Intelligent method and device for recognizing colorectal cancer

Through intelligent processing and identification models of intestinal lesions data, the missed problems of early intervention in bowel cancer are solved, the diagnostic accuracy and efficiency are improved, and the formulation of personalized treatment plans is realized.

CN118507026BActive Publication Date: 2024-11-22THE FIRST MEDICAL CENT CHINESE PLA GENERAL HOSPITAL
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410639234.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-22
Publication Date
2024-11-22
Estimated Expiration
2044-05-22

AI Technical Summary

Technical Problem

The existing technology is difficult to effectively intervene in the early stage of intestinal cancer, and the lack of intelligent lesion data correlation analysis and identification methods, resulting in the loss of early intervention.

Method used

The server comprehensively processes the current and historical lesion data of the patient's intestines, generates a bowel cancer recognition model, automatically identify lesion location and characteristic information, compares, identify lesion changes trends, and generates personalized critical time predictions.

Benefits of technology

It improves diagnostic accuracy and efficiency, realizes early identification and dynamic monitoring of intestinal cancer lesions, provides a more accurate diagnostic basis and personalized treatment plan, and optimizes doctor-patient communication.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118507026B_ABST
    Figure CN118507026B_ABST
Patent Text Reader

Abstract

The present invention provides an intelligent intestinal cancer identification method and device, comprising: a server processes the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion; the server extracts the historical lesion data of the patient's intestine to determine the second position information and second indication information corresponding to each first historical lesion; based on the first position information and the second position information, the current lesion and the historical lesion are screened and processed to obtain the statistically second historical lesion and the second current lesion; based on an intestinal cancer identification model, the first indication information and the second indication information of the second historical lesion are analyzed and processed to obtain intestinal cancer identification data; the intestinal cancer identification model generates a corresponding intestinal cancer display model for the patient based on the intestinal cancer identification data, the first position information and the first indication information.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to an intelligent-based method and device for recognizing colorectal cancer. Background Art

[0002] Bowel cancer develops slowly. Figure 1 Intestinal cancer generally goes through the process of hyperplastic polyps, small adenomatous polyps, large adenomatous polyps, severe atypical hyperplasia, pancreatic cancer, intestinal cancer, etc. It can take up to 5-10 years for hyperplastic polyps to develop into intestinal cancer. If intervention is carried out in the early stages of intestinal cancer development, the mortality rate can be significantly reduced.

[0003] In the prior art, the discovery of colorectal cancer still relies on the data of each examination. For example, when patient A had a colonoscopy in January, hyperplastic polyps were found, and when he had a colonoscopy in October, adenomatous polyps were found. When doctors evaluate the above data, they usually make a single evaluation and give the evaluation result that it is not colorectal cancer, and patient A is unaware of the changes in the lesions. Only when pancreatic cancer and colorectal cancer are found in subsequent colonoscopy examinations will they be taken seriously, but it is already difficult to intervene at this time, and early intervention is missed.

[0004] Therefore, how to intelligently perform correlation analysis and identification on lesion data and obtain identification data to assist doctors and patients in early intervention has become an urgent problem that needs to be solved. Summary of the invention

[0005] The embodiment of the present invention provides an intelligent method and device for recognizing colorectal cancer, which can intelligently perform correlation analysis and recognition on lesion data, and obtain recognition data to assist doctors and patients in early intervention.

[0006] A first aspect of an embodiment of the present invention provides an intelligent method for recognizing colorectal cancer, comprising:

[0007] The server processes the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion;

[0008] The server extracts historical lesion data of the patient's intestine to determine second location information and second indication information corresponding to each first historical lesion;

[0009] Based on the first position information and the second position information, the current lesion and the historical lesion are screened and processed to obtain a second historical lesion and a second current lesion after statistics;

[0010] Based on the colorectal cancer recognition model, the first indication information and the second indication information of the second current lesion and the second historical lesion are analyzed and processed to obtain colorectal cancer recognition data;

[0011] The colorectal cancer recognition model generates a colorectal cancer display model corresponding to the patient based on the colorectal cancer recognition data, the first location information and the first indication information.

[0012] Optionally, in a possible implementation of the first aspect, the server processes the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion, including:

[0013] The physician generates initial lesion data based on the operation information of the gastroenteroscopy, wherein the initial lesion data includes initial position information and first indication information;

[0014] The first posture information of the patient when the gastroenteroscopy is being operated is obtained, and the corresponding position direction is added to the initial position information based on the first posture information to obtain the first position information.

[0015] Optionally, in a possible implementation of the first aspect, the physician terminal generates initial lesion data based on operation information of a gastroenteroscopy, including:

[0016] After it is determined that the physician end has established a communication connection with the gastroenteroscope, the patient is examined based on the gastroenteroscope;

[0017] After the lesion is identified, the first indication information of the lesion extracted by the gastroenteroscopy is synchronized to the physician end;

[0018] The physician generates initial position information based on the angle at which the gastroenteroscope extracts the lesion, the preset size mark on the gastroenteroscope, and the intestinal information input by the physician.

[0019] Optionally, in a possible implementation manner of the first aspect, the acquiring first posture information of the patient when the gastroenteroscope is operated, and adding a corresponding position direction to the initial position information based on the first posture information to obtain the first position information includes:

[0020] The first posture information includes a positive side-lying posture and a reverse side-lying posture;

[0021] If the first position information is determined to be the lateral lying position, the angle of lesion extraction is analyzed based on the lateral lying position correspondence table to obtain the first position information based on the lateral lying position;

[0022] If the first position information is determined to be the reverse lateral position, the angle of lesion extraction is analyzed based on the reverse lateral position correspondence table to obtain the first position information based on the reverse lateral position.

[0023] Optionally, in a possible implementation of the first aspect, the corresponding front-side posture correspondence table and the back-side posture correspondence table are generated by the following steps, including:

[0024] Receiving the positive side angle interval and the corresponding positive side position configured by the management end, each positive side angle interval has a preset positive side position;

[0025] The reverse side angle interval and the corresponding reverse side position configured by the receiving management end are each configured such that each reverse side angle interval has a preset reverse side position.

