Enteroscope attached water supply system and method

Abstract

The invention relates to the field of big data medical detection, and discloses an enteroscope attached water supply system which comprises a first water injection scheme storing user physical examination data and position data of an enteroscope; inputting the physical examination data of the user, the position data of the enteroscope and the image data of the enteroscope; outputting a first water injection scheme according to the physical examination data and the position data of the user; adjusting the first water injection scheme into a second water injection scheme based on the image data; and injecting water into the intestinal tract through the enteroscope in response to the second water injection regimen. The size and the amount of foam in the intestinal tract are judged through image data to adjust the opening time and the opening flow of water injection flow, so that it can be guaranteed that water is injected with the minimum total water injection rate, discomfort of a user is avoided, impurities such as foam can be flushed away in time, observation and efficiency of a detection doctor are improved, and the total detection time is shortened.

Description

Technical Field

[0001] The present invention relates to the field of big data medical detection, and in particular to a colonoscope-attached water system capable of controlling water flow in a relatively intelligent manner. Background Art

[0002] During colonoscopy, when bubbles appear in the intestine, the doctor may start a flushing operation to ensure a clear view and accurate examination. This flushing operation is an important part of colonoscopy, and the specific content includes the following aspects:

[0003] 1. Necessity of flushing operation

[0004] The foam in the intestine may be formed by gas, mucus or incompletely discharged intestinal contents. These foams will block the field of view of the colonoscope and affect the doctor's observation and diagnosis of the intestinal mucosa. Therefore, during the examination, once foam is found in the intestine, the doctor needs to flush it in time to remove the foam and ensure the smooth progress of the examination.

[0005] 2. Specific steps of flushing operation

[0006] Turn off the air injection button: Before performing the flushing operation, the doctor will first turn off the air injection button on the colonoscope to avoid continuing to inject gas into the intestine during the flushing process, thereby increasing the pressure and discomfort in the intestine.

[0007] Inject water through the biopsy hole: The doctor will then inject an appropriate amount of warm water into the intestine through the biopsy hole on the colonoscope. This warm water can flush the intestinal wall and wash away the foam and mucus attached to the intestinal mucosa.

[0008] Observation and adjustment: During the water injection process, the doctor will closely observe the changes in the intestine to ensure that the foam is effectively removed. If there is a lot of foam or it is difficult to remove, the doctor may increase the amount of water injected or adjust the injection speed to achieve the best effect.

[0009] 3. Precautions for flushing operation

[0010] Control the amount of water injected: During the flushing process, the doctor needs to strictly control the amount of water injected to avoid too much or too little. Too much water may cause excessive intestinal stretching and discomfort, while too little water may not effectively remove the foam.

[0011] Observe the patient's reaction: During the flushing process, the doctor also needs to closely observe the patient's reaction, such as abdominal pain, bloating and other discomfort symptoms. If abnormal conditions occur, the doctor needs to adjust the operation in time or take other measures to ensure the patient's safety.

[0012] Ensure a clear field of vision: The ultimate goal of the flushing operation is to ensure a clear field of vision and the accuracy of the examination. Therefore, during the operation, the doctor needs to constantly adjust and observe to ensure a clear field of vision.

[0013] IV. Conclusion

[0014] Flushing during colonoscopy is an important step to ensure the smooth progress of the examination and the accuracy of the diagnosis. By closing the air injection button, injecting water through the biopsy channel, observing and adjusting, and sucking out residual air and water, the doctor can effectively remove the foam and mucus in the intestine and provide a clear field of vision for subsequent intestinal examinations. At the same time, the doctor also needs to pay attention to controlling the amount of water injected, observing the patient's reaction, and ensuring a clear field of vision to ensure the safety and effectiveness of the examination.

[0015] The following is common practice in this field:

[0016] Depending on the position of the colonoscope insertion, several factors may affect the water injection flow rate:

[0017] 1. Water injection flow rate at the initial stage of colonoscope insertion

[0018] Initial stage: When the colonoscope is first inserted, a small amount of water is usually injected slowly and evenly to reveal the shape of the intestinal cavity and help the colonoscope enter the intestine smoothly. The water injection flow rate at this time is relatively small to avoid rapid expansion of the intestine and discomfort to the patient.

[0019] 2. Water injection flow rate when the colonoscope passes through a specific part

[0020] The junction of the sigmoid colon and the descending colon: When the colonoscope passes through the junction of the sigmoid colon and the descending colon, since this part is prone to bends and stenosis, in order to straighten the sigmoid colon bend and reduce the traction of the colon mesentery, the water injection volume may be increased to make the water flow to the lower descending colon, thereby helping the colonoscope to pass through this part more easily. At this time, the water injection flow rate may be relatively large.

[0021] Other narrow or curved parts: When the colonoscope passes through other narrow or curved intestinal parts, such as the splenic flexure and hepatic flexure, the doctor may also increase the water injection flow rate according to the specific situation to ensure that the colonoscope can pass through these parts smoothly.

[0022] 3. Water injection flow after the colonoscope penetrates into the intestine

[0023] In-depth observation stage: When the colonoscope goes deep into the intestine, in order to observe the intestinal mucosa and lesions more clearly, the doctor may reduce the amount of water injection appropriately to avoid excessive water interfering with the field of vision. At this time, the water injection flow rate will be relatively small to maintain the appropriate filling degree of the intestine.

[0024] 4. Individual Differences and Doctors’ Experience

[0025] Individual Differences:

[0026] 1. Increase the water injection volume

[0027] Long intestinal length: For patients with long intestinal length, in order to fully expose the intestinal mucosa and reduce blind spots, the doctor may increase the amount of water injection appropriately to ensure that the colonoscope can smoothly reach the distal end of the intestine.

[0028] Intestinal stenosis or lesions: When there is intestinal stenosis or lesions, such as polyps, tumors, etc., in order to observe the lesions more clearly and avoid missing them, the doctor may increase the amount of water injected to expand the intestine to make the lesions more prominent and easier to observe.

[0029] 2. Reduce the amount of water injection

[0030] Good intestinal cleanliness: If the patient's intestinal cleanliness is good and there is less intestinal residue, the doctor may reduce the amount of water injection to avoid excessive water interfering with the field of vision or causing discomfort to the patient.

