation of gastrointestinal diseases. The main reasons are as follows: 1) Diagnosis of gastrointestinal diseases: Estimation of gastric content can help doctors determine the severity of gastrointestinal diseases and changes in the condition. For example, patients with hyperacidity need to control gastric acid secretion, and estimation of gastric content can help doctors determine the patient's maximum gastric emptying rate and thus formulate an appropriate diet plan; 2) Evaluation of the efficacy of drug therapy: Gastric content estimation can be used to evaluate the efficacy of drug therapy. For example, some drugs need to be taken on an empty stomach, and gastric content estimation can help doctors determine whether the patient is in an empty stomach state, thereby evaluating the efficacy of drug therapy. 3) Rehabilitation assessment of gastrointestinal diseases: Gastric content estimation can be used to evaluate the recovery of gastrointestinal diseases. For example, for patients recovering after surgery, gastric content estimation can help doctors determine the patient's recovery progress and thus adjust the rehabilitation plan.

[0003] However, existing gastric content estimation methods cannot effectively estimate the gastric content volume based on intelligent image recognition, and cannot provide timely management and treatment for patients, resulting in poor diagnosis and treatment effects for patients. Summary of the invention

[0004] The purpose of the present invention is to provide a method for estimating the amount of gastric contents based on intelligent image recognition, which can effectively estimate the amount of gastric contents based on intelligent image recognition, provide timely management and treatment for patients, improve the diagnosis and treatment effects of patients, and solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: The method for estimating gastric content volume based on intelligent image recognition comprises the following steps: S1: acquiring a perceived image of the patient's stomach contents based on a gastric ultrasound examination, processing the perceived image of the patient's stomach contents based on the gastric ultrasound examination, and determining a characteristic image of the patient's stomach contents based on the gastric ultrasound examination; S2: According to the demand for gastric content volume estimation based on intelligent image recognition, a gastric content volume estimation model based on intelligent image recognition is constructed, and the gastric content volume estimation model based on intelligent image recognition is tested and optimized to determine the optimal gastric content volume estimation model based on intelligent image recognition; S3: Based on the optimal gastric content volume estimation model based on intelligent image recognition, predict and analyze the real-time image of the patient's gastric content based on gastric ultrasound examination, estimate the patient's gastric content volume, and determine the patient's gastric content volume estimation value based on intelligent image recognition; S4: Based on the standard value of the patient's gastric content volume, conduct a comparative analysis on the patient's gastric content volume estimation value, determine the patient's gastric content volume estimation analysis results based on intelligent image recognition, and based on the patient's gastric content volume estimation analysis results, formulate a patient's gastric content volume estimation control plan based on intelligent image recognition to perform intelligent control on the patient's gastric content volume estimation.

[0006] Preferably, in S1, a perceived image of the patient's stomach contents based on gastric ultrasound examination is collected, and the following operations are performed: According to the estimation requirements of gastric content volume based on intelligent image recognition, the patient lies on the examination bed, and the patient lies on the side or on the back according to the examination requirements; Based on gastric ultrasound examination technology, a transparent colloid is applied to the patient's abdomen so that the ultrasound probe fits the abdomen; Place the ultrasound probe on a specific area of ​​the patient's abdomen, move the ultrasound probe back and forth, observe the position of the patient's stomach, emit ultrasonic waves through the ultrasound probe to detect the patient's stomach contents, form a visual image of the patient's stomach, and collect the perceived image of the patient's stomach contents based on gastric ultrasound examination.

[0007] Preferably, in S1, the perceived image of the patient's stomach contents based on gastric ultrasound examination is processed by performing the following operations: Acquire a perceptual image of the patient's gastric contents based on gastric ultrasound examination; Based on the filter, a filter processing is performed on the perceived image of the patient's stomach contents based on gastric ultrasound examination; Under the condition of retaining the detailed features of the gastric content perception image of the patient based on gastric ultrasound examination, the noise of the gastric content perception image of the patient based on gastric ultrasound examination is suppressed, and the noise in the gastric content perception image of the patient based on gastric ultrasound examination is removed; Based on image segmentation technology, the brightness and contrast of the patient's gastric content perception image based on gastric ultrasound examination are adjusted, where The gastric content perception image of the patient based on gastric ultrasound examination is divided into regions, and the average brightness of each region is analyzed. When the average brightness of a region is too low, the brightness of the region is increased; when the average brightness of a region is too high, the brightness of the region is reduced; The brightness of the perceived image of the patient's gastric contents based on gastric ultrasound examination is made consistent through interpolation or reconstruction techniques; Based on the adaptive adjustment method of global contrast and neighboring pixel relationship, the gray level difference of the pixels around each pixel in the patient's gastric content perception image based on gastric ultrasound examination is calculated; The contrast of the pixel is automatically adjusted according to the grayscale difference, so as to determine a high-definition gastric content perception image of the patient based on gastric ultrasound examination.

