A method and system for abnormal identification, labeling, and processing based on gastrointestinal endoscopic images.

Abstract

This invention discloses a method and system for abnormal identification and calibration processing based on gastrointestinal endoscopic images, belonging to the field of abnormal identification and calibration processing technology. The method involves: extracting lesion regions from endoscopic images; simultaneously acquiring endoscopic images and gastric acid value data within the same data acquisition cycle, and recording them as periodic lesion regions and periodic gastric acid value data, respectively; calculating the rate of change of periodic lesion regions; calculating the fluctuation value of periodic gastric acid value data; calculating the comprehensive abnormal value within all data acquisition cycles; calculating the abnormal difference value between different patients; setting a preset threshold; and analyzing and calibrating gastrointestinal abnormalities. By calculating the rate of change of lesion regions and the fluctuation value of gastric acid value, combined with comprehensive abnormal value analysis, this invention can comprehensively assess the structural and functional abnormalities of the digestive tract, providing more accurate diagnostic evidence. By calculating the abnormal difference value between patients, it can effectively identify group-related diseases, perform automated judgment and calibration, and improve the accuracy and efficiency of diagnosis.

Description

Technical Field

[0001] This invention relates to the field of service management technology, specifically to a method and system for abnormal identification, calibration, and processing based on gastrointestinal endoscopic images. Background Technology

[0002] With the rapid development of medical imaging technology, the application of endoscopy in the diagnosis of digestive tract diseases is becoming increasingly widespread. Gastrointestinal endoscopy allows for direct observation of the internal structures of the digestive tract, such as the esophagus, stomach, and duodenum, through visualized images, enabling the detection of abnormal lesions, such as ulcers, polyps, or tumors. Traditional endoscopic examinations typically rely on the doctor's experience, using visual observation and manual image analysis to determine digestive tract lesions. However, due to the subjectivity of manual analysis and the difficulty in processing large amounts of patient image data, automated image processing technology has gradually become a research focus. In recent years, with the continuous advancement of artificial intelligence technology, especially image recognition algorithms, the automated analysis capability of endoscopic images has been significantly improved, enabling more accurate lesion identification and analysis.

[0003] While existing automated gastrointestinal endoscopic image analysis technologies can assist doctors in disease diagnosis to some extent, several problems remain. First, current technologies focus primarily on identifying lesion areas within the images themselves, lacking comprehensive analysis with other physiological data of the patient, such as gastric acid secretion status. This may lead to incomplete judgments of lesions. Furthermore, when processing multi-period, multiple-acquisition image data, existing image processing technologies often struggle to effectively capture the changing trends of lesion areas over time, resulting in inaccurate judgments of disease progression. On the other hand, lesion characteristics among different patients are rarely correlated and analyzed in existing systems, making it impossible to discover group lesion patterns through horizontal comparisons. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for abnormal identification and calibration processing based on gastrointestinal endoscopic images, so as to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A method for abnormal identification and labeling based on gastrointestinal endoscopic images includes the following steps: dividing the memory into storage disks based on the number of patients and storing the patients' medical information in the storage disks; synchronously acquiring endoscopic images and gastric acid value data within the same data acquisition cycle; extracting lesion regions from the endoscopic images; denoting the lesion regions as periodic lesion regions and the gastric acid value data as periodic gastric acid value data; calculating the rate of change of the periodic lesion regions of the i-th patient at the t-th data acquisition cycle node; calculating the fluctuation value of the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node; calculating the comprehensive abnormal value of the i-th patient across all data acquisition cycles based on the rate of change and the fluctuation value; calculating the abnormal difference value between the i-th patient and the j-th patient based on the comprehensive abnormal value CO(i) of the i-th patient across all data acquisition cycles; and setting a preset threshold to analyze and perform gastrointestinal abnormality identification and labeling.

[0007] As a preferred embodiment of the abnormal identification and calibration processing method based on gastrointestinal endoscopic images described in this invention, a storage disk is partitioned in the memory based on the number of patients, wherein one storage disk is allocated for each patient. After authorization by the patient, the patient's diagnosis and treatment information is stored in the storage disk. The diagnosis and treatment information includes endoscopic images and gastric acid value data. The storage disk includes an endoscopic image database and a gastric acid value database, which are used to store endoscopic images and gastric acid value data, respectively.

[0008] Let G be the storage disk corresponding to the i-th patient. i Transfer storage disk G i The endoscopic image database in China is denoted as Drive G i The gastric acid value database in the middle is recorded as

[0009] As a preferred embodiment of the abnormal identification and calibration processing method based on gastrointestinal endoscopic images described in this invention, a unified data acquisition cycle is adopted. Within the same data acquisition cycle, endoscopic images and gastric acid value data are acquired simultaneously and stored separately in the endoscopic image database. And stomach acid value database In the process, image analysis technology is used to extract lesion areas from endoscopic images, denoted as Fa. The lesion area refers to the abnormal area in shape and color displayed on the endoscopic image.