[0031] Patients with poor tolerance: For patients with poor tolerance, such as the elderly and children, doctors may reduce the amount of water injected to avoid causing severe abdominal pain, bloating and other uncomfortable symptoms.

[0032] Doctors can make adjustments based on the above experience, but the doctor inserting the colonoscope to observe and control the amount of water injected will undoubtedly increase the difficulty of the inspection and the total insertion time. The longer the total insertion time, the more pain it will cause to the patient.

[0033] In other words, when flushing the colonoscope, it is necessary to maintain a reasonable amount of water, minimize the patient's discomfort symptoms such as abdominal pain and bloating, and ensure the clarity of the colonoscope's field of view and the accuracy of the examination. Therefore, controlling the timing and amount of water injection becomes the key to reflecting the level of a testing physician.

[0034] However, medical resources in my country are relatively scarce, and there are few high-level testing physicians, so many of the above-mentioned problems will occur during colonoscopy.

[0035] Therefore, there is a need to provide a colonoscope-attached water system that can more intelligently control the water flow. Summary of the invention

[0036] The technical problem to be solved by the present invention is to provide a water supply system for a colonoscope which can control the water flow more intelligently.

[0037] In a first aspect, the present invention provides a water supply system for a colonoscope, comprising:

[0038] A database storing a first water injection scheme corresponding to the user's physical examination data and the position data of the colonoscope;

[0039] An input module, which is used to input the user's physical examination data, the position data of the colonoscope, and the image data of the colonoscope;

[0040] A processing module, which is used to output a first water injection scheme according to the user's physical examination data and position data; and then adjust the first water injection scheme to a second water injection scheme based on the image data;

[0041] The water injection system injects water into the intestine through the enteroscope in response to the second water injection scheme.

[0042] In some embodiments, outputting a first water injection plan according to the user's physical examination data and location data includes:

[0043] Generate a basic scheme according to the age data in the physical examination data, wherein the basic scheme includes a maximum flow rate threshold and a maximum water volume threshold;

[0044] Determine whether the obesity index is in a normal weight interval, if not, determine whether the obesity index is in an obesity interval, if yes, reduce the maximum flow rate threshold and the maximum water volume threshold by a second percentage; otherwise, reduce the maximum flow rate threshold and the maximum water volume threshold by a first percentage;

[0045] The first water injection plan is generated according to the maximum flow rate value and the maximum water volume value.

[0046] In some embodiments, adjusting the first water injection scheme to a second water injection scheme based on the image data includes:

[0047] S321, obtaining the current position data of the colonoscope and the water injection amount data;

[0048] S322: output the real-time water injection flow rate pre-stored in the database according to the current position data of the colonoscope and the water injection amount data:

[0049] S323, dividing the image data into n*n number of sub-grid images;

[0050] S324, comparing each sub-grid image with the water injection image pre-stored in the database, and outputting the matching degree of the sub-grid image;

[0051] S325, constructing a rolling model, which is to construct a spherical model of a basic ball diameter and a groove of a basic radius, and cut out the cylindrical grooves that intersect with the basic radius in the spherical model to generate a spherical model of a basic size;

[0052] S326, judging whether the matching degree of one of the sub-network images of the abnormal frames of the image data exceeds the first threshold value, if it exceeds, outputting a rotation angle according to the matching degree of the largest sub-network image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model; if it does not exceed, forming a sub-field grid image with every 4 grids that can form a "field" shape in the image data, wherein adjacent sub-field grid images may have an intersection, dividing the water injection image pre-stored in the database into sub-water injection images with n / 2*n / 2 grids, obtaining the matching degree of each sub-field grid image and the sub-water injection image, judging whether the matching degree of the sub-field grid image of one of the abnormal frames exceeds the first threshold value, if so, outputting a rotation angle according to the matching degree of the largest sub-field grid image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model;

[0053] S327, determining whether the first flow rate is greater than the maximum flow rate threshold, if so, reducing the base radius by a third percentage, jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, reducing the number of grids n*n by a fourth percentage, and jumping to S323;

[0054] S328, determining whether the water volume data exceeds the maximum water volume threshold, if so, reducing the distance between the cylindrical model formed by the base radius and the spherical model by a fifth percentage, and jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, outputting the first flow rate at this time;

[0055] S329. Adjust the water injection flow rate of the colonoscope according to the first flow rate.

[0056] In some embodiments, the first percentage is 5% to 10%.

[0057] In some embodiments, the second percentage is 10% to 15%.

[0058] In a second aspect, the present invention provides a method for detecting water attached to a colonoscope, comprising:

[0059] A first water injection scheme corresponding to the user's physical examination data and the position data of the colonoscope is stored;

[0060] Input the user's physical examination data, colonoscope position data, and colonoscope image data;

[0061] Outputting a first water injection scheme according to the user's physical examination data and position data; and adjusting the first water injection scheme to a second water injection scheme based on the image data;

[0062] In response to the second water injection protocol, water is injected into the intestine through the enteroscope.

[0063] In some embodiments, outputting a first water injection plan according to the user's physical examination data and location data includes:

[0064] Generate a basic scheme according to the age data in the physical examination data, wherein the basic scheme includes a maximum flow rate threshold and a maximum water volume threshold;

[0065] Determine whether the obesity index is in a normal weight interval, if not, determine whether the obesity index is in an obesity interval, if yes, reduce the maximum flow rate threshold and the maximum water volume threshold by a second percentage; otherwise, reduce the maximum flow rate threshold and the maximum water volume threshold by a first percentage;

[0066] The first water injection plan is generated according to the maximum flow rate value and the maximum water volume value.