[0026] Optionally, in a possible implementation manner of the first aspect, the screening and processing of the current lesion and the historical lesion based on the first position information and the second position information to obtain the statistically analyzed second historical lesion and the second current lesion includes:

[0027] Acquire the first position number of the first position information and the second position information based on the normal side-lying state, and acquire the second position number of the first position information and the second position information based on the reverse side-lying state;

[0028] If the number of the first positions is greater than or equal to the number of the second positions, the first position information and the second position information based on the reverse side-lying are transformed into the forward side-lying position to obtain the first position information and the second position information based on the forward side-lying position;

[0029] If the number of the first positions is less than the number of the second positions, the first position information and the second position information based on the normal side-lying are transformed into the reverse side-lying position information to obtain the first position information and the second position information based on the reverse side-lying position information;

[0030] The first position information and the second position information of the corresponding position are determined as the associated position, and the second historical lesion and the second current lesion corresponding to the associated position are counted.

[0031] Optionally, in a possible implementation manner of the first aspect, converting the first position information and the second position information based on the reverse side-lying state into the normal side-lying state includes:

[0032] Retrieving a preset reverse position conversion table, and performing a front-to-side conversion on the first position information and the second position information based on the reverse position conversion table, wherein the reverse position conversion table has a corresponding angle conversion relationship between the reverse-side position and the front-side position;

[0033] The first position information and the second position information based on the normal side-lying are converted into the reverse side-lying, comprising:

[0034] A preset front-reverse position conversion table is retrieved, and the first position information and the second position information are converted to a reverse side position based on the front-reverse position conversion table, wherein the front-reverse position conversion table has a corresponding angle conversion relationship between the front side position and the reverse side position.

[0035] Optionally, in a possible implementation manner of the first aspect, determining the first position information and the second position information of the corresponding position as the associated position, and counting the second historical lesion and the second current lesion corresponding to the associated position, includes:

[0036] Counting the first position information and the second position information having the same intestinal information as a first statistical set;

[0037] Compare the angles of lesion extraction of the first position information and the second position information in the first statistical set and the size marks preset on the gastroenteroscope;

[0038] If it is determined that the absolute value of the angle difference between the corresponding first position information and the second position information is smaller than the preset angle value, and the absolute value of the size mark difference is smaller than the preset size value, the corresponding first position information and the second position information are used as associated positions.

[0039] Optionally, in a possible implementation manner of the first aspect, the analyzing and processing the first indication information and the second indication information of the second current lesion and the second historical lesion based on the intestinal cancer recognition model to obtain the intestinal cancer recognition data includes:

[0040] The first indication information and the second indication information at least include size information and type information;

[0041] If the colorectal cancer recognition model determines that the size information between the second historical lesion and the second current lesion is reduced or unchanged, then the first lesion label is added to the corresponding second historical lesion;

[0042] If the colorectal cancer recognition model determines that the size information between the second historical lesion and the second current lesion is enlarged, a corresponding indication change function is generated based on the first indication information and the second indication information, and a first lesion label or a second lesion label is added to the corresponding second historical lesion based on the indication change function;

[0043] If the colorectal cancer recognition model determines that the type information of the second historical lesion has changed, then a second lesion label is added;

[0044] The first intestinal cancer identification data is obtained based on the second historical lesion with the first lesion label or the second lesion label added.

[0045] Optionally, in a possible implementation manner of the first aspect, if the second historical lesion size information is enlarged, generating a corresponding indication change function based on the first indication information and the second indication information, and adding a first lesion label or a second lesion label to the corresponding second historical lesion based on the indication change function, includes:

[0046] Acquire the size and time corresponding to the first indication information and the second indication information respectively, and construct the indication change function corresponding to the first indication information and the second indication information based on the size and the time;

[0047] Extracting a first change rate corresponding to the indication change function;

[0048] If the first change rate is greater than or equal to a first preset value, adding a second lesion label to the corresponding second historical lesion, and generating a corresponding critical time based on the critical value;

[0049] If the first change rate is less than a first preset value, a first lesion label is added to the corresponding second historical lesion.

[0050] Optionally, in a possible implementation manner of the first aspect, if the first change rate is greater than or equal to a first preset value, adding a second lesion label to the corresponding second historical lesion, and generating a corresponding critical time based on a critical value, includes:

[0051] Substituting the critical size corresponding to the critical value into the indicator change function to obtain an initial critical time;

[0052] Obtain the historical critical time of other third historical lesions previously calculated for the patient, and obtain the corresponding critical time difference according to the historical critical time and the actual critical time;

[0053] The initial critical time is adjusted based on the critical time difference to obtain a determined critical time.

[0054] Optionally, in a possible implementation manner of the first aspect, adjusting the initial critical time based on the critical time difference to obtain the determined critical time includes:

[0055] If the critical time difference is equal to 0, the initial critical time is used as the determined critical time;

[0056] If the critical time difference is greater than 0, the initial critical time is extended and adjusted based on the critical time difference to obtain a determined critical time;

[0057] If the critical time difference is less than 0, the initial critical time is shortened and adjusted based on the critical time difference to obtain a determined critical time;

[0058] The critical time is determined by calculating the following formula,

[0059]

[0060] Among them, t his is the historical critical time, t pisis the actual critical time of history, T con is the critical time to be determined, T ini is the initial critical time, k1 is the extension weight, and k2 is the shortening weight.

[0061] Optionally, in a possible implementation of the first aspect, the intestinal cancer recognition model generates a intestinal cancer display model corresponding to the patient based on the intestinal cancer recognition data, the first location information, and the first indication information, including:

[0062] Determining corresponding second position information in the initial intestinal cancer display model;

[0063] Establishing a corresponding first virtual lesion based on the first indication information of the second current lesion at the second position information, and establishing a corresponding second virtual lesion based on the second indication information of the second historical lesion;

[0064] Determine an overlapping area and a non-overlapping area of ​​the first virtual lesion and the second virtual lesion;

[0065] The overlapping area is displayed in a first preset format, and the non-overlapping area is displayed in a second preset format.

[0066] Optionally, in a possible implementation of the first aspect, the method further includes:

[0067] Determine a center point of the first virtual lesion, and obtain a first outline of the first virtual lesion;

[0068] Calculating a first multiple between a critical size and the first virtual lesion size, and amplifying the first contour according to the center point based on the first multiple to generate a corresponding critical contour;

[0069] The critical time corresponding to each first virtual lesion is added to the corresponding critical contour to obtain an adjusted colorectal cancer display model.

