Gastrointestinal bleeding prediction analysis system and method based on information planning

Abstract

The application relates to the field of medical systems, in particular to a gastrointestinal hemorrhage prediction analysis system and method based on information planning, which analyzes the damage of food to the digestive tract and the influence of gastrointestinal hemorrhage abnormalities on the food through a multi-source data fusion method, judges the damage risk of food to the digestive tract, improves the edible safety of food, analyzes the chewing habit through a multi-layer perception machine neural network model, evaluates the influence of chewing on food, and then judges the damage of food to the digestive tract after chewing, and improves the accuracy of the damage risk judgment of food to the digestive tract.

Description

Technical Field

[0001] This application relates to the field of medical system technology, and in particular to a system and method for predicting and analyzing gastrointestinal bleeding based on information planning. Background Technology

[0002] Gastrointestinal bleeding is a common and serious clinical problem with complex causes closely related to factors such as food intake and the health of the digestive tract. Traditional methods for assessing gastrointestinal bleeding often rely on a single data source, such as focusing only on the pathological condition of the digestive tract itself or considering only certain characteristics of food, making it difficult to comprehensively and accurately predict the risk of bleeding. Moreover, research on the interaction between food and the digestive tract is insufficient, making it impossible to accurately determine the impact of different foods on the individual's digestive tract, resulting in limitations in dietary recommendations and risk warnings.

[0003] Current technologies for assessing the impact of food on the digestive tract lack comprehensive utilization of multi-source data. On the one hand, they fail to adequately consider factors such as the patient's oral chewing habits and food characteristics, making it difficult to accurately predict the state of food after chewing and the extent of damage to the digestive tract. On the other hand, assessments of the digestive tract do not effectively integrate information on lesions, vascular distribution, blood flow, and other aspects, resulting in inaccurate predictions of bleeding risk. Furthermore, the lack of a scientific and systematic risk assessment system in determining whether food is suitable for a patient prevents the provision of personalized dietary advice and timely food warnings.

[0004] In view of this, the applicant proposes a gastrointestinal bleeding prediction and analysis system and method based on information planning. Summary of the Invention

[0005] To overcome the defects and shortcomings of existing technologies, this application provides a gastrointestinal bleeding prediction and analysis system and method based on information planning. This application analyzes the damage of food to the digestive tract and the impact of abnormal gastrointestinal bleeding by means of multi-source data fusion, judges the risk of food damage to the digestive tract, and improves the safety of food consumption. By constructing a multilayer perceptron neural network model to analyze chewing habits, it assesses the impact of chewing on food, and then judges the damage of food to the digestive tract after chewing, thereby improving the accuracy of judging the risk of food damage to the digestive tract.

[0006] To achieve the above objectives, this application adopts the following technical solution: In the first aspect, this application provides a method for predicting and analyzing gastrointestinal bleeding based on information planning, comprising the following steps: S100. Obtain information on the patient's digestive tract and corresponding oral chewing habits, and also obtain information on the corresponding food. S200. Based on the patient's oral chewing habits and the corresponding food conditions, predict any abnormalities after food chewing. S300. Analyze the damage to the digestive tract caused by chewed food, taking into account the condition of the food after chewing and the condition of the digestive tract wall. S400, combined with the condition of gastrointestinal injury and the distribution of blood vessels in the corresponding location, to predict gastrointestinal bleeding. S500, in conjunction with the predicted gastrointestinal bleeding and damage, assesses the suitability of food and issues warnings for the corresponding foods.

[0007] In one implementation of this application, step S100 includes the following specific contents: Step 110: Obtain the internal condition of the patient's digestive tract through gastrointestinal endoscopy, including images of the patient's internal digestive tract and the distribution of blood vessels in the digestive tract. Construct a three-dimensional image of the patient's digestive tract using medical image reconstruction software for reference. Step 120: Obtain the type, shape, hardness, and volume of the corresponding food. The type of food is obtained through human input or image recognition. The hardness of the food is obtained through a hardness sensor. The volume and shape can be obtained through a 3D image acquisition terminal and stored. Step 130: By acquiring the patient's oral chewing habits, the chewing time and chewing force of the corresponding type of food are obtained. The oral enzyme activity and oral pH of the patient are also acquired through oral chewing habit statistics. The chewing and digestion of food are analyzed by analyzing the patient's oral enzyme activity, oral pH, and chewing. At the same time, the corresponding oral enzyme activity and oral pH are acquired through corresponding sensors and stored in corresponding storage components.