[0067] In some embodiments, adjusting the first water injection scheme to a second water injection scheme based on the image data includes:

[0068] S321, obtaining the current position data of the colonoscope and the water injection amount data;

[0069] S322: output the real-time water injection flow rate pre-stored in the database according to the current position data of the colonoscope and the water injection amount data:

[0070] S323, dividing the image data into n*n number of sub-grid images;

[0071] S324, comparing each sub-grid image with the water injection image pre-stored in the database, and outputting the matching degree of the sub-grid image;

[0072] S325, constructing a rolling model, which is to construct a spherical model of a basic ball diameter and a groove of a basic radius, and cut out the cylindrical grooves that intersect with the basic radius in the spherical model to generate a spherical model of a basic size;

[0073] S326, judging whether the matching degree of one of the sub-network images of the abnormal frames of the image data exceeds the first threshold value, if it exceeds, outputting a rotation angle according to the matching degree of the largest sub-network image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model; if it does not exceed, forming a sub-field grid image with every 4 grids that can form a "field" shape in the image data, wherein adjacent sub-field grid images may have an intersection, dividing the water injection image pre-stored in the database into sub-water injection images with n / 2*n / 2 grids, obtaining the matching degree of each sub-field grid image and the sub-water injection image, judging whether the matching degree of the sub-field grid image of one of the abnormal frames exceeds the first threshold value, if so, outputting a rotation angle according to the matching degree of the largest sub-field grid image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model;

[0074] S327, determining whether the first flow rate is greater than the maximum flow rate threshold, if so, reducing the base radius by a third percentage, jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, reducing the number of grids n*n by a fourth percentage, and jumping to S323;

[0075] S328, determining whether the water volume data exceeds the maximum water volume threshold, if so, reducing the distance between the cylindrical model formed by the base radius and the spherical model by a fifth percentage, and jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, outputting the first flow rate at this time;

[0076] S329. Adjust the water injection flow rate of the colonoscope according to the first flow rate.

[0077] In some embodiments, the first percentage is 5% to 10%.

[0078] In some embodiments, the second percentage is 10% to 15%.

[0079] The difference between the intestinal endoscope water system of the present invention and the prior art is that the intestinal endoscope water system of the present invention can first output a basic first water injection scheme for different user physical fitness and different intestinal endoscope insertion positions through the user's physical examination data and position data as the basis for water injection. Then, the size and amount of foam in the intestine are judged through image data to adjust the opening timing and flow rate of the water injection flow, thereby ensuring that the minimum total water injection volume is used for water injection to avoid user discomfort, and can also promptly flush out impurities such as foam, facilitate the observation and efficiency of the testing physician, and reduce the overall testing time.

[0080] The intestinal endoscope attached water system of the present invention is further described below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 It is a flow chart of the method of the water system attached to the enteroscope;

[0082] Figure 2 This is a demonstration diagram of the changes in the rolling model. DETAILED DESCRIPTION

[0083] like Figure 1 As shown, the present invention provides a water supply system for a colonoscope, comprising:

[0084] A database storing a first water injection scheme corresponding to the user's physical examination data and the position data of the colonoscope;

[0085] An input module, which is used to input the user's physical examination data, the position data of the colonoscope, and the image data of the colonoscope;

[0086] A processing module, which is used to output a first water injection scheme according to the user's physical examination data and position data; and then adjust the first water injection scheme to a second water injection scheme based on the image data;

[0087] The water injection system injects water into the intestine through the enteroscope in response to the second water injection scheme.

[0088] The present invention can first output a basic first water injection scheme for different user physical fitness and different intestinal endoscope insertion positions through the user's physical examination data and position data as the basis for water injection. Then, the size and amount of foam in the intestine are judged through image data to adjust the opening timing and flow rate of the water injection flow, thereby ensuring that the minimum total water injection volume is used to inject water and avoid user discomfort, and can also flush away impurities such as foam in time, which is convenient for the observation and efficiency of the testing physician and reduces the overall testing time.

[0089] Specific operation of water injection flow

[0090] Initial stage: When water injection begins, the flow rate can be slightly larger to quickly fill the intestines, but it should also be avoided to cause discomfort to the patient. Usually, the water injection flow rate in the initial stage may reach tens of milliliters to about 100 milliliters per minute.

[0091] Mid-stage: As the intestine gradually fills, the water injection flow rate needs to be gradually reduced to avoid excessive intestinal expansion and patient discomfort. At this stage, the flow rate may be adjusted to a dozen milliliters to several dozen milliliters per minute.

[0092] Ending stage: When the intestine is close to being full, the water injection flow rate needs to be further reduced, and even intermittent water injection can be used to ensure adequate cleaning of the intestine and the comfort of the patient.

[0093] This solution focuses on the control of opening and closing at the end stage, thereby reducing the problem of bubbles affecting observation. The above stages are determined by the number and position data of bubbles collected by the image data in each unit time period.

[0094] The user's physical examination data may be the user's body side data, which may include the user's height, weight, and age.

[0095] Among them, the position data of the colonoscope can be based on the scale on the colonoscope or other pressure sensors to know how deep the colonoscope has been inserted and the length of each section of the user's intestine, and to know which section of the intestine the end of the colonoscope is currently located in, so as to facilitate image data analysis according to different intestinal positions.

[0096] The image data of the colonoscope can be understood as the real-time image data collected after the colonoscope is inserted into the intestine.

[0097] The first water injection scheme may include a first maximum water injection volume and a first maximum water flow rate.

[0098] It should be noted that the first water injection plan is a first water injection plan generated based on corresponding data pre-stored in the database.

[0099] The second water injection scheme may include a second maximum water injection volume and a second maximum water flow rate.

[0100] It should be noted that the second water injection scheme is a second water injection scheme that is fine-tuned for the first water injection scheme according to the real-time changes of the image data.

[0101] In some embodiments, see Figure 1 , outputting a first water injection scheme according to the user's physical examination data and location data, including

[0102] Generate a basic scheme according to the age data in the physical examination data, wherein the basic scheme includes a maximum flow rate threshold and a maximum water volume threshold;

[0103] Determine whether the obesity index is in a normal weight interval, if not, determine whether the obesity index is in an obesity interval, if yes, reduce the maximum flow rate threshold and the maximum water volume threshold by a second percentage; otherwise, reduce the maximum flow rate threshold and the maximum water volume threshold by a first percentage;

[0104] The first water injection plan is generated according to the maximum flow rate value and the maximum water volume value.