[0008] Preferably, when the average brightness of a certain area is too low, the brightness of the area is increased, including: When the average brightness of a certain area is lower than a preset first brightness threshold, the area whose average brightness is lower than the preset first brightness threshold is taken as a first target area; Retrieving the current contrast value of the first target area; Retrieving an average brightness value of an image area connected to the first target area; Obtaining a first brightness adjustment amplitude ratio by using a current contrast value of the first target area and an average brightness value of an image area connected to the first target area; The first brightness adjustment amplitude ratio is obtained by the following formula:

[0009] Among them, U 01 represents the first brightness adjustment amplitude ratio; U0 represents the preset amplitude ratio reference value; n represents the number of image areas connected to the first target area; L i represents the brightness value of the i-th image area connected to the first target area; L m01 Indicates the brightness value of the first target area; L y01 represents the preset first brightness threshold; D represents the current contrast value of the first target area; D c Indicates the preset contrast reference value; The brightness of the first target area is enhanced according to the first brightness adjustment amplitude ratio.

[0010] Preferably, when the average brightness of a certain area is too high, the brightness of the area is reduced; including: When the average brightness of a certain area exceeds a preset second brightness threshold, the area whose average brightness exceeds the preset second brightness threshold is taken as a second target area; Retrieving the current contrast value of the second target area; Retrieving an average brightness value of an image area connected to the second target area; Obtaining a second brightness adjustment amplitude ratio by using a current contrast value of the second target area and an average brightness value of an image area connected to the second target area; The second brightness adjustment amplitude ratio is obtained by the following formula:

[0011] Among them, U 02 represents the second brightness adjustment amplitude ratio; U0 represents the preset amplitude ratio reference value; m represents the number of image areas connected to the second target area; L j represents the brightness value of the jth image area connected to the second target area; L m02 Indicates the brightness value of the second target area; L y02 Indicates the preset second brightness threshold; D d Indicates the current contrast value of the second target area; D c Indicates the preset contrast reference value; The brightness of the second target area is reduced according to the second brightness adjustment amplitude ratio.

[0012] Preferably, in S1, the perceived image of the patient's stomach contents based on gastric ultrasound examination is processed, and the following operations are further performed: Obtain high-definition gastric ultrasound-based sensory images of the patient's gastric contents; Based on image enhancement technology, high-definition images of patients' gastric contents are enhanced; Based on the edge detection algorithm, the edge of the high-definition patient's gastric content perception image is identified, and the edge of the patient's gastric content perception image is detected by calculating the gradient value of the neighborhood pixel points, and the edge is enhanced to highlight the edge in the high-definition patient's gastric content perception image; Based on image cropping and magnification technology, feature extraction is performed on the patient's gastric content perception image; Extracting features that can reflect the estimation of gastric content volume based on intelligent image recognition; Determine the characteristic image of the patient's gastric contents based on gastric ultrasound examination.

[0013] Preferably, in S2, a gastric content estimation model based on intelligent image recognition is constructed, and the following operations are performed: Obtain characteristic images of the patient's gastric contents based on gastric ultrasound examination; Classify the characteristic images of the patient's gastric contents based on gastric ultrasound examination; Determine a gastric content volume estimation training set and a gastric content volume estimation test set; According to the demand for gastric content volume estimation based on intelligent image recognition, a neural network model framework suitable for gastric content volume estimation based on intelligent image recognition is selected; Based on the gastric content volume estimation training set, a neural network model framework suitable for gastric content volume estimation based on intelligent image recognition is trained; A gastric content estimation model based on intelligent image recognition was determined.

[0014] Preferably, in S2, the gastric content volume estimation model based on intelligent image recognition is tested and optimized, and the following operations are performed: Obtain a gastric content volume estimation model based on intelligent image recognition; Based on the gastric content volume estimation test set, a performance test and evaluation of the gastric content volume estimation model based on intelligent image recognition is performed to determine the performance test and evaluation results of the gastric content volume estimation model; According to the performance test evaluation results of the gastric content volume estimation model, the gastric content volume estimation model based on intelligent image recognition is mined and analyzed; Determine the parameter adjustment optimization scheme based on the gastric content volume estimation model; According to the parameter adjustment optimization scheme based on the gastric content volume estimation model, the parameters of the gastric content volume estimation model based on intelligent image recognition are adjusted and optimized; Determine the optimal gastric content estimation model based on intelligent image recognition.