[0010] At the t-th data acquisition cycle node, acquire the endoscopic image database. Database of lesion areas, Fa, and gastric acid values ​​in endoscopic images The gastric acid value data were used, and the lesion area Fa was recorded as the periodic lesion area Fa. i,t (b) Record the gastric acid value data as periodic gastric acid value data WSi,t (h), where Fa i,t (b) represents the lesion region of the i-th patient at the t-th data acquisition cycle node in the b-th cycle, WS i,t (h) represents the gastric acid value data of the i-th patient at the h-th cycle node of the t-th data collection cycle.

[0011] As a preferred embodiment of the abnormal identification and calibration processing method based on gastrointestinal endoscopic images described in this invention, the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node is calculated. i,t (b) The rate of change is calculated using the following formula:

[0012]

[0013] Where, ΔFa i,t (b) represents the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node. i,t (b) the rate of change, where B represents the total number of periodic lesion regions, Fa i,1 (b) indicates the lesion area in the initial cycle.

[0014] In this invention, the formula calculates the rate of change of each lesion area relative to the initial lesion area to obtain an average rate of change. This rate of change reflects the development of the lesion over time, whether it is expanding or shrinking; the function of this formula is to identify the trend of change in the lesion area. By calculating the rate of change of each lesion area, it is possible to determine whether the disease is worsening or improving; if the rate of change of a certain lesion area ΔFa i,t (b) A significant increase may indicate that the lesion in the area is worsening, suggesting that the doctor needs to pay further attention.

[0015] Calculate the periodic gastric acid value data (WS) of the i-th patient at the t-th data acquisition cycle node. i,t The fluctuation value of (h) is calculated using the following formula:

[0016]

[0017] Among them, ΔWS i,t (h) represents the gastric acid value data of the i-th patient at the t-th data collection node. i,t The fluctuation value of (h), where H represents the total number of gastric acid value data in the period.

[0018] In this invention, the formula considers all collected gastric acid values ​​within each cycle, reflecting the fluctuations throughout the entire cycle, rather than just the deviation of a single data point; this calculation method is more comprehensive and accurate. Compared to calculating only the difference between a single gastric acid value and the average, this formula calculates the deviation of all data points, thus avoiding misjudgments due to a single outlier or isolated event. For example, if there is a significantly abnormally high gastric acid value within a certain cycle, but other data are relatively stable, this formula can reduce the impact of a single outlier on the fluctuation results; the result of the formula can be interpreted as the overall fluctuation range of gastric acid values ​​within a cycle. If the standard deviation (i.e., the fluctuation value) is large, it indicates that the patient's gastric acid values ​​fluctuate greatly within this cycle; if the fluctuation value is small, it indicates that the gastric acid values ​​are relatively stable. Under normal circumstances, gastric acid values ​​should remain within a certain range, while abnormal fluctuations may indicate unstable gastric acid secretion or symptoms such as acid reflux.

[0019] Based on the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node. i,t (b) Rate of change ΔFa i,t (b) and the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node (WS) i,t The fluctuation value ΔWS of (h) i,t (h) Calculate the comprehensive outlier value of the i-th patient across all data collection periods, using the following formula:

[0020]

[0021] Where CO(i) represents the comprehensive abnormal value of the i-th patient in all data collection cycles, T represents the total number of data collection cycles, and α and β represent the preset weighting coefficients of the periodic endoscopic image data and periodic gastric acid value data, respectively.

[0022] In this invention, the lesion area in the endoscopic image provides direct visualization information, helping doctors observe pathological changes inside the gastrointestinal tract, such as ulcers, polyps, and inflammation. It can show morphological abnormalities or damage to the digestive tract and is a key basis for diagnosing digestive diseases. Gastric acid value data reflects the functional status of the digestive tract, especially whether gastric acid secretion is normal. Excessive or insufficient gastric acid secretion will affect digestive function, thereby causing diseases such as gastroesophageal reflux, gastritis, or ulcers. The lesion area may show lesions in the digestive tract, but when combined with gastric acid value data, doctors can further understand whether these lesions are related to abnormal gastric acid secretion. For example, excessive gastric acid may aggravate ulcers, while insufficient gastric acid may lead to indigestion. Long-term abnormal gastric acid secretion may affect the physical structure of the digestive tract. For example, long-term excessive gastric acid may corrode the stomach wall, leading to gastric ulcers or esophageal damage. The lesion area can capture these changes, and by observing the fluctuations in gastric acid value, the causes of these structural changes can be traced. Moreover, some lesions may not be obvious in the images, but functional abnormalities will be shown through fluctuations in gastric acid value. Conversely, if the imaging shows a normal digestive tract structure but significantly abnormal gastric acid levels, doctors can suspect a potential functional digestive disorder, such as functional dyspepsia or gastroesophageal reflux disease.