[0105] In the present invention, the basic scheme corresponding to the age of the user pre-stored in the database is used as the basis for generating the first water injection scheme, and then the maximum water flow threshold and the maximum water volume threshold are adjusted according to the obesity index generated according to the height and weight of the user. When inserting the colonoscope, the testing physician can adjust the water flow size and the maximum water flow rate according to the physical examination data of different ages, heights, and weights. Avoid excessive intestinal extension and discomfort caused by too much water, and avoid severe abdominal pain, bloating and other uncomfortable symptoms caused by the elderly and children with poor tolerance. In addition, the obesity index determines the maximum flow rate threshold and the maximum water volume threshold that the user can bear to a certain extent. Therefore, adjusting the above thresholds based on the user's obesity level can improve the efficiency of the testing physician in detecting relatively obese users.

[0106] The present invention adjusts the maximum flow rate value and the maximum water volume value at different percentages according to different intervals of the obesity index, thereby providing different first water injection schemes for patients with different obesity levels, thereby maximizing the water injection efficiency of patients with good health, reducing the time for the user to insert the colonoscope, reducing pain, and improving efficiency. For users with poor health, while ensuring that the colonoscope is inserted as short as possible, the irritation during water injection is reduced, and adverse effects and risks are reduced.

[0107] The obesity index may be the BMI (Body Mass Index), which is an index obtained by dividing the weight in kilograms by the square of the height in meters, that is, the body mass index.

[0108] Body mass index (BMI) = weight (kg) ÷ height^2 (m)

[0109] For example: 70kg÷(1.75×1.75)=22.86, that is, the obesity index is 22.86.

[0110] Among them, the above basic plan can be an age-related plan, which can be shown in the following table:

[0111]

[0112] The maximum flow rate threshold can be understood as the maximum flow rate of water injection for the colonoscope that the person can withstand at this intestinal position. For example, a 20-year-old person is in good health and has a strong tolerance. In this case, we can inject water more quickly, thereby cleaning the feces and other impurities in the intestine through water injection, and on the other hand, allowing the intestine to have a certain amount of water injection to avoid bubbles affecting observation.

[0113] For example, the following table:

[0114]

[0115] The above basic solution is only an example of the maximum flow rate threshold and the maximum water volume value that match the patient's position data in the database. It is the basis we follow, and we will fine-tune it according to different obesity indexes before outputting it as the first water injection solution.

[0116] Among them, fine-tuning is performed according to different obesity indexes as follows:

[0117]

[0118]

[0119] That is, the first percentage is 5% to 10%, and the second percentage is 10% to 15%. The maximum flow rate threshold and the maximum water volume threshold are adjusted by the first percentage and the second percentage, so as to adjust the thresholds by using the user's obesity index to be more suitable for the size of each physical examination data. Preferably, for example, the first percentage is 5% and the second percentage is 10%.

[0120] The present invention adjusts the maximum flow rate threshold and the maximum water volume threshold to different degrees for people with different obesity indexes, so as to provide water injection schemes to different degrees for people with different obesity indexes.

[0121] In some embodiments, see Figure 1 , and then adjusting the first water injection scheme to a second water injection scheme based on the image data, including:

[0122] S321, obtaining the current position data of the colonoscope and the water injection amount data;

[0123] S322: output the real-time water injection flow rate pre-stored in the database according to the current position data of the colonoscope and the water injection amount data:

[0124] S323, dividing the image data into n*n number of sub-grid images;

[0125] S324, comparing each sub-grid image with the water injection image pre-stored in the database, and outputting the matching degree of the sub-grid image;

[0126] S325, constructing a rolling model, which is to construct a spherical model of a basic ball diameter and a groove of a basic radius, and cut out the cylindrical grooves that intersect with the basic radius in the spherical model to generate a spherical model of a basic size;

[0127] S326, judging whether the matching degree of one of the sub-network images of the abnormal frames of the image data exceeds the first threshold value, if it exceeds, outputting a rotation angle according to the matching degree of the largest sub-network image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model; if it does not exceed, forming a sub-field grid image with every 4 grids that can form a "field" shape in the image data, wherein adjacent sub-field grid images may have an intersection, dividing the water injection image pre-stored in the database into sub-water injection images with n / 2*n / 2 grids, obtaining the matching degree of each sub-field grid image and the sub-water injection image, judging whether the matching degree of the sub-field grid image of one of the abnormal frames exceeds the first threshold value, if so, outputting a rotation angle according to the matching degree of the largest sub-field grid image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model;

[0128] S327, determining whether the first flow rate is greater than the maximum flow rate threshold, if so, reducing the base radius by a third percentage, jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, reducing the number of grids n*n by a fourth percentage, and jumping to S323;

[0129] S328, determining whether the water volume data exceeds the maximum water volume threshold, if so, reducing the distance between the cylindrical model formed by the base radius and the spherical model by a fifth percentage, and jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, outputting the first flow rate at this time;

[0130] S329. Adjust the water injection flow rate of the colonoscope according to the first flow rate.

[0131] The present invention combines the above-mentioned real-time collected image data with the position data and the water injection amount data of the above-mentioned image data collected by the colonoscope, so as to obtain a general situation, and temporarily outputs the real-time water injection flow rate corresponding to this data in the database as a basis, and adjusts the size of this data through the image data, so that when it is time to start spraying water or when it is time to increase the water spraying amount, it automatically starts spraying water or increases the water spraying, thereby improving the efficiency of colonoscopy detection. Among them, the way to adjust the image data is to segment the real-time image data through the grid segmentation method, and use a single sub-grid image as a control object to compare it with the water injection image pre-stored in the database. If it is extremely similar or completely similar, that is, the matching degree is greater than the first threshold, water should be sprayed immediately at the first flow rate; however, this method also has defects, such as, an area that overlaps with the water injection image is located at the junction of four sub-grid images, that is, the four sub-grid images can be pieced together into an image with a roughly matching degree greater than the first value with the water injection image, but due to the segmentation of the grid, no sub-grid image has a matching degree greater than 1. Therefore, we adopt the method of constructing a rolling model, and let the matching degree of each sub-grid image affect the rolling angle of the rolling model, so as to adjust the water flow rate to achieve a first flow rate that matches the previous water flow rate, and directly output it when it meets the maximum flow rate threshold and the maximum water volume threshold, and control the water injection flow rate according to the conditions of meeting the maximum flow rate threshold and the maximum water volume threshold when it does not meet the maximum flow rate threshold and the maximum water volume threshold. That is, the water injection flow rate can be automatically adjusted in real time, making it easier for the testing physician to conduct efficient colonoscopy testing.