[0015] Preferably, in S3, based on the optimal gastric content volume estimation model based on intelligent image recognition, a real-time image of the patient's gastric content based on gastric ultrasound examination is predicted and analyzed, and the following operations are performed: According to the need for gastric content volume estimation based on intelligent image recognition, real-time images of the patient's gastric contents are collected based on gastric ultrasound examination; The real-time image of the patient's gastric contents based on gastric ultrasound examination is input into the optimal gastric contents volume estimation model based on intelligent image recognition; Based on the optimal gastric content volume estimation model based on intelligent image recognition, the real-time images of the patient's gastric content based on gastric ultrasound examination are predicted and analyzed to estimate the patient's gastric content volume; Determine the estimated value of the patient's gastric content volume based on intelligent image recognition.

[0016] Preferably, in S4, the estimated value of the patient's gastric content volume is compared and analyzed, and the following operations are performed: Obtain an estimate of the patient's gastric content volume based on intelligent image recognition; Based on the standard values ​​of gastric content volume of patients, the estimated values ​​of gastric content volume of patients were compared and analyzed; Determine the estimated analysis results of the patient's gastric content volume based on intelligent image recognition; Based on the patient's gastric content volume estimation and analysis results, formulate a patient gastric content volume estimation and control plan based on intelligent image recognition; Intelligently control the estimation of the patient's gastric content volume according to the patient's gastric content volume estimation and control scheme based on intelligent image recognition; Among them, when the estimated value of the patient's gastric content volume is within the standard value range of the patient's gastric content volume, the patient's gastric content volume estimation analysis result based on intelligent image recognition is that the patient's gastric content volume is normal; When the estimated value of the patient's gastric content volume is not within the standard value range of the patient's gastric content volume, the patient's gastric content volume estimation analysis result based on intelligent image recognition is that the patient's gastric content volume is abnormal; When the volume of the patient's gastric contents is abnormal, an intelligent early warning will be issued to promptly guide medical staff to manage and treat the patient.

[0017] Compared with the prior art, the present invention has the following beneficial effects: The present invention collects perceived images of gastric contents of patients, processes the perceived images of gastric contents of patients, determines characteristic images of gastric contents of patients, constructs a gastric contents volume estimation model based on intelligent image recognition according to the gastric contents volume estimation demand based on intelligent image recognition, tests and optimizes the gastric contents volume estimation model, determines an optimal gastric contents volume estimation model, predicts and analyzes the real-time images of gastric contents of patients based on the optimal gastric contents volume estimation model, estimates the gastric contents volume of patients, determines the estimated value of gastric contents volume of patients, compares and analyzes the estimated value of gastric contents volume of patients based on the standard value of gastric contents volume of patients, determines the analysis result of gastric contents volume estimation of patients, formulates the management and control plan of gastric contents volume estimation of patients based on the analysis result of gastric contents volume estimation of patients, performs intelligent management and control on the estimation of gastric contents volume of patients, can effectively estimate the gastric contents volume based on intelligent image recognition, can provide timely management and control treatment for patients, and can improve the diagnosis and treatment effects of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of the method for estimating the amount of gastric contents based on intelligent image recognition of the present invention; Figure 2 This is an algorithm diagram for comparative analysis of the patient's gastric content volume estimation value based on the patient's gastric content volume standard value according to the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] In order to solve the problem that the existing gastric content estimation method cannot effectively estimate the gastric content volume based on intelligent image recognition, cannot provide timely management and treatment for patients, and leads to poor diagnosis and treatment effects for patients, please refer to Figure 1-Figure 2 , this embodiment provides the following technical solutions: The method for estimating gastric content volume based on intelligent image recognition comprises the following steps: S1: acquiring a perceived image of the patient's stomach contents based on a gastric ultrasound examination, processing the perceived image of the patient's stomach contents based on the gastric ultrasound examination, and determining a characteristic image of the patient's stomach contents based on the gastric ultrasound examination; In this embodiment, the patient's gastric content perception image based on gastric ultrasound examination is acquired, and the following operations are performed: According to the estimation requirements of gastric content volume based on intelligent image recognition, the patient lies on the examination bed, and the patient lies on the side or on the back according to the examination requirements; Based on gastric ultrasound examination technology, a transparent colloid is applied to the patient's abdomen so that the ultrasound probe fits the abdomen; Place the ultrasound probe on a specific area of ​​the patient's abdomen, move the ultrasound probe back and forth, observe the position of the patient's stomach, emit ultrasonic waves through the ultrasound probe to detect the patient's stomach contents, form a visual image of the patient's stomach, and collect the perceived image of the patient's stomach contents based on gastric ultrasound examination.