[0070] A second aspect of an embodiment of the present invention provides an intelligent intestinal cancer identification device, comprising:

[0071] A processing module, configured to enable the server to process the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion;

[0072] An extraction module, configured to enable the server to extract historical lesion data of the patient's intestine, and determine second location information and second indication information corresponding to each first historical lesion;

[0073] A screening module, used for screening and processing the current lesion and the historical lesion based on the first position information and the second position information to obtain a second historical lesion and a second current lesion after statistics;

[0074] An analysis module, configured to analyze and process the first indication information and the second indication information of the second historical lesion based on a colorectal cancer recognition model to obtain colorectal cancer recognition data;

[0075] A generation module is used to enable the colorectal cancer recognition model to generate a colorectal cancer display model corresponding to the patient based on the colorectal cancer recognition data, the first position information and the first indication information.

[0076] The present invention provides a.

[0077] 1. This solution can improve the accuracy and efficiency of diagnosis. It can identify and analyze colorectal cancer lesions through intelligent methods, significantly improving the accuracy and efficiency of diagnosis. Specifically, the server comprehensively processes the current and historical lesion data of the patient's intestine. The system can automatically identify the location and characteristic information (including size, type, etc.) of colorectal cancer lesions and compare them with historical data. This process reduces reliance on the doctor's subjective judgment and can automatically track the development and changes of lesions, thereby providing doctors with more accurate diagnostic basis, especially when lesions change in the early stages, which helps to achieve earlier intervention and treatment.

[0078] 2. This solution achieves dynamic monitoring of changes in patient lesions by integrating the patient's current lesion information and historical lesion information. The system automatically identifies lesion change trends (such as increase or decrease in size) by comparing the location, size and other indicator information of the current lesion with the historical lesions through intelligent algorithms. In particular, the system can analyze the growth rate and change trend of lesions based on the colorectal cancer recognition model and generate personalized critical time predictions, which is of great significance for formulating early intervention and personalized treatment plans. By calculating the growth rate of lesions and predicting the time when they reach the critical size, doctors can make more accurate decisions on the timing and strategy of treatment and optimize the patient's treatment management plan.

[0079] 3. This solution can not only intelligently identify and analyze lesions, but also generate a patient's colorectal cancer display model based on the obtained data. This model provides doctors and patients with a more intuitive display of the disease status by intuitively displaying the location, size and predicted change trend of the lesions. Specifically, the system can display the current status and historical changes of the lesions on a virtual model of the intestine, including overlapping and non-overlapping areas of the lesions, as well as critical times and sizes. This intuitive display helps doctors better explain the progression of the disease to patients, while also making it easier for patients to understand their own condition and treatment needs, thereby improving the efficiency and quality of doctor-patient communication. This solution collectively embodies the advantages of the patent in intelligent identification, dynamic monitoring and intuitive display of disease status. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 This is a schematic diagram of the development of intestinal cancer provided by the present invention;

[0081] Figure 2 It is a flowchart of an intelligent method for recognizing colorectal cancer provided by an embodiment of the present invention;

[0082] Figure 3 is a schematic diagram of an outline provided by an embodiment of the present invention;

[0083] Figure 4 It is a structural schematic diagram of an intelligent colorectal cancer identification device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0084] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0085] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein.

[0086] It should be understood that in various embodiments of the present invention, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0087] It should be understood that in the present invention, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0088] It should be understood that in the present invention, "plurality" refers to two or more than two. "And / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "Contains A, B and C", "Contains A, B, C" means that A, B, and C are all included, "Contains A, B or C" means that one of A, B, and C is included, and "Contains A, B and / or C" means that any one, any two, or any three of A, B, and C are included.

[0089] It should be understood that in the present invention, "B corresponding to A", "B corresponding to A", "A corresponds to B" or "B corresponds to A" means that B is associated with A and B can be determined based on A. Determining B based on A does not mean determining B based only on A, but B can also be determined based on A and / or other information. A and B match when the similarity between A and B is greater than or equal to a preset threshold.

[0090] Depending on the context, "if" as used herein may be interpreted as "when" or "when" or "in response to determining" or "in response to detecting."

[0091] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0092] See also Figure 2 , is a flow chart of an intelligent method for recognizing colorectal cancer provided by an embodiment of the present invention, the method comprising:

[0093] S1. The server processes the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion.

[0094] In this step, after the communication connection between the physician and the gastroenteroscope is established, the patient is examined by gastroenteroscope. The physician examines the patient's intestines through the gastroenteroscope, and during the examination, the current lesion data, i.e., problem data, can be data on hyperplastic polyps, pancreatic polyps, severe atypical hyperplasia, pancreatic cancer, etc. In this solution, once the lesion is identified, the gastroenteroscope will synchronize the location and characteristics of the lesion to the physician.

[0095] In some embodiments, the server processes the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion, including:

[0096] The physician generates initial lesion data based on the operation information of the gastroenteroscopy, and the initial lesion data includes initial position information and first indication information.

[0097] The initial position information is used to indicate the position of the lesion in the intestine, and the first indication information is used to indicate the data of the lesion, such as the area and volume of the lesion. As known from the background technology, different first indication information corresponds to different types of lesions.

[0098] Specifically, the physician terminal generates initial lesion data based on the operation information of the gastroenteroscopy, including:

[0099] After it is determined that the physician side has established a communication connection with the gastroenteroscope, the patient is examined based on the gastroenteroscope. After it is determined that the lesion is identified, the gastroenteroscope extracts the first indication information of the lesion and synchronizes it to the physician side. In actual applications, after the physician side has established a communication connection with the gastroenteroscope, the physician examines the patient's intestines through the gastroenteroscope. When a lesion is detected, the gastroenteroscope extracts preliminary information about the lesion (such as location, size, etc.) and synchronizes it to the physician side.

[0100] The physician generates initial position information based on the angle at which the gastroenteroscope extracts the lesion, the preset size mark on the gastroenteroscope, and the intestinal information input by the physician.

[0101] Among them, since the intestine is cylindrical, the angle of the gastroenteroscope to extract the lesion can be, for example, a certain angle range in the 360 ​​degrees of the intestine; the preset size mark on the gastroenteroscope refers to the length information of the gastroenteroscope penetrating into the intestine, such as 10 centimeters. Combining the angle and size mark can obtain a relatively accurate position of the lesion in the intestine.

[0102] The first posture information of the patient when the gastroenteroscopy is being operated is obtained, and the corresponding position direction is added to the initial position information based on the first posture information to obtain the first position information.