[0023] As a preferred embodiment of the abnormal identification and calibration processing method based on gastrointestinal endoscopic images described in this invention, the abnormal difference value between the i-th patient and the j-th patient is calculated based on the comprehensive abnormal value CO(i) of the i-th patient throughout all data acquisition cycles, using the following formula:

[0024] CY(i,j)=|CO(i)-CO(j)|;

[0025] Where CY(i,j) represents the abnormal difference value between the i-th patient and the j-th patient, and CO(j) represents the comprehensive abnormal value of the j-th patient throughout the entire data collection period.

[0026] A comprehensive outlier threshold τ and an abnormal difference threshold θ are preset. If the comprehensive outlier CO(i) of the i-th patient in all data collection periods is greater than τ, and the abnormal difference between the i-th patient and the j-th patient is less than θ, then the i-th patient is determined to have a gastrointestinal abnormality. If there is a correlation between the i-th patient and the j-th patient, then all patients who are correlated with the i-th patient are identified and labeled as having gastrointestinal abnormalities.

[0027] An abnormality identification and calibration processing system based on gastrointestinal endoscopic images is disclosed. The system includes: a data storage module, a data acquisition and synchronization module, a data analysis and abnormality detection module, and a patient abnormality association module.

[0028] The data storage module: Based on the number of patients, it divides the memory into storage disks and stores the patients' diagnosis and treatment information on the storage disks.

[0029] The data acquisition and synchronization module: within the same data acquisition cycle, synchronously acquires endoscopic images and gastric acid value data; extracts lesion areas from the endoscopic images; and records the lesion areas as periodic lesion areas and the gastric acid value data as periodic gastric acid value data.

[0030] The data analysis and anomaly detection module: calculates the rate of change of the periodic lesion area of ​​the i-th patient at the t-th data acquisition cycle node; calculates the fluctuation value of the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node; and calculates the comprehensive anomaly value of the i-th patient throughout all data acquisition cycles based on the rate of change and the fluctuation value.

[0031] The patient abnormality association module calculates the abnormality difference between the i-th patient and the j-th patient based on the comprehensive abnormality value CO(i) of the i-th patient throughout all data collection periods; it also sets a preset threshold to analyze and perform gastrointestinal abnormality identification and calibration.

[0032] Furthermore, the data storage module includes an endoscopic image and gastric acid value storage unit and a patient storage allocation unit.

[0033] The endoscopic image and gastric acid value storage unit: Based on the number of patients, the memory is divided into storage disks, with one storage disk corresponding to each patient. After authorization by the patient, the patient's diagnosis and treatment information is stored in the storage disk. The diagnosis and treatment information includes endoscopic image and gastric acid value data. The storage disk includes an endoscopic image database and a gastric acid value database, which are used to store endoscopic image and gastric acid value data, respectively.

[0034] The patient storage allocation unit: The storage disk corresponding to the i-th patient is denoted as G. i Transfer storage disk G i The endoscopic image database in China is denoted as Drive G i The gastric acid value database in the middle is recorded as

[0035] Furthermore, the data acquisition and synchronization module includes an image acquisition unit and a periodic data setting unit.

[0036] The image acquisition unit: standardizes the data acquisition cycle, synchronously acquiring endoscopic images and gastric acid value data within the same data acquisition cycle, and storing them separately in the endoscopic image database. And stomach acid value database In the process, image analysis technology is used to extract lesion areas from endoscopic images, denoted as Fa. The lesion area refers to the abnormal area in shape and color displayed on the endoscopic image.

[0037] The periodic data setting unit: at the t-th data acquisition cycle node, acquires the endoscopic image database. Database of lesion areas, Fa, and gastric acid values ​​in endoscopic images The gastric acid value data were used, and the lesion area Fa was recorded as the periodic lesion area Fa. i,t (b) Record the gastric acid value data as periodic gastric acid value data WS i,t (h), where Fa i,t (b) represents the lesion region of the i-th patient at the t-th data acquisition cycle node in the b-th cycle, WS i,t (h) represents the gastric acid value data of the i-th patient at the h-th cycle node of the t-th data collection cycle.

[0038] Furthermore, the data analysis and anomaly detection module includes a rate of change and fluctuation value calculation unit and a comprehensive anomaly value calculation unit.

[0039] The rate of change and fluctuation value calculation unit calculates the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node. i,t (b) Rate of change ΔFa i,t (b); Calculate the periodic gastric acid value data WS of the i-th patient at the t-th data acquisition cycle node. i,t The fluctuation value ΔWS of (h) i,t (h).