[0021] In this embodiment, the patient's gastric content perception image based on gastric ultrasound examination is processed by performing the following operations: Acquire a perceptual image of the patient's gastric contents based on gastric ultrasound examination; Based on the filter, a filter processing is performed on the perceived image of the patient's stomach contents based on gastric ultrasound examination; Under the condition of retaining the detailed features of the gastric content perception image of the patient based on gastric ultrasound examination, the noise of the gastric content perception image of the patient based on gastric ultrasound examination is suppressed, and the noise in the gastric content perception image of the patient based on gastric ultrasound examination is removed; Based on image segmentation technology, the brightness and contrast of the patient's gastric content perception image based on gastric ultrasound examination are adjusted, where The gastric content perception image of the patient based on gastric ultrasound examination is divided into regions, and the average brightness of each region is analyzed. When the average brightness of a region is too low, the brightness of the region is increased; when the average brightness of a region is too high, the brightness of the region is reduced; The brightness of the perceived image of the patient's gastric contents based on gastric ultrasound examination is made consistent through interpolation or reconstruction techniques; Based on the adaptive adjustment method of global contrast and neighboring pixel relationship, the gray level difference of the pixels around each pixel in the patient's gastric content perception image based on gastric ultrasound examination is calculated; The contrast of the pixel is automatically adjusted according to the grayscale difference, so as to determine a high-definition gastric content perception image of the patient based on gastric ultrasound examination; Based on image enhancement technology, high-definition images of patients' gastric contents are enhanced; Based on the edge detection algorithm, the edge of the high-definition patient's gastric content perception image is identified, and the edge of the patient's gastric content perception image is detected by calculating the gradient value of the neighborhood pixel points, and the edge is enhanced to highlight the edge in the high-definition patient's gastric content perception image; Based on image cropping and magnification technology, feature extraction is performed on the patient's gastric content perception image; Extracting features that can reflect the estimation of gastric content volume based on intelligent image recognition; Determine the characteristic image of the patient's gastric contents based on gastric ultrasound examination.

[0022] Specifically, when the average brightness of a certain area is too low, the brightness of the area is increased, including: When the average brightness of a certain area is lower than a preset first brightness threshold, the area whose average brightness is lower than the preset first brightness threshold is taken as a first target area; Retrieving the current contrast value of the first target area; Retrieving an average brightness value of an image area connected to the first target area; Obtaining a first brightness adjustment amplitude ratio by using a current contrast value of the first target area and an average brightness value of an image area connected to the first target area; The first brightness adjustment amplitude ratio is obtained by the following formula:

[0023] Among them, U 01 represents the first brightness adjustment amplitude ratio; U0 represents the preset amplitude ratio reference value; n represents the number of image areas connected to the first target area; L i represents the brightness value of the i-th image area connected to the first target area; L m01Indicates the brightness value of the first target area; L y01 represents the preset first brightness threshold; D represents the current contrast value of the first target area; D c Indicates the preset contrast reference value; The brightness of the first target area is enhanced according to the first brightness adjustment amplitude ratio.

[0024] The technical effect of the above technical solution is that the solution can automatically identify the area with too low brightness in the image (i.e., the first target area) and perform brightness enhancement processing on it. This intelligent processing method can ensure that the overall brightness distribution of the image is more uniform and improve the visual experience. By considering the current contrast value (D) of the first target area and the preset contrast reference value (D c ), which can adjust the brightness more accurately. If the contrast of the first target area is low, a larger brightness adjustment amplitude may be required to improve the visual effect; conversely, if the contrast is already high, a smaller adjustment amplitude may be required to avoid over-enhancement. The scheme introduces the average brightness value (L i ), which helps to achieve a smoother brightness transition. By considering the brightness of adjacent areas, it is possible to avoid obvious brightness differences between the area after brightness enhancement and the surrounding areas, thereby maintaining the overall coherence of the image. The first brightness adjustment amplitude ratio (U 01 ), the solution can flexibly determine the amplitude of brightness adjustment based on multiple factors (including the brightness of adjacent areas, the brightness of the first target area, the preset brightness threshold and the contrast reference value, etc.). This flexibility enables the solution to adapt to different images and different brightness adjustment requirements. This technical solution can significantly improve the overall quality of the image by intelligently identifying areas with too low brightness and adjusting the brightness by comprehensively considering multiple factors such as contrast and brightness of adjacent areas. This helps to improve the visual effect of the image, making it clearer, easier to observe and understand.