[0103] It is worth mentioning that during the gastrointestinal endoscopy, the patient's corresponding posture information is different, generally including the forward side-lying posture or the reverse side-lying posture. Among them, the forward side-lying posture can be the patient facing the doctor forward, and the reverse side-lying posture can be the patient facing the doctor backward. It should be noted that since the intestine is cylindrical, the relative angle of the gastroenteroscope corresponding to the patient's intestine is different for different posture information. For example, the relative angle of the forward side-lying posture or the reverse side-lying posture differs by 180 degrees.

[0104] This solution will simultaneously record the patient's posture information during each examination, and will subsequently use this information to adjust the initial lesion location information in order to unify the multiple test data.

[0105] In the above embodiment, the step of obtaining the first posture information of the patient when the gastroenteroscopy is being operated, and adding the corresponding position direction to the initial position information based on the first posture information to obtain the first position information includes:

[0106] The first posture information includes a positive lateral lying posture and a reverse lateral lying posture. In this solution, the posture requirements for the patient are a positive lateral lying posture and a reverse lateral lying posture.

[0107] If the first posture information is judged to be the lateral lying posture, the angle of lesion extraction is analyzed based on the lateral lying posture correspondence table to obtain the first position information based on the lateral lying posture. In order to analyze the impact of the patient's posture on the lesion position information, the system presets the lateral lying posture correspondence table and the contralateral lying posture correspondence table. These tables are generated according to the angle intervals and corresponding position information configured by the management end, and are used to correct and transform the lesion position. In this step, if the patient is in the lateral lying posture, it is necessary to combine the lateral lying posture correspondence table to obtain the first position information based on the lateral lying posture.

[0108] If the first posture information is determined to be the reverse lateral position, the angle of lesion extraction is analyzed based on the reverse lateral position correspondence table to obtain the first position information based on the reverse lateral position. In this step, if the first posture information of the patient is the reverse lateral position, the angle of lesion extraction needs to be analyzed in combination with the reverse lateral position correspondence table to obtain the first position information based on the reverse lateral position.

[0109] The following steps are used to generate the corresponding positive and negative posture correspondence tables, including:

[0110] The positive side angle interval and the corresponding positive side position configured by the receiving management end are each positive side angle interval having a preset positive side position.

[0111] The reverse side angle interval and the corresponding reverse side position configured by the receiving management end are each configured such that each reverse side angle interval has a preset reverse side position.

[0112] Through the above steps, the corresponding table can be pre-determined in a preset form, and the corresponding angle area can be matched with the corresponding table to facilitate subsequent angle conversion.

[0113] S2, the server extracts the historical lesion data of the patient's intestine, and determines the second position information and second indication information corresponding to each first historical lesion.

[0114] It is worth mentioning that, since it is necessary to know the change of the lesion in order to predict the subsequent change trend, it is necessary to compare and analyze the historical data. Therefore, it is necessary to obtain the second position information and second indication information corresponding to each first historical lesion in the database. The second position information and second indication information refer to historical data.

[0115] S3, based on the first position information and the second position information, the current lesion and the historical lesion are screened and processed to obtain a second historical lesion and a second current lesion after statistics.

[0116] Since the lesion data needs to be compared, and only the lesion data at the same location are comparable, this solution needs to find lesions with corresponding locations for comparison, that is, the second historical lesion and the second current lesion mentioned above. It can be understood that there can be many groups of the second historical lesion and the second current lesion.

[0117] In some embodiments, the screening and processing of the current lesion and the historical lesion based on the first location information and the second location information to obtain the second historical lesion and the second current lesion after statistics includes:

[0118] S31, obtaining the first position number of the first position information and the second position information based on the forward side lying state, and obtaining the second position number of the first position information and the second position information based on the reverse side lying state.

[0119] In the above embodiment, it is mentioned that since the postures of lying on the side and lying on the opposite side will cause the angle to change, in order to better track and compare the changes of lesions of patients over time, thereby facilitating diagnosis and treatment decisions, the present solution will use the lesion location information of the unified posture to more accurately compare images at different time points to observe the changes of lesions.

[0120] S32, if the number of the first positions is greater than or equal to the number of the second positions, converting the first position information and the second position information based on the reverse side-lying state into the forward side-lying state to obtain the first position information and the second position information based on the forward side-lying state;

[0121] When performing body position conversion, in order to reduce the amount of data processing and improve conversion efficiency, this solution will compare the number of first positions and the number of second positions. If the number of first positions is greater than or equal to the number of second positions, it means that the number based on the forward side lying is larger, then the reverse side lying data can be converted into the forward side lying to obtain the first position information and the second position information based on the forward side lying.

[0122] In the above embodiment, the first position information and the second position information based on the reverse side-lying are converted into the normal side-lying state, including:

[0123] A preset reverse position conversion table is retrieved, and based on the reverse position conversion table, the first position information and the second position information are converted into front and side lying. The reverse position conversion table has a corresponding angle conversion relationship between the reverse side position and the front side position.

[0124] The present solution is provided with a preset reverse position conversion table, which includes the corresponding angle conversion relationship between the reverse side position and the forward side position. The angle conversion relationship can be, for example, 180 degrees, or can be pre-set by the physician based on experience. When performing the conversion, the first position information and the second position information can be combined with the degree conversion relationship for conversion.

[0125] S33: If the number of the first positions is less than the number of the second positions, the first position information and the second position information based on the normal side-lying state are transformed into the reverse side-lying state to obtain the first position information and the second position information based on the reverse side-lying state.

[0126] Different from step S32, the number of first positions in this step is less than the number of second positions, indicating that the number based on the reverse side lying is larger. Then the data of the normal side lying can be converted into the reverse side lying to obtain the first position information and the second position information based on the reverse side lying.

[0127] When the conversion is specifically performed, the conversion can be performed in combination with the same principle as in the above embodiment, and the first position information and the second position information based on the normal side-lying are converted into the reverse side-lying conversion, including:

[0128] The preset positive and negative position conversion table is retrieved, and the first position information and the second position information are converted to the reverse side based on the positive and negative position conversion table, wherein the positive and negative position conversion table has a corresponding angle conversion relationship between the positive side position and the reverse side position. This method is similar to the above embodiment and will not be described in detail.

[0129] S34, determining the first position information and the second position information of the corresponding position as the associated position, and counting the second historical lesion and the second current lesion corresponding to the associated position.

[0130] It is understandable that in order to compare the lesion data, it is necessary to find the data at the corresponding position for comparison. Therefore, it is necessary to find the corresponding first position information and second position information as a comparison group, and then find the second historical lesion and the second current lesion corresponding to the comparison group for comparison.