[0008] In one implementation of this application, the step S200 of predicting abnormal conditions after food chewing includes the following specific steps: Step 210: Obtain the type, shape, hardness, and volume of historical corresponding foods, as well as the chewing time and chewing force of historical patients for corresponding types of foods, the oral enzyme activity and oral pH of historical patients, and the shape and hardness of historical chewed foods. Construct a multilayer perceptron neural network model with the input of the type, shape, hardness, and volume of historical corresponding foods, the chewing time and chewing force of corresponding types of foods, the oral enzyme activity and oral pH of patients, and the output of the size and hardness of chewed foods. The specific steps of the multilayer perceptron neural network model are as follows: Divide the historical data into a 70% training set and a 30% test and validation set. Input the 70% weight and bias training set into the neural network model for training to obtain the initial multilayer perceptron neural network model. Use the 30% weight and bias test set to test the initial multilayer perceptron neural network model and output the initial multilayer perceptron neural network model output that satisfies the preset maximum accuracy of the size and hardness of the chewed food as the neural network model output. Step 220: Input the currently collected information on the type, shape, hardness, and volume of the corresponding food, the chewing time and force of the corresponding food, the patient's oral enzyme activity, and the oral pH into the corresponding multilayer perceptron neural network model to output the size and hardness of the chewed food. Step 230: Obtain the abnormal food chewing value by weighting and summing the ratios of the size and hardness of the chewed food to the corresponding safe food data. The ratios of the size and hardness of the chewed food to the corresponding safe food data are: the ratio of the size of the corresponding food to the size of the safe food, and the ratio of the hardness of the corresponding food to the hardness of the safe food.

[0009] In one implementation of this application, the damage to the digestive tract caused by chewed food in step S300 includes the following specific aspects: Step 310: Obtain the lesion image of the digestive tract of the corresponding personnel, and at the same time obtain the safe puncture pressure and the inner diameter of the digestive tract at each location. Obtain the abnormal value of the lesion by the average deviation between the pixel value of each point in the lesion image of the digestive tract and the pixel value of the location without lesion. Obtain the original damage value of the digestive tract by multiplying the abnormal value of the lesion by the lesion influence coefficient. Step 320: Obtain the safe puncture pressure and digestive tract diameter at the corresponding location. Obtain the pressure abnormality by dividing the standard pressure by the safe puncture pressure at the corresponding location, and obtain the digestive tract diameter abnormality by dividing the standard digestive tract diameter by the digestive tract diameter at the corresponding location. Step 330: The digestive tract vulnerability value and the food chewing abnormal value are weighted and summed to obtain the abnormal damage result of chewed food to the digestive tract. The abnormal damage result of chewed food to the digestive tract at all locations is summed to obtain the overall damage result.

[0010] In one implementation of this application, the gastrointestinal bleeding prediction in step S400 includes the following specific details: This method acquires information on the distribution of blood vessels and blood flow in various locations within the digestive tract. It also obtains abnormal results regarding the damage to the digestive tract caused by chewed food at each location. These abnormal results are compared with predefined abnormality thresholds. Locations where the abnormality is greater than or equal to the threshold are designated as abnormal digestive segments, while those less than the threshold are designated as safe digestive segments. By acquiring information on the abnormality, blood vessel distribution, and blood flow in these segments, a comprehensive understanding of the digestive tract's condition can be obtained. Furthermore, by acquiring this information on blood vessel distribution, blood flow, and abnormal food damage at various locations within the digestive tract, a comprehensive and detailed understanding of the entire digestive tract's physiological state and its impact on the digestive system can be achieved. Comparing the abnormal results with predefined thresholds helps to quickly and accurately locate potentially problematic areas within the digestive tract. The vascular depth hazard value is obtained by dividing the safe vascular depth by the vascular depth of the corresponding abnormal digestive segment. The vascular flow rate hazard value is obtained by dividing the vascular flow rate of the corresponding abnormal digestive segment by the safe vascular flow rate. The vascular depth hazard value and the vascular flow rate hazard value are weighted and summed to obtain the blood flow abnormality of the corresponding abnormal digestive segment. Multiplying the blood flow abnormality and injury abnormality results of the corresponding abnormal digestive segment yields the bleeding abnormality prediction for the corresponding abnormal digestive segment. Summing up the bleeding abnormality predictions of all corresponding abnormal digestive segments yields the gastrointestinal bleeding abnormality. Multiplying the blood flow abnormality and injury abnormality results of the corresponding abnormal digestive segment yields the bleeding abnormality prediction.