[0040] The comprehensive outlier calculation unit is based on the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node. i,t (b) Rate of change ΔFa i,t (b) and the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node (WS) i,t The fluctuation value ΔWS of (h) i,t (h) Calculate the comprehensive outlier CO(i) for the i-th patient across all data collection periods.

[0041] Furthermore, the patient abnormality correlation module includes an abnormality value calculation unit and a patient abnormality difference analysis unit.

[0042] The abnormal difference value calculation unit calculates the abnormal difference value CY9i,j between the i-th patient and the j-th patient based on the comprehensive abnormal value CO(i) of the i-th patient throughout all data acquisition cycles.

[0043] The patient abnormality difference analysis unit has preset comprehensive abnormality threshold τ and abnormal difference value threshold θ. If the comprehensive abnormality value CO(i) of the i-th patient in all data collection cycles is greater than τ, and the abnormal difference value CY(i,j) between the i-th patient and the j-th patient is less than θ, then the i-th patient is determined to have gastrointestinal abnormalities. If there is a correlation between the i-th patient and the j-th patient, then all patients with a correlation with the i-th patient are identified and labeled as having gastrointestinal abnormalities.

[0044] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: The method and system for abnormal identification and calibration processing based on gastrointestinal endoscopic images provided by this invention achieves individualized management of patient diagnosis and treatment information by dividing the data into independent storage disks, ensuring data security and privacy protection, while facilitating retrieval and analysis. Within a unified data acquisition cycle, endoscopic images and gastric acid value data are acquired synchronously. Lesion areas are extracted through image analysis, and changes in lesions and gastric acid values ​​are dynamically monitored. The rate of change in lesion areas and the fluctuation value of gastric acid values ​​are calculated, and combined with comprehensive abnormal value analysis, a comprehensive assessment of structural and functional abnormalities of the digestive tract can be achieved, providing more accurate diagnostic evidence. By calculating abnormal differences among patients, the system can effectively identify group-related diseases and perform automated judgment and calibration, improving the accuracy and efficiency of diagnosis, ultimately providing more comprehensive support for clinical decision-making. Attached Figure Description

[0045] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0046] Figure 1 This is a schematic diagram of the steps of an abnormality identification and calibration processing method based on gastrointestinal endoscopic images according to the present invention.

[0047] Figure 2 This is a schematic diagram of the structure of an abnormality identification and calibration processing system based on gastrointestinal endoscopic images according to the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Please see Figure 1 In this first embodiment, an abnormality identification and labeling processing method based on gastrointestinal endoscopic images is provided. The method includes the following steps:

[0050] Step S1: Based on the number of patients, divide the memory into storage disks and store the patients' diagnosis and treatment information in the storage disks.

[0051] Specifically, based on the number of patients, storage disks are partitioned in the memory, with one storage disk corresponding to each patient. After authorization by the patient, the patient's diagnosis and treatment information is stored in the storage disk. The diagnosis and treatment information includes endoscopic images and gastric acid value data. The storage disk includes an endoscopic image database and a gastric acid value database, which are used to store endoscopic images and gastric acid value data, respectively.

[0052] Furthermore, let G be the storage disk corresponding to the i-th patient. i Transfer storage disk G i The endoscopic image database in China is denoted as Drive G i The gastric acid value database in the middle is recorded as

[0053] Step S2: Within the same data acquisition cycle, simultaneously acquire endoscopic images and gastric acid value data; extract the lesion areas from the endoscopic images; record the lesion areas as periodic lesion areas and the gastric acid value data as periodic gastric acid value data.

[0054] Specifically, a unified data acquisition cycle is implemented, where endoscopic images and gastric acid value data are collected simultaneously within the same data acquisition cycle and stored separately in the endoscopic image database. And stomach acid value database In the process, image analysis technology is used to extract lesion areas from endoscopic images, denoted as Fa. The lesion area refers to the abnormal area in shape and color displayed on the endoscopic image.

[0055] Furthermore, at the t-th data acquisition cycle node, the endoscopic image database is acquired. Database of lesion areas, Fa, and gastric acid values ​​in endoscopic images The gastric acid value data were used, and the lesion area Fa was recorded as the periodic lesion area Fa. i,t (b) Record the gastric acid value data as periodic gastric acid value data WS i,t (h), where Fa i,t (b) represents the lesion region of the i-th patient at the t-th data acquisition cycle node in the b-th cycle, WS i,t (h) represents the gastric acid value data of the i-th patient at the h-th cycle node of the t-th data collection cycle.

[0056] Step S3: Calculate the rate of change of the periodic lesion area of ​​the i-th patient at the t-th data acquisition cycle node; calculate the fluctuation value of the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node; based on the rate of change and the fluctuation value, calculate the comprehensive abnormal value of the i-th patient throughout all data acquisition cycles.

