Nutrition whole-process management system for adult Crohn disease patient

Abstract

The invention relates to the technical field of medical data management, in particular to an adult Crohn disease patient nutrition whole-process management system, which comprises a risk screening module, which is used for identifying high-risk patients by adopting a standardized scale through integrating multiple data of patients; the nutrition evaluation module is used for outputting multi-dimensional comprehensive evaluation data based on a multi-dimensional evaluation mechanism in combination with the actual data of the high-risk patient; according to the invention, the scheme making module quantifies and balances three core objectives of nutritional supplement, drug side effects and treatment effects, and introduces drug-nutrition interaction constraints to generate a systematic collaborative optimization scheme, so that the pain point of the traditional separation treatment that they lose their other is solved, and the treatment collaborative effect is realized; through the monitoring and intervention module, the scheme is adjusted in real time through the index change rate and the dynamic intervention threshold value, and real-time dynamic adjustment can be carried out by adapting to factors such as patient individual differences, disease stages and intestinal function states.

Description

Technical Field

[0001] This invention relates to the field of medical data management technology, specifically to a comprehensive nutrition management system for adult Crohn's disease patients. Background Technology

[0002] Adult Crohn's disease patients are individuals diagnosed with Crohn's disease through a comprehensive diagnosis including clinical symptoms, imaging examinations, endoscopic examinations, and pathological biopsies. Crohn's disease is a chronic, relapsing inflammatory bowel disease of unknown etiology. Its core pathological feature is chronic granulomatous inflammation of the intestinal mucosa. The lesions can affect the entire digestive tract, with the terminal ileum and adjacent colon being the most common sites.

[0003] The patent with publication number CN113573417A describes in its specification that "this invention discloses an intelligent nutrition management system for kidney disease patients, including units for performing the following operations: generating patient nutrition management information by a kidney disease management server and sending it to a base station; in response to receiving the patient nutrition management information, the base station sends a first downlink control message to a mobile terminal on a first frequency band and a first symbol set; in response to receiving the first downlink control message sent by the base station, the mobile terminal begins to receive downlink data sent by the base station on the downlink resources indicated by the first downlink control message; in response to starting to send downlink data to the mobile terminal, the base station allocates a first PRACH resource set for sending random access preambles to a second mobile terminal; in response to allocating the first PRACH resources for sending random access preambles to the second mobile terminal, the base station sends a second downlink control message to the mobile terminal on a second frequency band and before a fourth time slot set."

[0004] While existing technologies have the advantages mentioned above, they also have disadvantages: Crohn's disease is a chronic inflammatory bowel disease characterized by recurrent episodes and persistent symptoms. Malnutrition is the most common systemic manifestation in adult Crohn's disease patients and can occur at any stage of the disease. In existing technologies, nutritional therapy and drug therapy are usually carried out separately, lacking systematic synergistic optimization. In addition, existing nutritional programs are mostly standardized formulas and cannot be dynamically adjusted in real time according to individual patient differences, disease stage, intestinal function, and other factors.

[0005] In conclusion, the development of a comprehensive nutrition management system for adult Crohn's disease patients remains a critical issue that urgently needs to be addressed in the field of medical data management technology. Summary of the Invention

[0006] The purpose of this invention is to address the problems in the existing technology, namely that Crohn's disease is a chronic inflammatory bowel disease characterized by recurrent attacks and persistent symptoms, malnutrition is the most common systemic manifestation in adult Crohn's disease patients and can occur at any stage of the disease, nutritional therapy and drug therapy are usually carried out separately in the existing technology, lacking systematic synergistic optimization, and existing nutritional programs are mostly standardized formulas that cannot be dynamically adjusted in real time according to individual differences of patients, disease stage, intestinal function status and other factors.

[0007] To achieve the above objectives, the present invention provides a comprehensive nutrition management system for adult Crohn's disease patients, comprising: The risk screening module integrates multiple patient data points and uses standardized scales to identify high-risk patients. The nutrition assessment module, based on a multi-dimensional assessment mechanism and combined with the actual data of the high-risk patients, outputs multi-dimensional comprehensive assessment data; The solution formulation module uses an artificial intelligence technology to build a decision support system and formulates a synergistic nutrition and drug treatment plan based on the multi-dimensional comprehensive evaluation data. The monitoring and intervention module tracks the feedback data after the implementation of the nutrition and drug synergy program in real time, updates the multi-dimensional comprehensive evaluation data, and then dynamically adjusts the nutrition and drug synergy program.