[0132] Among them, the current position data of the colonoscope can be understood as: the specific position of the user to be tested when the colonoscope is inserted. The position detection process can be obtained based on the existing technology and will not be repeated here. The collected current position data of the colonoscope may be: duodenum, upper jejunum, middle to lower jejunum, ileum, special cases (such as stenosis, adhesions, etc.). The general collection process is mainly to collect the position through ultrasound or calculate it according to the insertion depth or based on the experience of the testing physician.

[0133] The data on the amount of water injected can be understood as the sum of the amount of water injected into each section of the intestine.

[0134] Among them, S322, outputting the real-time water injection flow rate pre-stored in the database according to the position data of the current colonoscope and the data of the amount of water already injected, can be understood as: the real-time water injection flow rate corresponding to the position data and the data of the amount of water already injected of each previous colonoscope is pre-stored in the database, and only the position data and the data of the amount of water already injected of the colonoscope need to be input to output the real-time water injection flow rate of the colonoscope, thereby eliminating the need for the testing physician to adjust the size of the water flow through the intestine by himself during the process of inserting the colonoscope, thereby improving the detection efficiency of the colonoscope.

[0135] Among them, S323, dividing the image data into n*n sub-grid images, can be understood as: since there are many details in the image data that are unnecessary to observe, such as intestinal texture images, and there are many that need to be observed, such as polyps, ulcers, inflammation, tumors and other pathological images, and often such pathological images will not fill the entire image data, that is, they will not fill all the collected pictures, but only present the shape of a special pathological image in a local area.

[0136] Among them, S324, comparing each sub-grid image with the water injection image pre-stored in the database, and outputting the matching degree of the sub-grid image, can be understood as: Then, by splitting the image data into n*n sub-grid images in an n*n grid. Each sub-grid image is individually compared with the sub-water injection image that is also split from the water injection image as the lesion image through the n*n grid, so as to obtain the matching degree of this sub-grid image. It should be noted that the matching degree of this sub-grid image is the highest matching degree obtained after comparing one by one with all the water injection images pre-stored in the database that are split into n*n sub-water injection images according to the n*n grid. By comparing in this way, noise reduction and impurity removal can be performed on other positions of the image data, so as to retrieve relatively matching image data.

[0137] The output matching degree should be greater than the noise reduction threshold, that is, only a very small match means no match, and no output is needed, which is convenient for subsequent algorithm processing. The noise reduction threshold can be 20% to 70%, preferably 50%. In other words, only more than half of the matches will output the matching degree of this sub-grid image, so as to perform noise reduction on the matching degree.

[0138] Among them, the algorithm for matching the sub-grid image with the sub-watered image can be existing technology, and the output matching result can be 0-100%, which can be similar to the similarity comparison method between ordinary human faces and celebrity faces, or the image matching method of face recognition and fingerprint recognition.

[0139] Among them, S325, constructing a rolling model, which is to construct a spherical model of a basic ball diameter and a groove of a basic radius, cut out the cylindrical grooves that intersect with the basic radius in the spherical model, and generate a spherical model of a basic size. It can be understood that: in order to make the first flow velocity affected by the matching degree have the effect of pulling up and limiting the lowering, the present invention constructs a rolling model. This rolling model is as follows Figure 2 The shown is a linear sphere with an arc-shaped groove, whose groove is rotated from facing forward to facing left, that is, the maximum rotation is 90°. Then, Figure 2The proportion of the area of ​​the groove represented by the shadow in the middle to the area of ​​the circle corresponding to the sphere increases first and then decreases as the ball rolls, meeting the requirements of pulling up and limiting down. That is, the change is large for some matches, and the changes are not large for almost matches and complete matches, which makes the change trend of the influence of the matching degree of the sub-grid data more in line with the detection needs.

[0140] Among them, S326, judging whether the matching degree of one of the sub-network images of the product abnormal frames of the image data exceeds the first threshold, can be understood as: the abnormal frame can be each frame of the image data, or a frame that changes more than 5% from the previous frame, or the abnormal frame is a frame with a frame similarity of less than 95% with the image data of the previous second, that is, when the colonoscope is inserted inwardly, since the colonoscope is not inserted continuously, when it stops inserting, the change between the two frames of image data before and after is less than 5%, then, there is no need to judge and calculate each frame, thereby reducing computing pressure and computer load.

[0141] Then, for multiple continuous image data, we only extract a part of the abnormal frames for judgment and processing, and match the sub-network images split into this abnormal frame to obtain the matching degree, and compare it with the first threshold. The first threshold can be 80%, that is, whether the maximum matching degree of the sub-water injection image split according to the n*n grid with the water injection image pre-stored in the database exceeds 80%, so as to understand whether it is time to inject water and perform corresponding operations.

[0142] Among them, if it exceeds, the rotation angle is output according to the matching degree of the largest sub-network image, and the rolling model is rotated according to the rotation angle, and the real-time water injection flow is adjusted to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model. It can be understood as: if the matching degree exceeds 80%, the water injection should be started immediately, but the water injection amplitude of matching 80% and matching 100% should be different, or in other words, when the matching degree is 100%, the first flow rate should be increased to the maximum, and when the matching degree is 80%, the first flow rate can be reduced. So, how do we achieve the operation of pulling up and limiting between the matching degrees of 80% and 100%? It depends on the rotation angle of this rolling model, that is, input a steadily increasing matching degree of 80% to 100%, and output a first flow rate that pulls up and limits. For example, the following table:

[0143] Match Roll Angle Proportion First flow rate 80％ 0° 21.5％ 4ml / s 85％ 22.5° 19.9％ 4.428ml / s 90％ 45° 15.2％ 5.744ml / s 95％ 67.5° 8.2％ 7.704ml / s 100％ 90° 0％ 10ml / s

[0144] The algorithm of the present invention for deriving the matching degree and proportion through the above-mentioned sinusoidal relationship can be derived from the actual image. It is similar to the derivation formula of the initialization proportion and the sinusoidal image, which will not be described in detail here, but is only the calculation result of the image data.

[0145] The relationship between the proportion generated by the straight-line algorithm in the present invention and the first flow rate can be expressed by the following formula: First flow rate = -0.28 * proportion + 10.