[0057] Specifically, calculate the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node. i,t (b) The rate of change is calculated using the following formula:

[0058]

[0059] Where, ΔFa i,t (b) represents the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node. i,t (b) the rate of change, where B represents the total number of periodic lesion regions, Fa i,1 (b) indicates the lesion area in the initial cycle.

[0060] For example, assuming the total number of cyclic lesion regions B is 3, and the initial cyclic lesion region Fa of the first patient... 1,1 (1) = 0.6, Fa 1,1 (2) = 0.6, Fa 1,1 (3) = 06, the periodic lesion area Fa of the first patient at the 3rd data acquisition cycle node. 1,3 (1) = 0.7, Fa 1,3 (2) = 0.8, Fa 1,3 (1) = 0.9, substituting into the formula, we obtain the periodic lesion region Fa of the first patient at the third data acquisition cycle node. i,t (b) Rate of change ΔFa 1,3 (b)=0.17+0.29+0.375=0.835.

[0061] Furthermore, calculate the periodic gastric acid value data WS of the i-th patient at the t-th data acquisition cycle node. i,t The fluctuation value of (h) is calculated using the following formula:

[0062]

[0063] Among them, ΔWS i,t (h) represents the gastric acid value data of the i-th patient at the t-th data collection node. i,t The fluctuation value of (h), where H represents the total number of gastric acid value data in the period.

[0064] For example, assuming the total number of periodic gastric acid value data H is 4, the periodic gastric acid value data WS of the first patient at the 3rd data collection period node.1,3 (1) = 5, WS 1,3 (2) = 6, WS 1,3 (3) = 4, WS 1,3 (4) = 7, substituting into the formula, we obtain the fluctuation value ΔWS of the gastric acid value data of the first patient at the third data acquisition cycle node. 1,3 (h) = 0.12.

[0065] Furthermore, based on the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node... i,t (b) Rate of change ΔFa i,t (b) and the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node (WS) i,t The fluctuation value ΔWS of (h) i,t (h) Calculate the comprehensive outlier value of the i-th patient across all data collection periods, using the following formula:

[0066]

[0067] Where CO(i) represents the comprehensive abnormal value of the i-th patient in all data collection cycles, T represents the total number of data collection cycles, and α and β represent the preset weighting coefficients of the periodic endoscopic image data and periodic gastric acid value data, respectively.

[0068] For example, assuming the total number of data acquisition cycles T is 3, α and β are 0.7 and 0.8 respectively, ΔFa 1,1 (b) = 0.7, ΔWS 1,1 (h)=0.1, ΔFa 1,2 (b) = 0.8, ΔWS 1,2 (h) = 0.11, given ΔFa 1,3 (b) = 0.835, ΔWS 1,3 (h)=0.12, and substituting into the formula, we get the comprehensive outlier CO(1)=0.633.

[0069] Step S4: Based on the comprehensive outlier value CO(i) of the i-th patient throughout all data collection periods, calculate the abnormal difference value between the i-th patient and the j-th patient; preset a threshold, analyze and perform gastrointestinal abnormality identification and labeling.

[0070] Specifically, based on the comprehensive outlier CO(i) of the i-th patient throughout all data collection periods, the outlier difference between the i-th patient and the j-th patient is calculated using the following formula:

[0071] CY(i,j)=|CO(i)-CO(j)|;

[0072] Where CY(i,j) represents the abnormal difference value between the i-th patient and the j-th patient, and CO(j) represents the comprehensive abnormal value of the j-th patient throughout the entire data collection period.

[0073] Furthermore, a comprehensive outlier threshold τ and an abnormal difference threshold θ are preset. If the comprehensive outlier CO(i) of the i-th patient in all data collection periods is greater than τ, and the abnormal difference CY(i,j) between the i-th patient and the j-th patient is less than θ, then the i-th patient is determined to have a gastrointestinal abnormality. If there is a correlation between the i-th patient and the j-th patient, then all patients who are correlated with the i-th patient are identified and labeled as having gastrointestinal abnormalities.

[0074] For example, assuming j is 2, CO(2) is 0.62, the comprehensive outlier threshold τ and the outlier difference threshold θ are 0.5 and 0.1 respectively, the outlier difference value CY(1,2) = 0.013 is calculated by substituting into the formula. Given that CO(1) = 0.633, the comprehensive outlier value CO(1) > τ, and the outlier difference value CY(1,2) between the first patient and the second patient < θ, then it is determined that the first patient has a gastrointestinal abnormality, and there is a correlation between the first patient and the second patient. Therefore, the second patient is identified and labeled as having a gastrointestinal abnormality.