[0011] In one implementation of this application, step S500 involves combining the predicted gastrointestinal bleeding and damage to assess the risk of food suitability and issue a warning for the corresponding food, including the following specific details: The system obtains the overall damage results and abnormal gastrointestinal bleeding caused by the consumption of various foods, compares them with the corresponding thresholds, selects foods whose overall damage results and abnormal gastrointestinal bleeding are both less than or equal to the corresponding thresholds as safe foods, and classifies other foods as dangerous foods for early warning.

[0012] Secondly, this application also provides an information planning-based gastrointestinal bleeding prediction and analysis system, including: The data acquisition module acquires information about the patient's digestive tract and corresponding oral chewing habits, as well as information about the corresponding food. The food chewing analysis module uses the patient's oral chewing habits and the corresponding food conditions to predict abnormalities after food chewing. The food damage prediction module analyzes the damage to the digestive tract after chewing by combining the condition of the food after chewing and the condition of the digestive tract wall. The bleeding prediction module combines the condition of gastrointestinal injury with the distribution of blood vessels in the corresponding location to predict gastrointestinal bleeding. The risk assessment module combines the predicted gastrointestinal bleeding and damage to determine the suitability of food and issue warnings for the corresponding foods.

[0013] Thirdly, this application provides an electronic device comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes an information planning-based gastrointestinal bleeding prediction and analysis method by calling the computer program stored in the memory.

[0014] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform an information planning-based gastrointestinal bleeding prediction and analysis method.

[0015] Compared with the prior art, this application has the following advantages and beneficial effects: By using multi-source data fusion to analyze the impact of food on the digestive tract and abnormal gastrointestinal bleeding, the risk of food damage to the digestive tract can be assessed, thereby improving the safety of food consumption. By constructing a multilayer perceptron neural network model to analyze chewing habits, assess the impact of chewing on food, and then determine the damage of food to the digestive tract after chewing, the accuracy of assessing the risk of food damage to the digestive tract is improved. Attached Figure Description

[0016] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the overall process structure of an embodiment of the method of this application; Figure 2 This is a schematic diagram of the process structure for generating a three-dimensional model of the digestive tract in an embodiment of the method of this application; Figure 3 This is a schematic diagram of the flow structure of the multilayer perceptron neural network model in the embodiment of the method of this application; Figure 4 This is a schematic diagram of the flow structure of step S300 in the method embodiment of this application; Figure 5 This is a schematic diagram of the module composition structure of an embodiment of the system in this application. Detailed Implementation