[0025] In summary, this technical solution achieves precise adjustment of areas with low brightness in the image through an intelligent brightness adjustment method, combined with consideration of contrast and brightness of adjacent areas, thereby improving the overall quality of the image.

[0026] Specifically, when the average brightness of a certain area is too high, the brightness of the area is reduced; including: When the average brightness of a certain area exceeds a preset second brightness threshold, the area whose average brightness exceeds the preset second brightness threshold is taken as a second target area; Retrieving the current contrast value of the second target area; Retrieving an average brightness value of an image area connected to the second target area; Obtaining a second brightness adjustment amplitude ratio by using a current contrast value of the second target area and an average brightness value of an image area connected to the second target area; The second brightness adjustment amplitude ratio is obtained by the following formula:

[0027] Among them, U 02 represents the second brightness adjustment amplitude ratio; U0 represents the preset amplitude ratio reference value; m represents the number of image areas connected to the second target area; L j represents the brightness value of the jth image area connected to the second target area; L m02 Indicates the brightness value of the second target area; L y02 Indicates the preset second brightness threshold; D d Indicates the current contrast value of the second target area; D c Indicates the preset contrast reference value; The brightness of the second target area is reduced according to the second brightness adjustment amplitude ratio.

[0028] The technical effect of the above technical solution is that the solution can automatically identify the area with too high brightness in the image (i.e., the second target area) and reduce its brightness. This intelligent processing method helps to balance the overall brightness of the image, avoid the loss of details caused by overexposed areas, and improve the visual comfort of the image. By considering the current contrast value (D d ) and the preset contrast reference value (D c ), this scheme can more accurately control the brightness reduction. If the contrast of the second target area is high, a smaller brightness reduction may be required to maintain the details of the image; conversely, if the contrast is low, a larger adjustment may be required to avoid the image appearing too bland. The average brightness value (L j ), which helps to achieve a more natural brightness transition. By considering the brightness of adjacent areas, it can ensure that the brightness difference between the area after brightness reduction and the surrounding area is not too abrupt, thereby maintaining the overall coherence and naturalness of the image. The second brightness adjustment amplitude ratio (U 02), which can flexibly determine the extent of brightness reduction based on a variety of factors (including the brightness of adjacent areas, the brightness of the second target area, the preset brightness threshold and the contrast reference value, etc.). This flexibility enables the solution to adapt to different images and different brightness adjustment requirements, ensuring that the processed image is neither overexposed nor too dark. This technical solution can significantly improve the overall quality of the image by intelligently identifying areas with excessive brightness and reducing the brightness by comprehensively considering multiple factors such as contrast and brightness of adjacent areas. This helps to enhance the visual effect of the image, making it clearer, more delicate and easier to observe.

[0029] In summary, this technical solution achieves precise adjustment of areas with excessive brightness in the image through an intelligent brightness balance method, combined with considerations of contrast and brightness of adjacent areas, thereby improving the overall image quality and enhancing the visual experience.

[0030] S2: According to the demand for gastric content volume estimation based on intelligent image recognition, a gastric content volume estimation model based on intelligent image recognition is constructed, and the gastric content volume estimation model based on intelligent image recognition is tested and optimized to determine the optimal gastric content volume estimation model based on intelligent image recognition; In this embodiment, a gastric content estimation model based on intelligent image recognition is constructed, and the following operations are performed: Obtain characteristic images of the patient's gastric contents based on gastric ultrasound examination; Classify the characteristic images of the patient's gastric contents based on gastric ultrasound examination; Determine a gastric content volume estimation training set and a gastric content volume estimation test set; According to the demand for gastric content volume estimation based on intelligent image recognition, a neural network model framework suitable for gastric content volume estimation based on intelligent image recognition is selected; Based on the gastric content volume estimation training set, a neural network model framework suitable for gastric content volume estimation based on intelligent image recognition is trained; A gastric content estimation model based on intelligent image recognition was determined.