[0008] Beneficial effects Compared with known public technologies, the technical solution provided by this invention has the following beneficial effects: This invention quantifies and balances the three core objectives of nutritional supplementation, drug side effects, and therapeutic effects through a scheme formulation module. At the same time, it introduces drug-nutrient interaction constraints to generate a systematic synergistic optimization scheme, which can solve the pain point of traditional separate treatments that fail to address one aspect at the expense of another, and achieve synergistic therapeutic effects. Through the monitoring and intervention module, the scheme can be adjusted in real time by using the rate of change of indicators and dynamic intervention thresholds, which is conducive to adapting to factors such as individual patient differences, disease stage, and intestinal function status for real-time dynamic adjustment.

[0009] This invention integrates 12 patient-specific indicators through a risk screening module, and combines the weight fusion of entropy weight method and analytic hierarchy process to improve the accuracy of identifying high-risk groups. Through a nutritional assessment module, it comprehensively captures the patient's nutritional status, metabolic function, intestinal tolerance and drug interaction characteristics. Attached Figure Description

[0010] Figure 1 This is a system diagram of the adult Crohn's disease patient nutrition management system of the present invention. Detailed Implementation

[0011] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0012] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0013] The present invention will now be described in further detail with reference to the accompanying drawings: Example: like Figure 1 As shown, this invention provides a comprehensive nutrition management system for adult Crohn's disease patients, including: The risk screening module integrates multiple patient data points and uses standardized scales to identify high-risk patients. The nutrition assessment module, based on a multi-dimensional assessment mechanism and combined with the actual data of the high-risk patients, outputs multi-dimensional comprehensive assessment data; The solution formulation module uses an artificial intelligence technology to build a decision support system and formulates a synergistic nutrition and drug treatment plan based on the multi-dimensional comprehensive evaluation data. The monitoring and intervention module tracks the feedback data after the implementation of the nutrition and drug synergy program in real time, and updates the multi-dimensional comprehensive evaluation data to dynamically adjust the nutrition and drug synergy program. Furthermore, the operational procedures for the risk screening module include: The data include 12 indicators: age, gender, disease duration, Crohn's disease activity index, degree of intestinal inflammation, history of bowel resection, BMI, serum albumin, hemoglobin, diet adherence, history of malnutrition, and diabetic / nephropathy complications; the data are standardized to eliminate dimensional differences. Objective weights are calculated based on the entropy weight method. A judgment matrix is ​​constructed according to the 1-9 scale method. Gastroenterology experts judge the importance of the indicator data. Subjective weights are obtained after calculation and consistency test by the analytic hierarchy process. The objective weights and subjective weights are then merged to output the comprehensive weight. Specifically, for adult Crohn's disease patients aged 18-65, excluding those with severe liver or kidney failure, malignant tumors, or those who are pregnant or lactating, a closed-loop process of risk screening, nutritional assessment, treatment plan development, and monitoring and intervention was implemented to achieve synergistic management of nutrition and medication. Patient Zhang, a 35-year-old male, presented with "recurrent abdominal pain and diarrhea for 3 years, worsening with weight loss for 2 months" and was diagnosed with Crohn's disease of the ileocolonic type, in the active phase. Data on 12 indicators required by the risk screening module were collected through the hospital's HIS system, data interfaces of the laboratory department, and patient self-reported questionnaires, as shown in Table 1. Table 1

[0014] Z-score standardization combined with normalization is used to transform multiple data points into standardized values ​​in the [0,1] interval. The calculation formula is as follows: , In the formula, This is the mean of a clinical sample based on historical data from 500 adult Crohn's disease patients. The standard deviation is shown in Table 2, taking the key indicator as an example. Table 2

[0015] Based on the entropy weight method for calculating objective weights, the first step is to calculate the... Probability of item indicators : , Calculate the information entropy again : , Obtain objective weights : , Objective weighting results for some key indicators: , , , The analytic hierarchy process (AHP) was used to calculate subjective weights. This involved inviting five chief physicians from the gastroenterology department to perform pairwise comparisons of the importance of 12 indicators using a 1-9 scale (1 = equally important, 3 = slightly important, 5 = significantly important, 7 = strongly important, 9 = extremely important). A 12×12 judgment matrix was constructed, and the largest eigenvalue of the judgment matrix was calculated. Consistency indicators: Random Consistency Index Consistency ratio To meet the consistency requirement, subjective weights are then calculated. The subjective weights of some key indicators are obtained by solving the eigenvector method. The weighted average method is adopted, and objective weights are set accordingly. Subjective weighting The overall weight is: , Core indicator weighting results: .