[0017] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0018] Please see Figures 1 to 4 , Figure 1 This is a schematic diagram of the overall process of the information planning-based gastrointestinal bleeding prediction and analysis method provided in the embodiments of this application, which specifically includes the following steps: S100. Obtain information on the patient's digestive tract and corresponding oral chewing habits, and also obtain information on the corresponding food. In this embodiment, step S100 includes the following specific contents: Step 110: Obtain the internal condition of the patient's digestive tract through gastrointestinal endoscopy, including images of the patient's internal digestive tract and the distribution of blood vessels within the digestive tract. Construct a three-dimensional image of the patient's digestive tract using medical image reconstruction software for reference. Specific details include: Figure 2 As shown, firstly, the patient's digestive tract is explored using a digestive endoscope (such as a gastroscopy or colonoscopy) to obtain high-resolution images and vascular distribution data (such as narrow-band imaging / NBI-enhanced vascular observation). Subsequently, the acquired two-dimensional image sequence is imported into medical image processing software (such as 3D Slicer, Mimics, etc.), and a three-dimensional model of the digestive tract wall and vascular network is generated through image registration, segmentation, and three-dimensional reconstruction algorithms (such as volume rendering and surface drawing). At the same time, the limit pressure that the digestive tract mucosa of patients of corresponding ages can withstand during puncture is measured through in vitro tissue experiments and set as the safe puncture pressure. Step 120: Obtain the type, shape, hardness, and volume of the corresponding food. The type of food is obtained through human input or image recognition. The hardness of the food is obtained through a hardness sensor. The volume and shape can be obtained through a 3D image acquisition terminal and stored. Step 130: By acquiring the corresponding patient's oral chewing habits, the chewing time and chewing force of the corresponding type of food are obtained. The oral enzyme activity and oral pH of the patient are also acquired through oral chewing habit statistics. The chewing and digestion of food are analyzed by analyzing the patient's oral enzyme activity, oral pH, and chewing. At the same time, the corresponding oral enzyme activity and oral pH are acquired through the corresponding sensors and stored in the corresponding storage components. S200. Based on the patient's oral chewing habits and the corresponding food conditions, predict any abnormalities after food chewing. In this embodiment, step S200, which estimates abnormal conditions after food chewing, includes the following specific steps: Step 210: Obtain the type, shape, hardness, and volume of historical corresponding foods, as well as the chewing time and force of historical patients for these foods. Also obtain the oral enzyme activity and oral pH levels of historical patients, and the shape and hardness of chewed foods. Construct a multilayer perceptron neural network model with the inputs of historical corresponding food type, shape, hardness, volume, chewing time, chewing force, oral enzyme activity, and oral pH levels, and the output of the size and hardness of chewed foods. Figure 3 As shown, the specific steps include: First, the heterogeneous data needs to be cleaned and normalized. The physical properties of food (such as hardness and volume) are scaled to the [0,1] interval using Min-Max. Categorical variables (such as food types) are converted into sparse matrices through one-heat encoding. Oral biochemical parameters (such as pH value and amylase activity) are standardized using Z-score due to their large dimensional differences. For high-dimensional features (such as 3D scanned food morphology point clouds), principal component analysis (PCA) is used to retain 95% of the variance components, compressing the original thousands of dimensions of data to 10-50 dimensions, which reduces the amount of computation and avoids overfitting. Time series data (such as chewing force dynamic curves) can be used to extract time-domain features. The model incorporates features (mean, peak value) and frequency domain features (FFT energy spectrum) to enhance its ability to capture dynamic processes. The input layer dimension is determined by the number of features, including a total of 8 dimensions such as food type, shape, hardness, volume, chewing time and force for the corresponding food type, oral enzyme activity, and oral pH. The hidden layer uses a three-layer MLP structure (128, 64, and 32 neurons respectively), with each layer connected to a ReLU activation function to accelerate convergence and Dropout (ratio 0.3) set to prevent overfitting. The output layer is designed differently according to the task type: predicting continuous variables of food shape and hardness after chewing, using linear activation. Historical data was divided into a 70% training set and a 30% test / validation set. The 70% weight and bias training set was input into the neural network model for training to obtain an initial multilayer perceptron neural network model. The initial multilayer perceptron neural network model was tested using the 30% weight and bias test set. The output of the initial multilayer perceptron neural network model that satisfies the preset maximum accuracy of the size and hardness of chewed food was used as the neural network model output. The morphology of chewed food is affected by the initial physical properties (hardness, viscoelasticity), oral motion (tongue pressure, biting force), and biochemical environment (salivary enzymatic hydrolysis, pH). MLP can model high-dimensional nonlinear relationships (such as the interaction between enzyme activity and the degradation rate of starchy foods), which is superior to traditional statistical models. Step 220: Input the currently collected information on the type, shape, hardness, and volume of the corresponding food, the chewing time and force of the corresponding food, the