[0031] In this embodiment, the gastric content estimation model based on intelligent image recognition is tested and optimized, and the following operations are performed: Obtain a gastric content volume estimation model based on intelligent image recognition; Based on the gastric content volume estimation test set, a performance test and evaluation of the gastric content volume estimation model based on intelligent image recognition is performed to determine the performance test and evaluation results of the gastric content volume estimation model; According to the performance test evaluation results of the gastric content volume estimation model, the gastric content volume estimation model based on intelligent image recognition is mined and analyzed; Determine the parameter adjustment optimization scheme based on the gastric content volume estimation model; According to the parameter adjustment optimization scheme based on the gastric content volume estimation model, the parameters of the gastric content volume estimation model based on intelligent image recognition are adjusted and optimized; Determine the optimal gastric content estimation model based on intelligent image recognition.

[0032] S3: Based on the optimal gastric content volume estimation model based on intelligent image recognition, predict and analyze the real-time image of the patient's gastric content based on gastric ultrasound examination, estimate the patient's gastric content volume, and determine the patient's gastric content volume estimation value based on intelligent image recognition; In this embodiment, based on the optimal gastric content volume estimation model based on intelligent image recognition, a real-time image of the patient's gastric content based on gastric ultrasound examination is predicted and analyzed, and the following operations are performed: According to the need for gastric content volume estimation based on intelligent image recognition, real-time images of the patient's gastric contents are collected based on gastric ultrasound examination; The real-time image of the patient's gastric contents based on gastric ultrasound examination is input into the optimal gastric contents volume estimation model based on intelligent image recognition; Based on the optimal gastric content volume estimation model based on intelligent image recognition, the real-time images of the patient's gastric content based on gastric ultrasound examination are predicted and analyzed to estimate the patient's gastric content volume; Determine the estimated value of the patient's gastric content volume based on intelligent image recognition.

[0033] S4: Based on the standard value of the patient's gastric content volume, conduct a comparative analysis on the patient's gastric content volume estimation value, determine the patient's gastric content volume estimation analysis results based on intelligent image recognition, and based on the patient's gastric content volume estimation analysis results, formulate a patient's gastric content volume estimation control plan based on intelligent image recognition to perform intelligent control on the patient's gastric content volume estimation.

[0034] In this embodiment, the estimated values ​​of the patient's gastric content volume are compared and analyzed, and the following operations are performed: Obtain an estimate of the patient's gastric content volume based on intelligent image recognition; Based on the standard values ​​of gastric content volume of patients, the estimated values ​​of gastric content volume of patients were compared and analyzed; Determine the estimated analysis results of the patient's gastric content volume based on intelligent image recognition; Based on the patient's gastric content volume estimation and analysis results, formulate a patient gastric content volume estimation and control plan based on intelligent image recognition; Intelligently control the estimation of the patient's gastric content volume according to the patient's gastric content volume estimation and control scheme based on intelligent image recognition; Among them, when the estimated value of the patient's gastric content volume is within the standard value range of the patient's gastric content volume, the patient's gastric content volume estimation analysis result based on intelligent image recognition is that the patient's gastric content volume is normal; When the estimated value of the patient's gastric content volume is not within the standard value range of the patient's gastric content volume, the patient's gastric content volume estimation analysis result based on intelligent image recognition is that the patient's gastric content volume is abnormal; When the volume of the patient's gastric contents is abnormal, an intelligent early warning will be issued to promptly guide medical staff to manage and treat the patient.

[0035] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0036] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for estimating gastric content based on intelligent image recognition, characterized in that: The steps include: S1: acquiring a perceived image of the patient's stomach contents based on a gastric ultrasound examination, processing the perceived image of the patient's stomach contents based on a gastric ultrasound examination, and determining a characteristic image of the patient's stomach contents based on a gastric ultrasound examination; S2: According to the demand for gastric content volume estimation based on intelligent image recognition, a gastric content volume estimation model based on intelligent image recognition is constructed, and the gastric content volume estimation model based on intelligent image recognition is tested and optimized to determine the optimal gastric content volume estimation model based on intelligent image recognition; S3: Based on the optimal gastric content volume estimation model based on intelligent image recognition, predict and analyze the real-time image of the patient's gastric content based on gastric ultrasound examination, estimate the patient's gastric content volume, and determine the patient's gastric content volume estimation value based on intelligent image recognition; S4: Based on the standard value of the patient's gastric content volume, conduct a comparative analysis on the patient's gastric content volume estimation value, determine the patient's gastric content volume estimation analysis results based on intelligent image recognition, and based on the patient's gastric content volume estimation analysis results, formulate a patient's gastric content volume estimation control plan based on intelligent image recognition to perform intelligent control on the patient's gastric content volume estimation.