[0075] Please see Figure 2 In this second embodiment, an abnormal identification and calibration processing system based on gastrointestinal endoscopic images is provided. The system includes: a data storage module, a data acquisition and synchronization module, a data analysis and abnormal detection module, and a patient abnormal association module.

[0076] The data storage module: Based on the number of patients, it divides the memory into storage disks and stores the patients' diagnosis and treatment information on the storage disks.

[0077] The data acquisition and synchronization module: within the same data acquisition cycle, synchronously acquires endoscopic images and gastric acid value data; extracts lesion areas from the endoscopic images; and records the lesion areas as periodic lesion areas and the gastric acid value data as periodic gastric acid value data.

[0078] The data analysis and anomaly detection module: calculates the rate of change of the periodic lesion area of ​​the i-th patient at the t-th data acquisition cycle node; calculates the fluctuation value of the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node; and calculates the comprehensive anomaly value of the i-th patient throughout all data acquisition cycles based on the rate of change and the fluctuation value.

[0079] The patient abnormality association module calculates the abnormality difference between the i-th patient and the j-th patient based on the comprehensive abnormality value CO(i) of the i-th patient throughout all data collection periods; it also sets a preset threshold to analyze and perform gastrointestinal abnormality identification and calibration.

[0080] Furthermore, the data storage module includes an endoscopic image and gastric acid value storage unit and a patient storage allocation unit.

[0081] The endoscopic image and gastric acid value storage unit: Based on the number of patients, the memory is divided into storage disks, with one storage disk corresponding to each patient. After authorization by the patient, the patient's diagnosis and treatment information is stored in the storage disk. The diagnosis and treatment information includes endoscopic image and gastric acid value data. The storage disk includes an endoscopic image database and a gastric acid value database, which are used to store endoscopic image and gastric acid value data, respectively.

[0082] The patient storage allocation unit: The storage disk corresponding to the i-th patient is denoted as G. i Transfer storage disk G i The endoscopic image database in China is denoted as Drive G i The gastric acid value database in the middle is recorded as

[0083] Furthermore, the data acquisition and synchronization module includes an image acquisition unit and a periodic data setting unit.

[0084] The image acquisition unit: standardizes the data acquisition cycle, synchronously acquiring endoscopic images and gastric acid value data within the same data acquisition cycle, and storing them separately in the endoscopic image database. And stomach acid value database In the process, image analysis technology is used to extract lesion areas from endoscopic images, denoted as Fa. The lesion area refers to the abnormal area in shape and color displayed on the endoscopic image.

[0085] The periodic data setting unit: at the t-th data acquisition cycle node, acquires the endoscopic image database. Database of lesion areas, Fa, and gastric acid values ​​in endoscopic images The gastric acid value data were used, and the lesion area Fa was recorded as the periodic lesion area Fa. i,t (b) Record the gastric acid value data as periodic gastric acid value data WS i,t (h), where Fa i,t (b) represents the lesion region of the i-th patient at the t-th data acquisition cycle node in the b-th cycle, WS i,t (h) represents the gastric acid value data of the i-th patient at the h-th cycle node of the t-th data collection cycle.

[0086] Furthermore, the data analysis and anomaly detection module includes a rate of change and fluctuation value calculation unit and a comprehensive anomaly value calculation unit.

[0087] The rate of change and fluctuation value calculation unit calculates the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node. i,t (b) Rate of change ΔFa i,t (b); Calculate the periodic gastric acid value data WS of the i-th patient at the t-th data acquisition cycle node. i,t The fluctuation value ΔWS of (h) i,t (h).

[0088] The comprehensive outlier calculation unit is based on the periodic lesion region Fa of the i-th patient at the t-th data acquisition cycle node. i,t (b) Rate of change ΔFa i,t (b) and the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node (WS) i,t The fluctuation value ΔWS of (h) i,t (h) Calculate the comprehensive outlier CO(i) for the i-th patient across all data collection periods.

[0089] Furthermore, the patient abnormality correlation module includes an abnormality value calculation unit and a patient abnormality difference analysis unit.

[0090] The abnormal difference value calculation unit calculates the abnormal difference value CY(i,j) between the i-th patient and the j-th patient based on the comprehensive abnormal value CO(i) of the i-th patient throughout all data collection periods.

[0091] The patient abnormality difference analysis unit has preset comprehensive abnormality threshold τ and abnormal difference value threshold θ. If the comprehensive abnormality value CO(i) of the i-th patient in all data collection cycles is greater than τ, and the abnormal difference value CY(i,j) between the i-th patient and the j-th patient is less than θ, then the i-th patient is determined to have gastrointestinal abnormalities. If there is a correlation between the i-th patient and the j-th patient, then all patients with a correlation with the i-th patient are identified and labeled as having gastrointestinal abnormalities.