patient's oral enzyme activity, and the oral pH into the corresponding multilayer perceptron neural network model to output the size and hardness of the chewed food. Step 230: Obtain the abnormal food chewing value by weighted summing the ratios of the size and hardness of the chewed food to the corresponding safe food data. The ratios are: the ratio of the size of the chewed food to the size of the safe food, and the ratio of the hardness of the chewed food to the hardness of the safe food. The safe food hardness and size are obtained by medical personnel based on assessments of foods that will not cause harm to ordinary people. The safe food size is the inner diameter of narrowed areas of the digestive tract. The process is modeled using a multilayer perceptron (MLP). The nonlinear relationship between individual oral parameters (such as salivary enzyme activity and biting force) and food characteristics, real-time prediction of food hardness and particle size after chewing, and calculation of outliers to quantify swallowing risk are beneficial to improving the accuracy of swallowing risk assessment. The weights here are obtained as follows: observe the effects of foods of different sizes and hardness on human swallowing and digestion after chewing. For example, select a large number of subjects of different ages and health conditions, have them eat foods with different characteristics, record the probability of swallowing difficulties, indigestion, etc., and determine the relative importance of size ratio and hardness ratio in affecting swallowing risk based on these statistical data, and then determine the weights. S300. Analyze the damage to the digestive tract caused by chewed food, taking into account the condition of the food after chewing and the condition of the digestive tract wall. In this embodiment, as Figure 4 As shown, the damage to the digestive tract caused by chewed food in step S300 includes the following specific details: Step 310: Obtain images of lesions in the digestive tract of the corresponding personnel, and simultaneously obtain the safe puncture pressure and internal diameter of the digestive tract at each location. Obtain the abnormal lesion value by the average deviation between the pixel values ​​of each point in the lesion image and the pixel values ​​of locations without lesions. Obtain the original damage value of the digestive tract by multiplying the abnormal lesion value by the lesion influence coefficient. The pixel values ​​of locations without lesions are obtained by image restoration. The lesion influence coefficient is obtained by averaging after statistics. The location is divided into at least twenty digestive segments of the same length, with one digestive segment corresponding to one location. The number of locations for each personnel is set to be the same to ensure comparability. Step 320: Obtain the safe puncture pressure and gastrointestinal diameter at the corresponding location. The pressure anomaly is obtained by dividing the standard pressure by the safe puncture pressure at the corresponding location, and the gastrointestinal diameter anomaly is obtained by dividing the standard gastrointestinal diameter by the gastrointestinal diameter at the corresponding location. The standard pressure and standard gastrointestinal diameter are averages from a sample of all ages. For children or the elderly, the safe puncture pressure and gastrointestinal diameter at the corresponding location may be lower, making them more susceptible to injury. The pressure anomaly and the gastrointestinal diameter anomaly are then weighted and summed to obtain the gastrointestinal tolerance anomaly. Typically, the gastrointestinal vulnerability value is obtained by weighted summing of abnormal gastrointestinal tolerance and original gastrointestinal damage values. It should be noted that when calculating abnormal gastrointestinal tolerance, it is necessary to perform weighted summation of abnormal pressure and abnormal gastrointestinal diameter. This can be achieved by collecting a large amount of case data on different gastrointestinal health conditions and analyzing the degree of influence of abnormal pressure and abnormal diameter on gastrointestinal vulnerability. For example, in one embodiment, statistical methods such as regression analysis are used, with gastrointestinal vulnerability value as the dependent variable and abnormal pressure and abnormal diameter as independent variables, to fit a regression equation. The coefficients of the independent variables in the equation are the corresponding weights. Step 330: The digestive tract vulnerability value and the abnormal food chewing value are weighted and summed to obtain the abnormal damage result of chewed food to the digestive tract. The abnormal damage results of chewed food to the digestive tract at all locations are summed to obtain the overall damage result. The digestive tract vulnerability value reflects the health status and tolerance of the digestive tract itself, while the abnormal food chewing value reflects the processing of food before it enters the digestive tract. Both factors affect the damage of food to the digestive tract. By using a weighted summation method, these two important factors can be combined to comprehensively assess the degree of damage to the digestive tract. It should be noted that the weighted summation of the digestive tract vulnerability value and the abnormal food chewing value to obtain the abnormal damage result of chewed food to the digestive tract can also be achieved using a combination of clinical research and expert experience. On the one hand, through follow-up studies of a large number of patients, the combined impact of digestive tract vulnerability value and abnormal food chewing value on digestive tract damage can be analyzed; on the other hand, expert opinions on the importance of these two factors in damage assessment can be solicited to determine appropriate weights. S400, combined with the condition of gastrointestinal injury and the distribution of blood vessels in the corresponding location, to predict gastrointestinal bleeding. In this embodiment, the gastrointestinal bleeding prediction in step S400 includes the following specific