2. The method for estimating gastric content based on intelligent image recognition according to claim 1, characterized in that: In S1, a gastric content perception image of a patient based on gastric ultrasound examination is acquired, and the following operations are performed: According to the estimation requirements of gastric content volume based on intelligent image recognition, the patient lies on the examination bed, and the patient lies on the side or on the back according to the examination requirements; Based on gastric ultrasound examination technology, a transparent colloid is applied to the patient's abdomen so that the ultrasound probe fits the abdomen; Place the ultrasound probe on a specific area of ​​the patient's abdomen, move the ultrasound probe back and forth, observe the position of the patient's stomach, emit ultrasonic waves through the ultrasound probe to detect the patient's stomach contents, form a visual image of the patient's stomach, and collect the perceived image of the patient's stomach contents based on gastric ultrasound examination.

3. The method for estimating gastric content based on intelligent image recognition according to claim 2, characterized in that: In S1, the patient's gastric content perception image based on gastric ultrasound examination is processed, and the following operations are performed: Acquire a perceptual image of the patient's gastric contents based on gastric ultrasound examination; Based on the filter, a filter processing is performed on the perceived image of the patient's stomach contents based on gastric ultrasound examination; Under the condition of retaining the detailed features of the gastric content perception image of the patient based on gastric ultrasound examination, the noise of the gastric content perception image of the patient based on gastric ultrasound examination is suppressed, and the noise in the gastric content perception image of the patient based on gastric ultrasound examination is removed; Based on image segmentation technology, the brightness and contrast of the patient's gastric content perception image based on gastric ultrasound examination are adjusted, where The gastric content perception image of the patient based on gastric ultrasound examination is divided into regions, and the average brightness of each region is analyzed. When the average brightness of a region is too low, the brightness of the region is increased; when the average brightness of a region is too high, the brightness of the region is reduced; The brightness of the perceived image of the patient's gastric contents based on gastric ultrasound examination is made consistent through interpolation or reconstruction techniques; Based on the adaptive adjustment method of global contrast and neighboring pixel relationship, the gray level difference of pixels around each pixel in the patient's gastric content perception image based on gastric ultrasound examination is calculated; The contrast of the pixel is automatically adjusted according to the grayscale difference, so as to determine a high-definition gastric content perception image of the patient based on gastric ultrasound examination.

4. The method for estimating gastric content based on intelligent image recognition according to claim 3, characterized in that: When the average brightness of an area is too low, increase the brightness of the area, including: When the average brightness of a certain area is lower than a preset first brightness threshold, the area whose average brightness is lower than the preset first brightness threshold is taken as a first target area; Retrieving the current contrast value of the first target area; Retrieving an average brightness value of an image area connected to the first target area; Obtaining a first brightness adjustment amplitude ratio by using a current contrast value of the first target area and an average brightness value of an image area connected to the first target area; The first brightness adjustment amplitude ratio is obtained by the following formula: Among them, U 01 represents the first brightness adjustment amplitude ratio; U0 represents the preset amplitude ratio reference value; n represents the number of image areas connected to the first target area; L i represents the brightness value of the i-th image area connected to the first target area; L m01 Indicates the brightness value of the first target area; L y01 represents the preset first brightness threshold; D represents the current contrast value of the first target area; D c Indicates the preset contrast reference value; The brightness of the first target area is enhanced according to the first brightness adjustment amplitude ratio.

5. The method for estimating gastric content based on intelligent image recognition according to claim 3, characterized in that: When the average brightness of an area is too high, reduce the brightness of the area; including: When the average brightness of a certain area exceeds a preset second brightness threshold, the area whose average brightness exceeds the preset second brightness threshold is taken as a second target area; Retrieving the current contrast value of the second target area; Retrieving an average brightness value of an image area connected to the second target area; Obtaining a second brightness adjustment amplitude ratio by using a current contrast value of the second target area and an average brightness value of an image area connected to the second target area; The second brightness adjustment amplitude ratio is obtained by the following formula: Among them, U 02 represents the second brightness adjustment amplitude ratio; U0 represents the preset amplitude ratio reference value; m represents the number of image areas connected to the second target area; L j represents the brightness value of the jth image area connected to the second target area; L m02 Indicates the brightness value of the second target area; L y02 Indicates the preset second brightness threshold; D d Indicates the current contrast value of the second target area; D c Indicates the preset contrast reference value; The brightness of the second target area is reduced according to the second brightness adjustment amplitude ratio.