[0146] Among them, if it does not exceed, every 4 sub-grid images that can form a "field" shape in the image data are combined to form a sub-field grid image. It can be understood that: when the matching degree does not exceed 80%, it is still possible that the matching water injection image appears at the intersection of 4 sub-grid images. Therefore, it does not meet the requirements. Then, the density of this grid is halved, that is, the four sub-grid images adjacent in a "field" shape are used as a sub-field grid image to compare with the sub-water injection image with the same grid density halved, so as to find the water injection image in the middle of the intersection of the four sub-grid images, which is convenient to start the water injection action, that is, adjust to the first flow rate.

[0147] Among them, adjacent sub-field grid images can have an intersection. It can be understood that: the first sub-field grid image on the left and the sub-field grid image adjacent to it on the right have two identical sub-grid images; the same is true for its lower part.

[0148] Among them, the water injection images pre-stored in the database are divided into sub-water injection images with a (n / 2) * (n / 2) grid. That is to say, on the premise that the sub-water injection images remain unchanged, the grid is expanded horizontally by one time and vertically by one time, so as to obtain a larger sub-water injection image for re-matching with the sub-field grid image.

[0149] Among them, the matching degree between each sub-field grid image and the sub-water injection image is obtained. The method for calculating the matching degree is the prior art.

[0150] Among them, it is judged whether the matching degree of one of the sub-field grid images of the abnormal frame exceeds the first threshold. It can be understood that: this time, it is judged whether the maximum matching degree of the larger sub-field grid image exceeds 80%.

[0151] Among them, if so, the rotation angle is output according to the matching degree of the largest sub-field grid image, the rolling model is rotated according to the rotation angle, and the real-time water injection flow rate is adjusted to increase to the first flow rate according to the proportion of the projection of the groove in the corresponding circle of the spherical model. It can be understood that: see Figure 2 as shown. Figure 2 On the left is 0 degrees, in the middle is the 90-degree limit, and on the right is the 180-degree rolling angle situation that does not need to be considered to avoid bugs. Specifically, when the matching degree is 80% and the rolling angle is 0 degrees, the shadow area generated by the groove occupies the area of the corresponding circle of the ball to the maximum extent. At this time, the proportion is 21.5%, and the corresponding first flow rate is 4 ml / s. And so on, see the above table.

[0152] Among them, it is determined whether the first flow rate is greater than the maximum flow rate threshold. If it is greater, the basic radius is reduced by a third percentage and the process jumps to S325. It can be understood that the third percentage can be 5%, that is, if it exceeds the maximum flow rate threshold, the first flow rate is reduced by 5%.

[0153] Among them, the number of jumps is recorded. When the number of jumps exceeds the third threshold, the number of grids n*n is reduced by the fourth percentage, and jumps to S323. It can be understood that if the number of times the first flow rate is reduced is still too many, then the first flow rate will no longer be reduced, but it is determined that an erroneous operation may occur, thereby reducing n by the fourth percentage. The fourth percentage can be 10%, that is, the original n is 10, that is, the original grid is 10*10, and after reducing the fourth percentage, it becomes a 9*9 grid. For special cases, the reduced n is rounded and obtained in a rounded manner, that is, the original n is 3, and it is reduced by 10% in a rounded manner, so that n is 2. The present invention recalculates the above steps from S323 through image data divided by a smaller number of grids n*n and a smaller number of grids (n / 2)*(n / 2) grids, so as to avoid the first flow rate exceeding the maximum threshold.

[0154] The third threshold may be 3 times, that is, after n is reduced by 10% for three times, the number of grids is reduced again.

[0155] Among them, it is judged whether the water injection volume data exceeds the maximum water volume threshold. If so, the distance between the cylindrical and spherical models formed by the basic radius is reduced by the fifth percentage, and the process jumps to S325, and the number of jumps is recorded. When the number of jumps exceeds the third threshold, the number of grids n*n is reduced by the fourth percentage, and the process jumps to S323. It can be understood that if the water injection volume has exceeded the maximum water injection volume, water can still be injected, but the water flow rate should be reduced, that is, the first flow rate should be reduced, so as not to cause greater pain to the user. Then, collect the water volume data of the water pump. When the water volume data exceeds the maximum water volume threshold, reduce the distance between the cylinder formed by the basic radius and the spherical model by the fifth percentage, which can be 10%. The above groove is obtained by cutting a cylinder stretched by the basic radius in the spherical model with a first distance. If the first distance is reduced by 10%, the groove will inevitably expand. Since the first flow rate obtained by the above rolling model has the effect of pulling up and limiting the lower limit, it will reduce the first flow rate to a large extent, and still keep the first flow rate adaptively changing, that is, jump to s325 and recalculate the first flow rate. However, if the above first distance decreases too many times, that is, more than 3 times, the first flow rate at this time should be output. In other words, it is reduced to a fixed value at this time, and the first flow rate is no longer affected by any image data. That is, it is determined to be a fault or an image data acquisition error at this time, and the first flow rate is no longer changed, but a fixed value is output.

[0156] In some embodiments, see Figure 1 , the first percentage is 5% to 10%.

[0157] In some embodiments, see Figure 1 , the second percentage is 10% to 15%.

[0158] like Figure 1 As shown, the present invention provides a method for detecting water attached to a colonoscope, comprising:

[0159] A first water injection scheme corresponding to the user's physical examination data and the position data of the colonoscope is stored;

[0160] Input the user's physical examination data, colonoscope position data, and colonoscope image data;

[0161] Outputting a first water injection scheme according to the user's physical examination data and position data; and adjusting the first water injection scheme to a second water injection scheme based on the image data;

[0162] In response to the second water injection protocol, water is injected into the intestine through the enteroscope.

[0163] The present invention can first output a basic first water injection scheme for different user physical fitness and different intestinal endoscope insertion positions through the user's physical examination data and position data as the basis for water injection. Then, the size and amount of foam in the intestine are judged through image data to adjust the opening timing and flow rate of the water injection flow, thereby ensuring that the minimum total water injection volume is used to inject water and avoid user discomfort, and can also flush away impurities such as foam in time, which is convenient for the observation and efficiency of the testing physician and reduces the overall testing time.