[0131] The step of determining the first position information and the second position information of the corresponding position as the associated position, and counting the second historical lesion and the second current lesion corresponding to the associated position includes:

[0132] The first position information and the second position information with the same intestinal information are counted as a first statistical set. The intestinal information is, for example, intestinal information such as the rectum and small intestine, and different intestinal information corresponds to different lesions. Therefore, this solution needs to find the first position information and the second position information of the same intestinal information as a set to compare the same intestinal information.

[0133] The angles of lesion extraction of the first position information and the second position information in the first statistical set and the size marks preset on the gastroenteroscopy are compared. After obtaining the first statistical set, this solution needs to compare the first position information and the second position information to determine whether the two can be used as a comparison group.

[0134] If it is determined that the absolute value of the angle difference between the corresponding first position information and the second position information is smaller than the preset angle value, and the absolute value of the size mark difference is smaller than the preset size value, the corresponding first position information and the second position information are used as associated positions.

[0135] It is understandable that when comparing, this solution needs to compare two dimensions: angle and size. If the difference in angle is within a certain range, and the difference in size mark is also within a certain range, then these two position information will be identified as related positions. This means that these two position information are likely to point to the state of the same lesion at different time points. It is worth mentioning that this solution uses preset angle values ​​and preset size values ​​for judgment, which can allow some errors to exist.

[0136] S4, analyzing and processing the first indication information and the second indication information of the second historical lesion based on the colorectal cancer recognition model to obtain colorectal cancer recognition data.

[0137] This solution will use the colorectal cancer recognition model to identify and judge the corresponding data to obtain colorectal cancer recognition data.

[0138] In some embodiments, the analyzing and processing the first indication information and the second indication information of the second historical lesion based on the intestinal cancer recognition model to obtain the intestinal cancer recognition data includes:

[0139] S41: The first indication information and the second indication information at least include size information and type information.

[0140] The size information refers to the size of the lesion, such as volume, etc. The type information is related to the size information and may refer to the type of the lesion (such as adenoma, polyp, etc.).

[0141] S42: If the colorectal cancer recognition model determines that the size information between the second historical lesion and the second current lesion is reduced or unchanged, a first lesion label is added to the corresponding second historical lesion.

[0142] If the colorectal cancer recognition model determines that the size information of the second historical lesion shows that the lesion size has become smaller or remains unchanged, this means that the lesion may be benign and has not developed into a malignant state. The model will then add a first lesion label to this historical lesion.

[0143] S43, if the colorectal cancer recognition model determines that the size information between the second historical lesion and the second current lesion is increasing, it generates a corresponding indication change function based on the first indication information and the second indication information, and adds a first lesion label or a second lesion label to the corresponding second historical lesion based on the indication change function.

[0144] If the model determines that the size of the lesion has increased, this may indicate that the condition is at risk of worsening. In this case, the colorectal cancer recognition model will further generate an indication change function, which describes the change pattern of the lesion over time based on the first and second indication information. The role of the colorectal cancer recognition model here is to assist doctors in determining the nature and development trend of the lesion by analyzing the change trend of the lesion. The distinction between benign and malignant lesions is crucial for patients' treatment plans and prognosis.

[0145] Wherein, if the second historical lesion size information is enlarged, generating a corresponding indication change function based on the first indication information and the second indication information, and adding a first lesion label or a second lesion label to the corresponding second historical lesion based on the indication change function, including:

[0146] The size and time corresponding to the first indication information and the second indication information are obtained, and the indication change function corresponding to the first indication information and the second indication information is constructed based on the size and time. First, the model obtains the size information and time information related to the first indication information and the second indication information. This means that the model will consider the size measurement values ​​of the lesion at different time points. Based on the size and time data collected above, the model constructs an indication change function, which describes the size change trend of the lesion over time. The indication change function can be a linear function.

[0147] Extract the first change rate corresponding to the indication change function. This solution model extracts the first change rate of the indication change function, which is a quantitative indicator to measure the growth rate of the lesion size over time. The first change rate can be the slope of the indication change function.

[0148] If the first change rate is greater than or equal to the first preset value, a second lesion label is added to the corresponding second historical lesion, and a corresponding critical time is generated based on the critical value; if the first change rate is less than the first preset value, a first lesion label is added to the corresponding second historical lesion.

[0149] The model compares the extracted first rate of change with a first preset value. If the first rate of change is greater than or equal to the first preset value, this may mean that the lesion is growing faster, indicating high risk or malignant potential. In this case, the model will mark the corresponding second historical lesion as a second lesion label and calculate a critical time based on this rate of change. This critical time may refer to the estimated time for the lesion to reach a specific size or deteriorate to a specific degree. If the first rate of change is less than the first preset value, this may indicate that the lesion is growing slower and may be benign or low risk. Therefore, the model will mark this lesion as the first lesion label.

[0150] If the first change rate is greater than or equal to a first preset value, a second lesion label is added to the corresponding second historical lesion, and a corresponding critical time is generated based on the critical value, including:

[0151] Substitute the critical size corresponding to the critical value into the indication change function to obtain the initial critical time. In this solution, substitute the critical size corresponding to the critical value (which may refer to a lesion size threshold that requires intervention or attention) into the indication change function. The estimated time for the lesion size to reach the critical size is obtained by calculation, that is, the initial critical time.

[0152] Obtain the historical critical time of other third historical lesions previously calculated by the patient, and obtain the corresponding critical time difference according to the historical critical time and the actual critical time. It is worth mentioning that if the patient has other lesions before, or the lesion data of other patients, and the historical critical time of those lesions has been calculated, these data will be used to further correct the critical time. This scheme needs to compare the historical critical time and the actual critical time to obtain the critical time difference.

[0153] The initial critical time is adjusted based on the critical time difference to obtain a determined critical time. This solution adjusts the initial critical time based on the obtained critical time difference. This adjustment is to make the predicted critical time more accurate, taking into account the individual differences in the development of lesions in patients.

[0154] The adjusting the initial critical time based on the critical time difference to obtain the determined critical time includes:

[0155] If the critical time difference is equal to 0, the initial critical time is taken as the determined critical time. If the historical critical time and the actual critical time are the same (i.e., the difference is 0), it means that the calculation is very accurate, and no adjustment is required, and the initial critical time is taken as the determined critical time. Of course, this situation is unlikely to occur.