6. The method for estimating gastric content based on intelligent image recognition according to claim 3, characterized in that: In S1, the perceived image of the patient's stomach contents based on gastric ultrasound examination is processed, and the following operations are performed: Obtain high-definition gastric ultrasound-based sensory images of the patient's gastric contents; Based on image enhancement technology, high-definition images of patients' gastric contents are enhanced; The edge of the high-definition patient's gastric content perception image is identified based on the edge detection algorithm, and the edge of the patient's gastric content perception image is detected by calculating the gradient value of the neighborhood pixel points, and the edge is enhanced to highlight the edge in the high-definition patient's gastric content perception image; Based on image cropping and magnification technology, feature extraction is performed on the patient's gastric content perception image; Extracting features that can reflect the estimation of gastric content volume based on intelligent image recognition; Determine the characteristic image of the patient's gastric contents based on gastric ultrasound examination.

7. The method for estimating gastric content based on intelligent image recognition according to claim 6, characterized in that: In S2, a gastric content estimation model based on intelligent image recognition is constructed, and the following operations are performed: Obtain characteristic images of the patient's gastric contents based on gastric ultrasound examination; Classify the characteristic images of the patient's gastric contents based on gastric ultrasound examination; Determine a gastric content volume estimation training set and a gastric content volume estimation test set; According to the demand for gastric content volume estimation based on intelligent image recognition, a neural network model framework suitable for gastric content volume estimation based on intelligent image recognition is selected; Based on the gastric content volume estimation training set, a neural network model framework suitable for gastric content volume estimation based on intelligent image recognition is trained; A gastric content estimation model based on intelligent image recognition was determined.

8. The method for estimating gastric content based on intelligent image recognition according to claim 7, characterized in that: In S2, the gastric content estimation model based on intelligent image recognition is tested and optimized, and the following operations are performed: Obtain a gastric content volume estimation model based on intelligent image recognition; Based on the gastric content volume estimation test set, a performance test and evaluation of the gastric content volume estimation model based on intelligent image recognition is performed to determine the performance test and evaluation results of the gastric content volume estimation model; According to the performance test evaluation results of the gastric content volume estimation model, the gastric content volume estimation model based on intelligent image recognition is mined and analyzed; Determine the parameter adjustment optimization scheme based on the gastric content volume estimation model; According to the parameter adjustment optimization scheme based on the gastric content volume estimation model, the parameters of the gastric content volume estimation model based on intelligent image recognition are adjusted and optimized; Determine the optimal gastric content estimation model based on intelligent image recognition.

9. The method for estimating gastric content based on intelligent image recognition according to claim 8, characterized in that: In S3, based on the optimal gastric content volume estimation model based on intelligent image recognition, a real-time image of the patient's gastric content based on gastric ultrasound examination is predicted and analyzed, and the following operations are performed: According to the need for gastric content volume estimation based on intelligent image recognition, real-time images of the patient's gastric contents are collected based on gastric ultrasound examination; The real-time image of the patient's gastric contents based on gastric ultrasound examination is input into the optimal gastric contents volume estimation model based on intelligent image recognition; Based on the optimal gastric content volume estimation model based on intelligent image recognition, the real-time images of the patient's gastric content based on gastric ultrasound examination are predicted and analyzed to estimate the patient's gastric content volume; Determine the estimated value of the patient's gastric content volume based on intelligent image recognition.

10. The method for estimating gastric content based on intelligent image recognition according to claim 9, characterized in that: In S4, the estimated values ​​of the patient's gastric contents are compared and analyzed, and the following operations are performed: Obtain an estimate of the patient's gastric content volume based on intelligent image recognition; Based on the standard values ​​of gastric contents of patients, the estimated values ​​of gastric contents of patients were compared and analyzed; Determine the estimated analysis results of the patient's gastric content volume based on intelligent image recognition; Based on the patient's gastric content volume estimation and analysis results, formulate a patient gastric content volume estimation and control plan based on intelligent image recognition; Intelligently control the estimation of the patient's gastric content volume according to the patient's gastric content volume estimation and control scheme based on intelligent image recognition; Among them, when the estimated value of the patient's gastric content volume is within the standard value range of the patient's gastric content volume, the patient's gastric content volume estimation analysis result based on intelligent image recognition is that the patient's gastric content volume is normal; When the estimated value of the patient's gastric content volume is not within the standard value range of the patient's gastric content volume, the patient's gastric content volume estimation analysis result based on intelligent image recognition is that the patient's gastric content volume is abnormal; When the volume of the patient's gastric contents is abnormal, an intelligent early warning will be issued to promptly guide medical staff to manage and treat the patient.

Images