[0016] Furthermore, the operational procedures for the risk screening module also include: The method uses standardized scales to identify high-risk patients and calculates their comprehensive nutritional risk score. :

[0017] In the formula, Indicates the first The comprehensive nutritional risk score of each patient Indicates the first The weight of each screening indicator Indicates the first The patient The standardized value of the indicator Represents the coefficient of the interaction term. Indicates the first Crohn's disease activity index of the patients Indicates the first Patients with a history of bowel resection were used to determine screening thresholds based on ROC curves. : Obtain historical data to construct an ROC curve and calculate the Youden index: , In the formula, Indicates the Yoden Index, This refers to the proportion of patients who are actually malnourished but are correctly identified as high-risk. This refers to the proportion of patients who are actually not malnourished but are correctly identified as low-risk. corresponding As a high-risk assessment threshold, At that time, among them, This indicates the high-risk threshold. If a patient is identified as high-risk, they will proceed to the nutrition assessment module; otherwise, they will be identified as low-risk and a basic nutrition guidance plan will be output. Specifically, the interaction term coefficients were determined based on training using 500 historical data examples. To highlight the synergistic risk of "CDAI + history of bowel resection," the sum of the weights and standardized values ​​of the 12 indicators is first calculated: Then calculate the interaction items: In the formula, , , , Finally, calculate the comprehensive score. Historical data from 800 adult Crohn's disease patients were obtained: 320 patients were diagnosed with malnutrition based on PG-SGA score and serum markers, and 480 patients were not malnourished. Calculations were made for each patient. Value, iterate through all possible values. Threshold, calculate the corresponding Values, some key threshold data are shown in Table 3: Table 3

[0018] ROC curve analysis was conducted at... When, corresponding threshold However, considering the clinical need to "prioritize reducing the rate of missed diagnoses," the final selection was... corresponding As a high-risk threshold, the sensitivity is 0.89 with a false negative rate of 11%; the specificity is 0.85 with a false negative rate of 15%, balancing accuracy and clinical applicability. If a patient is identified as high-risk, the nutritional assessment module is automatically triggered, and a multi-dimensional precision assessment process is initiated. If a patient is identified as low-risk, a basic nutritional guidance plan is provided, such as a daily energy intake of 1800-2000 kcal, dietary fiber of 25g, and monthly monitoring of weight and serum ALB.

[0019] Furthermore, the operational procedures for the nutrition assessment module include: For the high-risk patients, an assessment indicator system was constructed, consisting of 4 primary assessment dimensions and 16 secondary indicators. ,in, As a first-level dimension, The indicators are secondary indicators. Dimension 1 is nutritional status, including: BMI standardized, ALB standardized value, lymphocyte count, and PG-SGA assessment; Dimension 2 is metabolic function, including: basal metabolic rate, energy expenditure coefficient, glucose metabolism abnormality coefficient, and lipid metabolism abnormality coefficient; Dimension 3 is intestinal tolerance, including: number of food intolerances, frequency of diarrhea, erythrocyte sedimentation rate, C-reactive protein, and procalcitonin; Dimension 4 is drug-nutrient interaction, including: dosage of immunosuppressants, duration of hormone use, degree of vitamin D deficiency, degree of vitamin B12 deficiency, degree of folic acid deficiency, degree of iron deficiency, and degree of calcium deficiency.