details: This method acquires information on the distribution of blood vessels and blood flow in various locations within the digestive tract. It also obtains abnormal results regarding the damage to the digestive tract caused by chewed food at each location. These abnormal results are compared to preset abnormality thresholds. Locations where the abnormality is greater than or equal to the threshold are designated as abnormal digestive segments, while those less than the threshold are designated as safe digestive segments. The method acquires information on the abnormality, blood vessel distribution, and blood flow in these segments to provide a comprehensive understanding of the digestive tract's condition. By acquiring information on the distribution of blood vessels and blood flow in various locations within the digestive tract... By analyzing abnormal results of food damage to the digestive tract, a comprehensive and detailed understanding of the physiological state of the entire digestive tract and its impact on the digestive system can be obtained. By comparing the abnormal results of damage at each location with set thresholds, abnormal digestive segments and safe digestive segments can be identified. This helps to quickly and accurately locate potentially problematic areas in the digestive tract, providing a basis for subsequent targeted analysis and treatment. Different locations in the digestive tract have different structures and functions, and their blood vessel distribution and blood flow also vary. During the digestion process of food at different locations in the digestive tract, the degree of damage to the digestive tract will also be different. By quantifying abnormal results of damage and comparing them with thresholds, it is possible to objectively determine whether a location is in an abnormal state. The vascular depth hazard value is obtained by dividing the safe vascular depth by the vascular depth of the corresponding abnormal digestive segment. The vascular flow hazard value is obtained by dividing the vascular flow rate of the corresponding abnormal digestive segment by the safe vascular flow rate. The blood flow anomaly of the corresponding abnormal digestive segment is obtained by weighted summing of the vascular depth hazard value and the vascular flow hazard value. Here, the safe vascular depth is the average depth of human blood vessels, and the safe vascular flow rate is the average flow rate of human blood vessels. By calculating the vascular depth hazard value and the vascular flow hazard value, the vascular condition of the abnormal digestive segment is compared with the average level in the human body, allowing for a quantitative assessment of the risk level faced by the blood vessels in that area. The blood flow anomaly is obtained by weighted summing of the vascular depth hazard value and the vascular flow hazard value, comprehensively considering both vascular depth and flow rate. Key factors, more comprehensively reflecting the blood flow status of abnormal digestive segments, provide a more accurate basis for subsequent assessment of bleeding risk. Vascular depth and blood flow are important factors affecting vascular stability and bleeding risk. Excessively shallow vascular depth or abnormally increased blood flow may make vessels more susceptible to damage, thus increasing the likelihood of bleeding. Using the average human level as a reference standard allows for a more objective assessment of the risk level of vessels in abnormal digestive segments. When calculating the blood flow abnormality of the corresponding abnormal digestive segment, a weighted sum of vascular depth and vascular flow risk values ​​is required. Weights can be obtained through animal experiments or human simulation experiments to simulate bleeding conditions caused by gastrointestinal injury under different vascular depths and flow rates. Measuring indicators such as bleeding volume under different conditions and analyzing the impact of vascular depth and flow rate on bleeding risk are also crucial for determining the weights. Multiplying the blood flow abnormality and damage abnormality results of the corresponding abnormal digestive segment yields the bleeding abnormality prediction for that segment. Summing up the bleeding abnormality predictions of all corresponding abnormal digestive segments yields the gastrointestinal bleeding abnormality. Multiplying the blood flow abnormality and damage abnormality results of the corresponding abnormal digestive segment yields the bleeding abnormality prediction. By comprehensively considering vascular condition and gastrointestinal damage, the bleeding risk of the abnormal digestive segment can be more accurately predicted. Summing up the bleeding abnormality predictions of all corresponding abnormal digestive segments yields the gastrointestinal bleeding abnormality, which provides an overall assessment of the bleeding risk of the entire gastrointestinal tract and offers important reference for clinicians to formulate treatment plans and assess the severity of the condition. S500, combining the prediction of gastrointestinal bleeding and damage, makes a risk assessment of whether food is suitable and issues warnings for the corresponding food. This embodiment includes the following specific contents: The overall damage and gastrointestinal bleeding abnormalities resulting from the consumption of various foods are obtained and compared with corresponding thresholds. Foods with both overall damage and gastrointestinal bleeding abnormalities less than or equal to the corresponding thresholds are designated as safe foods, while other foods are designated as dangerous foods. This serves as a warning for dangerous foods. The thresholds are obtained by surveying a large number of healthy individuals, recording the overall damage and gastrointestinal bleeding abnormalities in their digestive tract after consuming various foods, and statistically analyzing the distribution characteristics of these data, such as the mean and standard deviation. Typically, the threshold is set at a certain percentile of the normal population data, such as the 95th percentile, and conditions exceeding this value are considered abnormal.