[0156] If the critical time difference is greater than 0, the initial critical time is extended and adjusted based on the critical time difference to obtain a determined critical time. If the historical critical time is longer than the actual critical time (ie, the difference is greater than 0), the initial critical time needs to be extended.

[0157] If the critical time difference is less than 0, the initial critical time is shortened and adjusted based on the critical time difference to obtain the determined critical time. If the historical critical time is shorter than the actual critical time (ie, the difference is less than 0), the initial critical time needs to be shortened.

[0158] The critical time is determined by calculating the following formula,

[0159]

[0160] Among them, t his is the historical critical time, t pos is the actual critical time of history, T con is the critical time to be determined, T ini is the initial critical time, k1 is the extension weight, and k2 is the shortening weight.

[0161] In the above formula, Represents the time adjustment coefficient. If the difference is larger, the larger the adjustment required will be. Represents the time adjustment coefficient. If the difference is larger, the larger the adjustment required will be.

[0162] Through the above adjustment process, the final determined critical time can better reflect the individual differences and historical development trends of the patient's lesions. Doctors can use this time as a reference to decide the monitoring frequency and intervention timing of the lesions to optimize the patient's treatment management plan. Among them, the extension weight is less than the shortening weight, which means that a more radical adjustment is adopted when shortening the critical time than when extending it. The choice of weights may be based on clinical data, statistical analysis, or expert experience, aiming to provide a reasonable adjustment to ensure that the determined critical time is neither too conservative nor too radical.

[0163] S44, if the colorectal cancer recognition model determines that the type information of the second historical lesion has changed, a second lesion label is added. It is understandable that if the recognition model determines that the type information of the second historical lesion is different from the previous record or has changed, the system will automatically add a second lesion label to identify the change. The first lesion label and the second lesion label may represent the lesion detection results at different time points or under different conditions.

[0164] S45, obtaining first intestinal cancer identification data based on the second historical lesion to which the first lesion label or the second lesion label is added.

[0165] In this way, the system can track the historical changes of lesions and provide doctors with detailed data analysis, which helps doctors better understand the development of the disease and provide patients with more precise treatment plans.

[0166] S5, the colorectal cancer recognition model generates a colorectal cancer display model corresponding to the patient based on the colorectal cancer recognition data, the first location information and the first indication information.

[0167] This solution can generate a patient-specific colorectal cancer display model. This model combines colorectal cancer identification data, location information, and indication information to more intuitively display the characteristics of colorectal cancer lesions.

[0168] In some embodiments, the intestinal cancer recognition model generates an intestinal cancer display model corresponding to the patient based on the intestinal cancer recognition data, the first location information and the first indication information, including:

[0169] The corresponding second position information is determined in the initial intestinal cancer display model. The initial intestinal cancer display model is the initial intestinal virtual model corresponding to the patient, which can be a digital twin model. The second position information is the location of the historical lesion, which needs to be marked in the initial intestinal virtual model.

[0170] A corresponding first virtual lesion is established based on the first indication information of the second current lesion at the second position information, and a corresponding second virtual lesion is established based on the second indication information of the second historical lesion. It can be understood that this step is to obtain the shape of the lesion at the historical moment and the current moment, and display it at the corresponding position of the initial intestinal virtual model. It can be understood that the shape size at different moments is different.

[0171] Determine the overlapping area and non-overlapping area of ​​the first virtual lesion and the second virtual lesion. Display the overlapping area in a first preset form, and display the non-overlapping area in a second preset form. This solution needs to find the overlapping area and non-overlapping area between corresponding contours and display them differently. The first preset form display can be, for example, a blue form display, and the second preset form display can be a red form display.

[0172] Based on the above embodiment, it also includes:

[0173] Determine the center point of the first virtual lesion and obtain a first contour of the first virtual lesion. Figure 3It is understandable that this step requires obtaining the first contour of the first virtual lesion. The method for determining the center point of the first virtual lesion may be determined by a doctor, or the contour of the first virtual lesion may be processed into coordinates to obtain coordinate information of multiple pixel points in the contour, and then the maximum Y and minimum Y are found in the multiple coordinate information, and the Y coordinate value of the center point is obtained by combining the average value of the maximum Y and the minimum Y, and the maximum X and the minimum X are found in the multiple coordinate information, and the X coordinate value of the center point is obtained by combining the average value of the maximum X and the minimum X, and the position of the center point is obtained by combining the X coordinate value of the center point and the Y coordinate value of the center point.

[0174] Calculate the first multiple between the critical size and the first virtual lesion size, and based on the first multiple, magnify the first contour according to the center point to generate a corresponding critical contour. It can be understood that this solution needs to obtain the multiple of the critical size greater than the first virtual lesion size, and the above-mentioned first multiple is a multiple greater than 1, and then magnify the first contour according to the first multiple to obtain a predicted contour and generate a corresponding critical contour. The critical contour is the predicted contour corresponding to the critical size.

[0175] The critical time corresponding to each first virtual lesion is added to the corresponding critical contour to obtain an adjusted intestinal cancer display model. This solution also adds the critical time to the critical contour to display the predicted time of the contour.

[0176] See also Figure 4 , is a schematic diagram of the structure of an intelligent intestinal cancer identification device provided in an embodiment of the present invention, the intelligent intestinal cancer identification device comprising:

[0177] A processing module, configured to enable the server to process the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion;

[0178] An extraction module, configured to enable the server to extract historical lesion data of the patient's intestine, and determine second location information and second indication information corresponding to each first historical lesion;

[0179] A screening module, used for screening and processing the current lesion and the historical lesion based on the first position information and the second position information to obtain a second historical lesion and a second current lesion after statistics;

[0180] an analysis module, configured to analyze and process the first indication information and the second indication information of the second historical lesion based on a colorectal cancer recognition model to obtain colorectal cancer recognition data;

[0181] A generation module is used to enable the colorectal cancer recognition model to generate a colorectal cancer display model corresponding to the patient based on the colorectal cancer recognition data, the first position information and the first indication information.

[0182] The present invention also provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided by the various embodiments described above.

[0183] Among them, the storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, the storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (Application Specific Integrated Circuits, referred to as ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the storage medium can also exist in a communication device as discrete components. The storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0184] The present invention also provides a program product, which includes an execution instruction, which is stored in a storage medium. At least one processor of a device can read the execution instruction from the storage medium, and at least one processor executes the execution instruction so that the device implements the methods provided in the above various embodiments.