[0164] In some embodiments, see Figure 1 , outputting a first water injection scheme according to the user's physical examination data and location data, including

[0165] Generate a basic scheme according to the age data in the physical examination data, wherein the basic scheme includes a maximum flow rate threshold and a maximum water volume threshold;

[0166] Determine whether the obesity index is in a normal weight interval, if not, determine whether the obesity index is in an obesity interval, if yes, reduce the maximum flow rate threshold and the maximum water volume threshold by a second percentage; otherwise, reduce the maximum flow rate threshold and the maximum water volume threshold by a first percentage;

[0167] The first water injection plan is generated according to the maximum flow rate value and the maximum water volume value.

[0168] In the present invention, the basic scheme corresponding to the age of the user pre-stored in the database is used as the basis for generating the first water injection scheme, and then the maximum water flow threshold and the maximum water volume threshold are adjusted according to the obesity index generated according to the height and weight of the user. When inserting the colonoscope, the testing physician can adjust the water flow size and the maximum water flow rate according to the physical examination data of different ages, heights, and weights. Avoid excessive intestinal extension and discomfort caused by too much water, and avoid severe abdominal pain, bloating and other uncomfortable symptoms caused by the elderly and children with poor tolerance. In addition, the obesity index determines the maximum flow rate threshold and the maximum water volume threshold that the user can bear to a certain extent. Therefore, adjusting the above thresholds based on the user's obesity level can improve the efficiency of the testing physician in detecting relatively obese users.

[0169] The present invention adjusts the maximum flow rate value and the maximum water volume value at different percentages according to different intervals of the obesity index, thereby providing different first water injection schemes for patients with different obesity levels, thereby maximizing the water injection efficiency of patients with good health, reducing the time for the user to insert the colonoscope, reducing pain, and improving efficiency. For users with poor health, while ensuring that the colonoscope is inserted as short as possible, the irritation during water injection is reduced, and adverse effects and risks are reduced.

[0170] In some embodiments, see Figure 1 , and then adjusting the first water injection scheme to a second water injection scheme based on the image data, including:

[0171] S321, obtaining the current position data of the colonoscope and the water injection amount data;

[0172] S322: output the real-time water injection flow rate pre-stored in the database according to the current position data of the colonoscope and the water injection amount data:

[0173] S323, dividing the image data into n*n number of sub-grid images;

[0174] S324, comparing each sub-grid image with the water injection image pre-stored in the database, and outputting the matching degree of the sub-grid image;

[0175] S325, constructing a rolling model, which is to construct a spherical model of a basic ball diameter and a groove of a basic radius, and cut out the cylindrical grooves that intersect with the basic radius in the spherical model to generate a spherical model of a basic size;

[0176] S326, judging whether the matching degree of one of the sub-network images of the abnormal frames of the image data exceeds the first threshold value, if it exceeds, outputting a rotation angle according to the matching degree of the largest sub-network image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model; if it does not exceed, forming a sub-field grid image with every 4 grids that can form a "field" shape in the image data, wherein adjacent sub-field grid images may have an intersection, dividing the water injection image pre-stored in the database into sub-water injection images with n / 2*n / 2 grids, obtaining the matching degree of each sub-field grid image and the sub-water injection image, judging whether the matching degree of the sub-field grid image of one of the abnormal frames exceeds the first threshold value, if so, outputting a rotation angle according to the matching degree of the largest sub-field grid image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model;

[0177] S327, determining whether the first flow rate is greater than the maximum flow rate threshold, if so, reducing the base radius by a third percentage, jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, reducing the number of grids n*n by a fourth percentage, and jumping to S323;

[0178] S328, determining whether the water volume data exceeds the maximum water volume threshold, if so, reducing the distance between the cylindrical model formed by the base radius and the spherical model by a fifth percentage, and jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, outputting the first flow rate at this time;

[0179] S329. Adjust the water injection flow rate of the colonoscope according to the first flow rate.

[0180] The present invention combines the above-mentioned real-time collected image data with the position data and the water injection amount data of the above-mentioned image data collected by the colonoscope, so as to obtain a general situation, and temporarily outputs the real-time water injection flow rate corresponding to this data in the database as a basis, and adjusts the size of this data through the image data, so that when it is time to start spraying water or when it is time to increase the water spraying amount, it automatically starts spraying water or increases the water spraying, thereby improving the efficiency of colonoscopy detection. Among them, the way to adjust the image data is to segment the real-time image data through the grid segmentation method, and use a single sub-grid image as a control object to compare it with the water injection image pre-stored in the database. If it is extremely similar or completely similar, that is, the matching degree is greater than the first threshold, water should be sprayed immediately at the first flow rate; however, this method also has defects, such as, an area that overlaps with the water injection image is located at the junction of four sub-grid images, that is, the four sub-grid images can be pieced together into an image with a roughly matching degree greater than the first value with the water injection image, but due to the segmentation of the grid, no sub-grid image has a matching degree greater than 1. Therefore, we adopt the method of constructing a rolling model, and let the matching degree of each sub-grid image affect the rolling angle of the rolling model, so as to adjust the water flow rate to achieve a first flow rate that matches the previous water flow rate, and directly output it when it meets the maximum flow rate threshold and the maximum water volume threshold, and control the water injection flow rate according to the conditions of meeting the maximum flow rate threshold and the maximum water volume threshold when it does not meet the maximum flow rate threshold and the maximum water volume threshold. That is, the water injection flow rate can be automatically adjusted in real time, making it easier for the testing physician to conduct efficient colonoscopy testing.

[0181] In some embodiments, see Figure 1 , the first percentage is 5% to 10%.

[0182] In some embodiments, see Figure 1 , the second percentage is 10% to 15%.

[0183] The embodiments described above are merely descriptions of preferred implementation modes of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A water supply system for a colonoscope, characterized in that: include A database storing a first water injection scheme corresponding to the user's physical examination data and the position data of the colonoscope; An input module, which is used to input the user's physical examination data, the position data of the colonoscope, and the image data of the colonoscope; A processing module, which is used to output a first water injection scheme according to the user's physical examination data and position data; and then adjust the first water injection scheme to a second water injection scheme based on the image data; The water injection system injects water into the intestine through the enteroscope in response to the second water injection scheme.