[0019] The advantages of the above embodiments are: by analyzing the impact of food on the digestive tract and abnormal gastrointestinal bleeding through multi-source data fusion, the risk of food damage to the digestive tract can be assessed, thereby improving the safety of food consumption.

[0020] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of the information planning-based gastrointestinal bleeding prediction and analysis system provided in this application embodiment, including: The data acquisition module acquires information about the patient's digestive tract and corresponding oral chewing habits, as well as information about the corresponding food. The food chewing analysis module uses the patient's oral chewing habits and the corresponding food conditions to predict abnormalities after food chewing. The food damage prediction module analyzes the damage to the digestive tract after chewing by combining the condition of the food after chewing and the condition of the digestive tract wall. The bleeding prediction module combines the condition of gastrointestinal injury with the distribution of blood vessels in the corresponding location to predict gastrointestinal bleeding. The risk assessment module combines the predicted gastrointestinal bleeding and damage to determine the suitability of food and issue warnings for the corresponding foods.

[0021] The steps for implementing the corresponding functions of each parameter and unit module in the information planning-based gastrointestinal bleeding prediction and analysis system of this application can be referred to the parameters and steps in the embodiments of the information planning-based gastrointestinal bleeding prediction and analysis method described above, and will not be repeated here.

[0022] at the same time, Figure 5 The arrow in the image indicates the direction of data transmission.

[0023] Embodiments of this application also provide an electronic device, including a memory, a processor, and a communication bus; the memory and the processor are connected via the communication bus. The memory stores a gastrointestinal bleeding prediction and analysis method based on information planning, as provided in the above embodiments, which can be loaded and executed by the processor 320.

[0024] The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory 310 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the information planning-based gastrointestinal bleeding prediction and analysis method provided in the above embodiments, etc. The data storage area may store data involved in the information planning-based gastrointestinal bleeding prediction and analysis method provided in the above embodiments, etc.

[0025] The processor may include one or more processing cores. The processor executes instructions, programs, code sets, or instruction sets stored in memory, and calls data stored in memory to perform various functions and process data as described in this application. The processor may be at least one of the following: Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), controller, microcontroller, and microprocessor. It is understood that, for different devices, the electronic devices used to implement the functions of the processor 320 described above may also be other types, and this application embodiment does not specifically limit the specific devices used.

[0026] A communication bus may include a pathway for transmitting information between the aforementioned components. The communication bus 330 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Communication buses can be categorized as address buses, data buses, control buses, etc.

[0027] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in the above embodiments for the prediction and analysis of gastrointestinal bleeding based on information planning.

[0028] In this embodiment, a computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device. The computer-readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, the computer-readable storage medium can be a portable computer disk, a hard disk, a USB flash drive, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), spoofing random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory stick, floppy disk, optical disk, magnetic disk, mechanical encoding device, or any combination thereof.

[0029] 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.

[0030] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A method for predicting and analyzing gastrointestinal bleeding based on information planning, characterized in that, Includes the following steps: Obtain information about the patient's digestive tract and corresponding oral chewing habits, as well as information about the corresponding food. The patient's oral chewing habits and the corresponding food conditions are used to predict abnormalities after food chewing. The damage to the digestive tract caused by chewed food is analyzed by combining the condition of the food after chewing and the condition of the digestive tract lining. Gastrointestinal bleeding is predicted by combining the extent of gastrointestinal damage with the corresponding vascular distribution. By combining the predicted gastrointestinal bleeding and damage, a risk assessment is made regarding the suitability of food, and corresponding warnings are issued for the food.

2. The gastrointestinal bleeding prediction and analysis method based on information planning according to claim 1, characterized in that, The process of predicting abnormalities after food chewing includes the following specific steps: The model acquires historical data on the type, shape, hardness, and volume of corresponding foods, as well as historical data on the chewing time and force of corresponding foods by patients, oral enzyme activity and oral pH levels of patients, and the shape and hardness of chewed foods. It constructs a multilayer perceptron neural network model that takes the type, shape, hardness, and volume of corresponding foods, chewing time and force of corresponding foods, oral enzyme activity and oral pH levels of patients as inputs, and outputs the size and hardness of chewed foods. The data collected on the type, shape, hardness, and volume of the corresponding food, as well as the chewing time, chewing force, oral enzyme activity, and oral pH of the patient, are imported into the corresponding multilayer perceptron neural network model to output the size and hardness of the chewed food. Abnormal food chewing values ​​are obtained by weighting and summing the ratios of the size and hardness of chewed food to the corresponding safe food data.

3. The gastrointestinal bleeding prediction and analysis method based on information planning according to claim 2, characterized in that, The extent of damage to the digestive tract caused by chewed food includes the following specific details: The system acquires images of lesions in the digestive tract of the corresponding personnel, as well as safe puncture pressure and internal diameter of the digestive tract at various locations. The abnormal lesion value is obtained by the average deviation between the pixel values ​​of each point in the lesion image and the pixel values ​​of locations without lesions. The original damage value of the digestive tract is obtained by multiplying the abnormal lesion value by the lesion influence coefficient. Obtain the safe puncture pressure and digestive tract diameter at the corresponding location. Weighted summation of abnormal pressure and abnormal digestive tract diameter yields the digestive tract tolerance abnormality. Weighted summation of digestive tract tolerance abnormality with the original digestive tract damage value yields the digestive tract vulnerability value. The abnormal damage result of chewed food to the digestive tract is obtained by weighted summation of the digestive tract vulnerability value and the food chewing abnormal value. The overall damage result is obtained by summing the abnormal damage results of chewed food to the digestive tract at all locations.