[0092] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0093] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for abnormal identification and labeling processing based on gastrointestinal endoscopic images, characterized in that, The method includes the following steps: Step S1: Based on the number of patients, divide the memory into storage disks and store the patients' endoscopic images and gastric acid values ​​in the storage disks; Step S2: Within the same data acquisition cycle, simultaneously acquire endoscopic images and gastric acid value data; extract lesion areas from the endoscopic images; record the lesion areas as periodic lesion areas and the gastric acid value data as periodic gastric acid value data respectively; Step S3: Calculate the rate of change of the patient's periodic lesion area at the data acquisition cycle node; calculate the fluctuation value of the patient's periodic gastric acid value data at the data acquisition cycle node; based on the rate of change and the fluctuation value, calculate the patient's comprehensive abnormal value throughout all data acquisition cycles; Step S4: Based on the comprehensive abnormal values ​​of patients throughout the entire data collection period, calculate the abnormal difference values ​​between different patients, analyze and perform gastrointestinal abnormality identification and labeling; The specific implementation process of step S2 includes: A standardized data acquisition cycle is implemented, during which endoscopic images and gastric acid value data are acquired synchronously and stored separately in the endoscopic image database. And stomach acid value database In, among them, This represents the storage disk partition corresponding to the i-th patient. Indicates storage disk Endoscopic image database in China Indicates storage disk The database of gastric acid values; using image analysis technology to extract lesion areas from endoscopic images, denoted as Fa, where the lesion area refers to the abnormal area in shape and color displayed on the endoscopic image; At the t-th data acquisition cycle node, acquire the endoscopic image database. Database of lesion areas, Fa, and gastric acid values ​​in endoscopic images The gastric acid value data were used, and the lesion area Fa was recorded as the periodic lesion area. The gastric acid value data is recorded as periodic gastric acid value data. ,in, This represents the lesion region of the i-th patient at the b-th cycle node of the t-th data acquisition cycle. This represents the gastric acid value data of the i-th patient at the h-th data collection cycle node in the t-th data collection cycle. The specific implementation process of step S3 includes: Calculate the periodic lesion region of the i-th patient at the t-th data acquisition cycle node. The rate of change is calculated using the following formula: ； in, This represents the periodic lesion region of the i-th patient at the t-th data acquisition cycle node. The rate of change, where B represents the total number of lesion regions in the periodic cycle. Indicates the lesion area in the initial cycle; Calculate the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node. The fluctuation value is calculated using the following formula: ； in, This represents the periodic gastric acid value data of the i-th patient at the t-th data collection period node. The fluctuation value, H represents the total number of gastric acid value data in the period; Based on the periodic lesion area of ​​the i-th patient at the t-th data acquisition cycle node. rate of change and the periodic gastric acid value data of the i-th patient at the t-th data collection cycle node. fluctuation value Calculate the overall outlier value for the i-th patient across all data collection periods using the following formula: ； in, This represents the total outlier values ​​for the i-th patient across all data collection periods, where T represents the total number of data collection periods. and These represent the weighting coefficients for the preset periodic endoscopic image data and periodic gastric acid value data, respectively.

2. The abnormal identification and labeling processing method based on gastrointestinal endoscopic images according to claim 1, characterized in that, The specific implementation process of step S1 includes: Based on the number of patients, storage disks are partitioned in the memory, with one storage disk corresponding to each patient. After authorization by the patient, the patient's diagnosis and treatment information is stored in the storage disk. The diagnosis and treatment information includes endoscopic images and gastric acid value data. The storage disk includes an endoscopic image database and a gastric acid value database, which are used to store endoscopic images and gastric acid value data, respectively. Let the storage disk corresponding to the i-th patient be denoted as . , storage disk The endoscopic image database in China is denoted as , storage disk The gastric acid value database in the middle is recorded as .

3. The abnormal identification and labeling processing method based on gastrointestinal endoscopic images according to claim 2, characterized in that, The specific implementation process of step S4 includes: Based on the comprehensive outlier values ​​of the i-th patient throughout all data collection periods Calculate the abnormal difference value between the i-th patient and the j-th patient using the following formula: ； in, This represents the abnormal difference value between the i-th patient and the j-th patient. This represents the total outlier values ​​for the j-th patient across all data collection periods; Preset comprehensive outlier threshold and abnormal difference threshold If the overall outlier value of the i-th patient during all data collection periods is... And the abnormal difference value between the i-th patient and the j-th patient If the i-th patient is found to have a gastrointestinal abnormality and there is a correlation between the i-th patient and the j-th patient, then all patients who are related to the i-th patient will be identified and labeled as having gastrointestinal abnormalities.