[0185] In the above-mentioned terminal or server embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0186] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent method for recognizing colorectal cancer, characterized in that: include: The server processes the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion; The server extracts historical lesion data of the patient's intestine to determine second location information and second indication information corresponding to each first historical lesion; Based on the first position information and the second position information, the current lesion and the historical lesion are screened and processed to obtain a second historical lesion and a second current lesion after statistics; Based on the colorectal cancer recognition model, the first indication information and the second indication information of the second current lesion and the second historical lesion are analyzed and processed to obtain colorectal cancer recognition data; The intestinal cancer recognition model generates an intestinal cancer display model corresponding to the patient based on the intestinal cancer recognition data, the first location information and the first indication information; The method of screening and processing the current lesion and the historical lesion based on the first position information and the second position information to obtain the second historical lesion and the second current lesion after statistics includes: Acquire the first position number of the first position information and the second position information based on the normal side-lying state, and acquire the second position number of the first position information and the second position information based on the reverse side-lying state; If the number of the first positions is greater than or equal to the number of the second positions, the first position information and the second position information based on the reverse side-lying are transformed into the forward side-lying position to obtain the first position information and the second position information based on the forward side-lying position; If the number of the first positions is less than the number of the second positions, the first position information and the second position information based on the normal side-lying are transformed into the reverse side-lying position information to obtain the first position information and the second position information based on the reverse side-lying position information; Determine the first position information and the second position information of the corresponding position as the associated position, and count the second historical lesion and the second current lesion corresponding to the associated position; The first position information and the second position information based on the reverse side-lying are converted into the forward side-lying state, comprising: Retrieving a preset reverse position conversion table, and performing a front-to-side conversion on the first position information and the second position information based on the reverse position conversion table, wherein the reverse position conversion table has a corresponding angle conversion relationship between the reverse-side position and the front-side position; The first position information and the second position information based on the normal side-lying are converted into the reverse side-lying, comprising: Retrieving a preset forward and reverse position conversion table, and performing reverse-side lying conversion on the first position information and the second position information based on the forward and reverse position conversion table, wherein the forward and reverse position conversion table has a corresponding angle conversion relationship between the forward side position and the reverse side position; The step of determining the first position information and the second position information of the corresponding position as the associated position, and counting the second historical lesion and the second current lesion corresponding to the associated position, comprises: Counting the first position information and the second position information having the same intestinal information as a first statistical set; Compare the angles of lesion extraction of the first position information and the second position information in the first statistical set and the size marks preset on the gastroenteroscope; If it is determined that the absolute value of the angle difference between the corresponding first position information and the second position information is smaller than the preset angle value, and the absolute value of the size mark difference is smaller than the preset size value, the corresponding first position information and the second position information are used as the associated position; The analyzing and processing the first indication information and the second indication information of the second current lesion and the second historical lesion based on the intestinal cancer recognition model to obtain intestinal cancer recognition data includes: The first indication information and the second indication information at least include size information and type information; If the colorectal cancer recognition model determines that the size information between the second historical lesion and the second current lesion is reduced or unchanged, then the first lesion label is added to the corresponding second historical lesion; If the colorectal cancer recognition model determines that the size information between the second historical lesion and the second current lesion is enlarged, a corresponding indication change function is generated based on the first indication information and the second indication information, and a first lesion label or a second lesion label is added to the corresponding second historical lesion based on the indication change function; If the colorectal cancer recognition model determines that the type information of the second historical lesion has changed, then a second lesion label is added; Obtaining first intestinal cancer identification data based on the second historical lesion with the first lesion label or the second lesion label added; If the second historical lesion size information is enlarged, generating a corresponding indication change function based on the first indication information and the second indication information, and adding a first lesion label or a second lesion label to the corresponding second historical lesion based on the indication change function, including: Acquire the size and time corresponding to the first indication information and the second indication information respectively, and construct the indication change function corresponding to the first indication information and the second indication information based on the size and the time; Extracting a first change rate corresponding to the indication change function; If the first change rate is greater than or equal to a first preset value, adding a second lesion label to the corresponding second historical lesion, and generating a corresponding critical time based on the critical value; If the first change rate is less than a first preset value, a first lesion label is added to the corresponding second historical lesion.

2. The intelligent intestinal cancer identification method according to claim 1, characterized in that: The server processes the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion, including: The physician generates initial lesion data based on the operation information of the gastroenteroscopy, wherein the initial lesion data includes initial position information and first indication information; The first posture information of the patient when the gastroenteroscopy is being operated is obtained, and the corresponding position direction is added to the initial position information based on the first posture information to obtain the first position information.

3. The intelligent intestinal cancer identification method according to claim 2, characterized in that: The physician terminal generates initial lesion data based on the operation information of the gastroenteroscopy, including: After it is determined that the physician end has established a communication connection with the gastroenteroscope, the patient is examined based on the gastroenteroscope; After the lesion is identified, the first indication information of the lesion extracted by the gastroenteroscopy is synchronized to the physician end; The physician generates initial position information based on the angle at which the gastroenteroscope extracts the lesion, the preset size mark on the gastroenteroscope, and the intestinal information input by the physician.

4. The intelligent method for recognizing colorectal cancer according to claim 3, characterized in that: The step of obtaining the first posture information of the patient when the gastroenteroscopy is being operated, and adding a corresponding position direction to the initial position information based on the first posture information to obtain the first position information includes: The first posture information includes a positive side-lying posture and a reverse side-lying posture; If the first position information is determined to be the lateral lying position, the angle of lesion extraction is analyzed based on the lateral lying position correspondence table to obtain the first position information based on the lateral lying position; If the first position information is determined to be the reverse lateral position, the angle of lesion extraction is analyzed based on the reverse lateral position correspondence table to obtain the first position information based on the reverse lateral position.

5. The intelligent method for recognizing colorectal cancer according to claim 4, characterized in that: The corresponding positive and negative posture correspondence tables and negative posture correspondence tables are generated by the following steps, including: Receiving the positive side angle interval and the corresponding positive side position configured by the management end, each positive side angle interval has a preset positive side position; The reverse side angle interval and the corresponding reverse side position configured by the receiving management end are each configured such that each reverse side angle interval has a preset reverse side position.