4. The gastrointestinal bleeding prediction and analysis method based on information planning according to claim 2, characterized in that, The prediction of gastrointestinal bleeding includes the following specific details: The system obtains the distribution of blood vessels and blood flow in the blood vessels at various locations in the digestive tract. It also obtains abnormal results of food damage to the digestive tract after chewing at each location. The abnormal results of food damage to the digestive tract after chewing at each location are compared with the set abnormal damage threshold. The location where the abnormal results of food damage to the digestive tract after chewing are greater than or equal to the set abnormal damage threshold is set as the abnormal digestion segment, and the location where the abnormal results of food damage to the digestive tract after chewing are less than the set abnormal damage threshold is set as the safe digestion segment. The vascular depth hazard value is obtained by dividing the safe vascular depth by the vascular depth of the corresponding abnormal digestive segment. The vascular flow rate hazard value is obtained by dividing the vascular flow rate of the corresponding abnormal digestive segment by the safe vascular flow rate. The vascular depth hazard value and the vascular flow rate hazard value are weighted and summed to obtain the blood flow abnormality of the corresponding abnormal digestive segment. The abnormal blood flow and abnormal injury results of the corresponding abnormal digestive segment are multiplied to obtain the bleeding abnormality prediction of the corresponding abnormal digestive segment. The bleeding abnormality predictions of all corresponding abnormal digestive segments are summed to obtain the gastrointestinal bleeding abnormality.

5. The gastrointestinal bleeding prediction and analysis method based on information planning according to claim 4, characterized in that, The method of combining the predicted gastrointestinal bleeding and injury to assess the suitability of food and issue warnings for corresponding foods includes the following specific content: The system obtains the overall damage results and abnormal gastrointestinal bleeding caused by the consumption of various foods, compares them with the corresponding thresholds, selects foods whose overall damage results and abnormal gastrointestinal bleeding are both less than or equal to the corresponding thresholds as safe foods, and classifies other foods as dangerous foods for early warning.

6. The gastrointestinal bleeding prediction and analysis method based on information planning according to claim 2, characterized in that, The construction of the multilayer perceptron neural network model includes the following specific contents: Historical data was divided into a 70% training set and a 30% test / validation set. The 70% weight and bias training set was input into the neural network model for training to obtain an initial multilayer perceptron neural network model. The initial multilayer perceptron neural network model was tested using the 30% weight and bias test set. The output of the initial multilayer perceptron neural network model that satisfies the preset maximum accuracy of the size and hardness of the chewed food was used as the neural network model output.

7. The gastrointestinal bleeding prediction and analysis method based on information planning according to claim 1, characterized in that, The acquisition of the patient's digestive tract condition and corresponding oral chewing habits, as well as the corresponding food condition, includes the following specific contents: acquiring the internal condition of the patient's digestive tract through digestive endoscopy, including images of the patient's internal digestive tract and the distribution of blood vessels in the digestive tract, and constructing a three-dimensional image of the patient's digestive tract for reference using medical image construction software. Obtain information on the type, shape, hardness, and volume of the corresponding food. By obtaining information on the patient's oral chewing habits, the duration and intensity of chewing for different types of food, as well as the patient's oral enzyme activity and oral pH levels, can be obtained.

8. A gastrointestinal bleeding prediction and analysis system based on information planning, used to implement the gastrointestinal bleeding prediction and analysis method based on information planning as described in any one of claims 1-7, characterized in that, Specifically, it includes: The data acquisition module acquires information about the patient's digestive tract and corresponding oral chewing habits, as well as information about the corresponding food. The food chewing analysis module uses the patient's oral chewing habits and the corresponding food conditions to predict abnormalities after food chewing. The food damage prediction module analyzes the damage to the digestive tract after chewing by combining the condition of the food after chewing and the condition of the digestive tract wall. The bleeding prediction module combines the condition of gastrointestinal injury with the distribution of blood vessels in the corresponding location to predict gastrointestinal bleeding. The risk assessment module combines the predicted gastrointestinal bleeding and damage to determine the suitability of food and issue warnings for the corresponding foods.

9. An electronic device, comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; characterized in that the processor executes the information planning-based gastrointestinal bleeding prediction and analysis method as described in any one of claims 1-7 by calling the computer program stored in the memory.

Images