4. An abnormality identification and calibration processing system based on gastrointestinal endoscopic images, executing the abnormality identification and calibration processing method based on gastrointestinal endoscopic images as described in any one of claims 1-3, characterized in that, The system includes: a data storage module, a data acquisition and synchronization module, a data analysis and anomaly detection module, and a patient anomaly association module; The data storage module: based on the number of patients, divides the memory into storage disks and stores the patients' diagnosis and treatment information on the storage disks; The data acquisition and synchronization module: synchronously acquires endoscopic images and gastric acid value data within the same data acquisition cycle; extracts lesion areas from the endoscopic images; and records the lesion areas as periodic lesion areas and the gastric acid value data as periodic gastric acid value data. The data analysis and anomaly detection module: calculates the rate of change of the periodic lesion area of ​​the i-th patient at the t-th data acquisition cycle node; calculates the fluctuation value of the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node; and calculates the comprehensive anomaly value of the i-th patient throughout all data acquisition cycles based on the rate of change and the fluctuation value. The patient anomaly association module is based on the comprehensive anomaly values ​​of the i-th patient across all data acquisition periods. Calculate the abnormal difference value between the i-th patient and the j-th patient; set a preset threshold, analyze and perform gastrointestinal abnormality identification and labeling.

5. The abnormal identification and calibration processing system based on gastrointestinal endoscopic images according to claim 4, characterized in that: The data storage module includes an endoscopic image and gastric acid value storage unit and a patient storage allocation unit; The endoscopic image and gastric acid value storage unit: Based on the number of patients, the memory is divided into storage disks, with one storage disk corresponding to each patient. After authorization by the patient, the patient's diagnosis and treatment information is stored in the storage disk. The diagnosis and treatment information includes endoscopic image and gastric acid value data. The storage disk includes an endoscopic image database and a gastric acid value database, which are used to store endoscopic image and gastric acid value data, respectively. The patient storage allocation unit: The storage disk corresponding to the i-th patient is denoted as... , storage disk The endoscopic image database in China is denoted as , storage disk The gastric acid value database in the middle is recorded as .

6. The abnormal identification and calibration processing system based on gastrointestinal endoscopic images according to claim 5, characterized in that: The data acquisition and synchronization module includes an image acquisition unit and a periodic data setting unit; The image acquisition unit: standardizes the data acquisition cycle, synchronously acquiring endoscopic images and gastric acid value data within the same data acquisition cycle, and storing them separately in the endoscopic image database. And stomach acid value database In the process of extracting lesion areas from endoscopic images using image analysis technology, denoted as Fa, the lesion areas refer to abnormal areas of shape and color displayed on endoscopic images; The periodic data setting unit: at the t-th data acquisition cycle node, acquires the endoscopic image database. Database of lesion areas, Fa, and gastric acid values ​​in endoscopic images The gastric acid value data were used, and the lesion area Fa was recorded as the periodic lesion area. The gastric acid value data is recorded as periodic gastric acid value data. ,in, This represents the lesion region of the i-th patient at the b-th cycle node of the t-th data acquisition cycle. This represents the gastric acid value data of the i-th patient at the h-th cycle node of the t-th data collection cycle.

7. The abnormal identification and calibration processing system based on gastrointestinal endoscopic images according to claim 6, characterized in that: The data analysis and anomaly detection module includes a rate of change and fluctuation value calculation unit and a comprehensive anomaly value calculation unit; The rate of change and fluctuation value calculation unit calculates the periodic lesion area of ​​the i-th patient at the t-th data acquisition cycle node. rate of change ; Calculate the periodic gastric acid value data of the i-th patient at the t-th data acquisition cycle node. fluctuation value ; The comprehensive outlier calculation unit is based on the periodic lesion region of the i-th patient at the t-th data acquisition cycle node. rate of change and the periodic gastric acid value data of the i-th patient at the t-th data collection cycle node. fluctuation value Calculate the comprehensive outlier value of the i-th patient across all data collection periods. .

8. The abnormal identification and calibration processing system based on gastrointestinal endoscopic images according to claim 7, characterized in that: The patient abnormality correlation module includes an abnormality value calculation unit and a patient abnormality difference analysis unit; The abnormal difference value calculation unit is based on the comprehensive abnormal value of the i-th patient throughout all data acquisition periods. Calculate the abnormal difference value between the i-th patient and the j-th patient. ; The patient abnormality analysis unit: preset comprehensive abnormal value threshold. and abnormal difference threshold If the overall outlier value of the i-th patient during all data collection periods is... And the abnormal difference value between the i-th patient and the j-th patient If the i-th patient is found to have a gastrointestinal abnormality and there is a correlation between the i-th patient and the j-th patient, then all patients who are related to the i-th patient will be identified and labeled as having gastrointestinal abnormalities.

Images