6. The intelligent method for recognizing colorectal cancer according to claim 1, characterized in that: If the first change rate is greater than or equal to a first preset value, adding a second lesion label to the corresponding second historical lesion, and generating a corresponding critical time based on the critical value, including: Substituting the critical size corresponding to the critical value into the indicator change function to obtain an initial critical time; Obtain the historical critical time of other third historical lesions previously calculated for the patient, and obtain the corresponding critical time difference according to the historical critical time and the actual critical time; The initial critical time is adjusted based on the critical time difference to obtain a determined critical time.

7. The intelligent method for recognizing colorectal cancer according to claim 6, characterized in that: The adjusting the initial critical time based on the critical time difference to obtain the determined critical time includes: If the critical time difference is equal to 0, the initial critical time is used as the determined critical time; If the critical time difference is greater than 0, the initial critical time is extended and adjusted based on the critical time difference to obtain a determined critical time; If the critical time difference is less than 0, the initial critical time is shortened and adjusted based on the critical time difference to obtain a determined critical time; The critical time is determined by calculating the following formula, in, For the critical time of history, is the actual critical time of history, To determine the critical time, is the initial critical time, To extend the weight, To shorten the weight.

8. The intelligent method for recognizing colorectal cancer according to claim 7, characterized in that: The intestinal cancer recognition model generates an intestinal cancer display model corresponding to the patient based on the intestinal cancer recognition data, the first location information and the first indication information, including: Determining corresponding second position information in the initial intestinal cancer display model; Establishing a corresponding first virtual lesion based on the first indication information of the second current lesion at the second position information, and establishing a corresponding second virtual lesion based on the second indication information of the second historical lesion; Determine an overlapping area and a non-overlapping area of ​​the first virtual lesion and the second virtual lesion; The overlapping area is displayed in a first preset format, and the non-overlapping area is displayed in a second preset format.

9. The intelligent method for recognizing colorectal cancer according to claim 8, characterized in that: Also includes: Determine a center point of the first virtual lesion, and obtain a first outline of the first virtual lesion; Calculating a first multiple between a critical size and the first virtual lesion size, and amplifying the first contour according to the center point based on the first multiple to generate a corresponding critical contour; The critical time corresponding to each first virtual lesion is added to the corresponding critical contour to obtain an adjusted colorectal cancer display model.

10. An intelligent intestinal cancer identification device, characterized in that: include: A processing module, configured to enable the server to process the current lesion data of the patient's intestine to obtain first position information and first indication information corresponding to each first current lesion; An extraction module, configured to enable the server to extract historical lesion data of the patient's intestine, and determine second location information and second indication information corresponding to each first historical lesion; A screening module, used for screening and processing the current lesion and the historical lesion based on the first position information and the second position information to obtain a second historical lesion and a second current lesion after statistics; An analysis module, configured to analyze and process the first indication information and the second indication information of the second current lesion and the second historical lesion based on a colorectal cancer recognition model to obtain colorectal cancer recognition data; A generating module, configured to enable the intestinal cancer recognition model to generate an intestinal cancer display model corresponding to the patient based on the intestinal cancer recognition data, the first position information and the first indication information; The method of screening and processing the current lesion and the historical lesion based on the first position information and the second position information to obtain the second historical lesion and the second current lesion after statistics includes: Acquire the first position number of the first position information and the second position information based on the normal side-lying state, and acquire the second position number of the first position information and the second position information based on the reverse side-lying state; If the number of the first positions is greater than or equal to the number of the second positions, the first position information and the second position information based on the reverse side-lying are transformed into the forward side-lying position to obtain the first position information and the second position information based on the forward side-lying position; If the number of the first positions is less than the number of the second positions, the first position information and the second position information based on the normal side-lying are transformed into the reverse side-lying position information to obtain the first position information and the second position information based on the reverse side-lying position information; Determine the first position information and the second position information of the corresponding position as the associated position, and count the second historical lesion and the second current lesion corresponding to the associated position; The first position information and the second position information based on the reverse side-lying are converted into the forward side-lying state, comprising: Retrieving a preset reverse position conversion table, and performing a front-to-side conversion on the first position information and the second position information based on the reverse position conversion table, wherein the reverse position conversion table has a corresponding angle conversion relationship between the reverse-side position and the front-side position; The first position information and the second position information based on the normal side-lying are converted into the reverse side-lying, comprising: Retrieving a preset forward and reverse position conversion table, and performing reverse-side lying conversion on the first position information and the second position information based on the forward and reverse position conversion table, wherein the forward and reverse position conversion table has a corresponding angle conversion relationship between the forward side position and the reverse side position; The step of determining the first position information and the second position information of the corresponding position as the associated position, and counting the second historical lesion and the second current lesion corresponding to the associated position, comprises: Counting the first position information and the second position information having the same intestinal information as a first statistical set; Compare the angles of lesion extraction of the first position information and the second position information in the first statistical set and the size marks preset on the gastroenteroscope; If it is determined that the absolute value of the angle difference between the corresponding first position information and the second position information is smaller than the preset angle value, and the absolute value of the size mark difference is smaller than the preset size value, the corresponding first position information and the second position information are used as the associated position; The analyzing and processing the first indication information and the second indication information of the second current lesion and the second historical lesion based on the intestinal cancer recognition model to obtain intestinal cancer recognition data includes: The first indication information and the second indication information at least include size information and type information; If the colorectal cancer recognition model determines that the size information between the second historical lesion and the second current lesion is reduced or unchanged, then the first lesion label is added to the corresponding second historical lesion; If the colorectal cancer recognition model determines that the size information between the second historical lesion and the second current lesion is enlarged, a corresponding indication change function is generated based on the first indication information and the second indication information, and a first lesion label or a second lesion label is added to the corresponding second historical lesion based on the indication change function; If the colorectal cancer recognition model determines that the type information of the second historical lesion has changed, then a second lesion label is added; Obtaining first intestinal cancer identification data based on the second historical lesion with the first lesion label or the second lesion label added; If the second historical lesion size information is enlarged, generating a corresponding indication change function based on the first indication information and the second indication information, and adding a first lesion label or a second lesion label to the corresponding second historical lesion based on the indication change function, including: Acquire the size and time corresponding to the first indication information and the second indication information respectively, and construct the indication change function corresponding to the first indication information and the second indication information based on the size and the time; Extracting a first change rate corresponding to the indication change function; If the first change rate is greater than or equal to a first preset value, adding a second lesion label to the corresponding second historical lesion, and generating a corresponding critical time based on the critical value; If the first change rate is less than a first preset value, a first lesion label is added to the corresponding second historical lesion.