2. The enteroscope-attached water system according to claim 1, characterized in that: Outputting a first water injection plan according to the user's physical examination data and location data, including Generate a basic scheme according to the age data in the physical examination data, wherein the basic scheme includes a maximum flow rate threshold and a maximum water volume threshold; Determine whether the obesity index is in a normal weight interval, if not, determine whether the obesity index is in an obesity interval, if yes, reduce the maximum flow rate threshold and the maximum water volume threshold by a second percentage; otherwise, reduce the maximum flow rate threshold and the maximum water volume threshold by a first percentage; The first water injection plan is generated according to the maximum flow rate value and the maximum water volume value.

3. The enteroscope-attached water system according to claim 2, characterized in that: Then adjusting the first water injection scheme to a second water injection scheme based on the image data includes: S321, obtaining the current position data of the colonoscope and the water injection amount data; S322: output the real-time water injection flow rate pre-stored in the database according to the current position data of the colonoscope and the water injection amount data: S323, dividing the image data into n*n number of sub-grid images; S324, comparing each sub-grid image with the water injection image pre-stored in the database, and outputting the matching degree of the sub-grid image; S325, constructing a rolling model, which is to construct a spherical model of a basic ball diameter and a groove of a basic radius, and cut out the cylindrical grooves that intersect with the basic radius in the spherical model to generate a spherical model of a basic size; S326, judging whether the matching degree of one of the sub-network images of the abnormal frames of the image data exceeds the first threshold value, if it exceeds, outputting a rotation angle according to the matching degree of the largest sub-network image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model; if it does not exceed, forming a sub-field grid image with every 4 grids that can form a "field" shape in the image data, wherein adjacent sub-field grid images may have an intersection, dividing the water injection image pre-stored in the database into sub-water injection images with n / 2*n / 2 grids, obtaining the matching degree of each sub-field grid image and the sub-water injection image, judging whether the matching degree of the sub-field grid image of one of the abnormal frames exceeds the first threshold value, if so, outputting a rotation angle according to the matching degree of the largest sub-field grid image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model; S327, determining whether the first flow rate is greater than the maximum flow rate threshold, if so, reducing the base radius by a third percentage, jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, reducing the number of grids n*n by a fourth percentage, and jumping to S323; S328, determining whether the water volume data exceeds the maximum water volume threshold, if so, reducing the distance between the cylindrical model formed by the base radius and the spherical model by a fifth percentage, and jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, outputting the first flow rate at this time; S329. Adjust the water injection flow rate of the colonoscope according to the first flow rate.

4. The enteroscope-attached water system according to claim 3, characterized in that: The first percentage is 5% to 10%.

5. The enteroscope-attached water system according to claim 3, characterized in that: The second percentage is 10% to 15%.

6. A method for detecting water attached to a colonoscope, characterized in that: include A first water injection scheme corresponding to the user's physical examination data and the position data of the colonoscope is stored; Input the user's physical examination data, colonoscope position data, and colonoscope image data; Outputting a first water injection scheme according to the user's physical examination data and position data; and adjusting the first water injection scheme to a second water injection scheme based on the image data; In response to the second water injection protocol, water is injected into the intestine through the enteroscope.

7. The enteroscope-attached water system according to claim 6, characterized in that: Outputting a first water injection plan according to the user's physical examination data and location data, including Generate a basic scheme according to the age data in the physical examination data, wherein the basic scheme includes a maximum flow rate threshold and a maximum water volume threshold; Determine whether the obesity index is in a normal weight interval, if not, determine whether the obesity index is in an obesity interval, if yes, reduce the maximum flow rate threshold and the maximum water volume threshold by a second percentage; otherwise, reduce the maximum flow rate threshold and the maximum water volume threshold by a first percentage; The first water injection plan is generated according to the maximum flow rate value and the maximum water volume value.

8. The method for detecting water attached to a colonoscope according to claim 7, characterized in that: Then adjusting the first water injection scheme to a second water injection scheme based on the image data includes: S321, obtaining the current position data of the colonoscope and the water injection amount data; S322: output the real-time water injection flow rate pre-stored in the database according to the current position data of the colonoscope and the water injection amount data: S323, dividing the image data into n*n number of sub-grid images; S324, comparing each sub-grid image with the water injection image pre-stored in the database, and outputting the matching degree of the sub-grid image; S325, constructing a rolling model, which is to construct a spherical model of a basic ball diameter and a groove of a basic radius, and cut out the cylindrical grooves that intersect with the basic radius in the spherical model to generate a spherical model of a basic size; S326, judging whether the matching degree of one of the sub-network images of the abnormal frames of the image data exceeds the first threshold value, if it exceeds, outputting a rotation angle according to the matching degree of the largest sub-network image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model; if it does not exceed, forming a sub-field grid image with every 4 grids that can form a "field" shape in the image data, wherein adjacent sub-field grid images may have an intersection, dividing the water injection image pre-stored in the database into sub-water injection images with n / 2*n / 2 grids, obtaining the matching degree of each sub-field grid image and the sub-water injection image, judging whether the matching degree of the sub-field grid image of one of the abnormal frames exceeds the first threshold value, if so, outputting a rotation angle according to the matching degree of the largest sub-field grid image, rotating the rolling model according to the rotation angle, and adjusting the real-time water injection flow rate to increase to the first flow rate according to the proportion of the projection of the groove in the perfect circle corresponding to the spherical model; S327, determining whether the first flow rate is greater than the maximum flow rate threshold, if so, reducing the base radius by a third percentage, jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, reducing the number of grids n*n by a fourth percentage, and jumping to S323; S328, determining whether the water volume data exceeds the maximum water volume threshold, if so, reducing the distance between the cylindrical model formed by the base radius and the spherical model by a fifth percentage, and jumping to S325, and recording the number of jumps, when the number of jumps exceeds the third threshold, outputting the first flow rate at this time; S329. Adjust the water injection flow rate of the colonoscope according to the first flow rate.

9. The method for detecting water attached to a colonoscope according to claim 8, characterized in that: The first percentage is 5% to 10%.

10. The method for detecting water attached to a colonoscope according to claim 8, characterized in that: The second percentage is 10% to 15%.

Images