[0020] Specifically, taking a 35-year-old male adult with Crohn's disease as an example, he was diagnosed with active Crohn's disease in the terminal ileum with a CDAI score of 280, had a history of one partial bowel resection, and was identified as a high-risk patient by the risk screening module due to a weight loss of 8 kg in 3 months and a serum albumin (ALB) level of 29.5 g / L. He then entered the nutritional assessment process, and obtained four secondary indicator data through anthropometric measurements, laboratory tests, and clinical assessments—his actual BMI was 17.8 kg / m². 2 After Z-score standardization, the result is ALB measured at 29.5 g / L, after standardization... Lymphocyte count 1.8 × 10⁻⁶ 9 / L, quantized as The patient was assessed as Grade C using the PG-SGA assessment scale, indicating severe malnutrition, quantified as follows: The basal metabolic rate was measured to be 1420 kcal / d using an indirect calorimetry method. Combined with the disease activity level, the energy expenditure coefficient was calculated to be 1.3. Fasting blood glucose was 6.8 mmol / L, slightly elevated, and the glucose metabolism abnormality coefficient was quantified as follows: Triglycerides were 2.1 mmol / L, slightly elevated, and the lipid metabolism abnormality coefficient was quantified as follows: By analyzing patients' food diaries, it was found that they were intolerant to four types of food: milk, wheat, soy, and nuts. Over the past week, the average number of times I had diarrhea was 5, which was quantified as... Intestinal permeability was detected using the lactulose / mannitol ratio method, and the result was 0.18, which is higher than the normal threshold of 0.03. Serum C-reactive protein 12 mg / L, tumor necrosis factor-α 28 pg / mL, and inflammatory factor levels quantified as follows: The patient is currently taking the immunosuppressant azathioprine 50 mg / day. The dose-response coefficient was calculated based on liver and kidney function. ; Previous use of glucocorticoids for 2 months, duration quantified as follows: Serum vitamin D level was 12 ng / mL, and the deficiency severity coefficient was... Serum zinc level was 7.2 μmol / L, indicating mild deficiency. The deficiency severity coefficient was calculated as follows: Ultimately, a complete assessment indicator system for this patient was formed. The assessment results can directly provide a basis for subsequent treatment plans. For example, if the patient has poor intestinal tolerance, the treatment plan module can select short-peptide enteral nutrition preparations and avoid formulas containing milk or wheat. In combination with vitamin D deficiency, adding 800 IU of vitamin D supplementation daily to the treatment plan has clear clinical application value and helps to avoid the one-sidedness of traditional single-indicator assessment. The 4 dimensions and 16 indicators cover the entire chain of intake-absorption-metabolism-drug effects. Through the assessment, patients can not only clarify the cause of weight loss, but also find vitamin D deficiency caused by hormones, which is conducive to providing direction for precise intervention.

[0021] Furthermore, the operational procedures for the nutrition assessment module also include: The evaluation was quantified based on four primary assessment dimensions. A fuzzy judgment matrix was constructed using fuzzy evaluations of the importance of each dimension from gastroenterologists and nutritionists, along with triangular fuzzy numbers. ,in, Represents the fuzzy judgment matrix. Represents the first in the matrix The assessment dimension and the first fuzzy importance relationships among the evaluation dimensions, and calculate the weights of each dimension. ,expression: , In the formula, Indicates the first The weights of each evaluation dimension are determined based on the weights of each dimension. Then calculate the multi-dimensional comprehensive evaluation value : , In the formula, This represents data from a multi-dimensional comprehensive evaluation. Indicates the first Scores for each primary evaluation dimension Indicates the first The first primary evaluation dimension The quantitative value of each secondary indicator, , The closer to 0, the worse the nutritional status, and the stronger the intervention required; Specifically, as previously mentioned, 16 secondary indicator quantification values ​​of a 32-year-old male at high risk of Crohn's disease have been obtained. For example, first, the quantitative values ​​of the secondary indicators under the four primary dimensions of the patient are organized into basic data, such as the nutritional status dimension. Metabolic function dimension Then, three chief physicians of gastroenterology and three clinical nutrition experts were invited to conduct a fuzzy evaluation of the importance of four dimensions based on the characteristics of poor intestinal absorption and drug-induced nutrient metabolism in adult Crohn's disease patients. For example, the experts unanimously agreed that nutritional status had the greatest impact on the assessment results, while drug-nutrient interactions needed to be considered in conjunction with the treatment plan but had a slightly lower weight. A 4×4 fuzzy judgment matrix was constructed using triangular fuzzy numbers, and the results were then calculated according to the formula... The weights of each dimension are obtained by summing the fuzzy values ​​of each dimension in the matrix and dividing by the sum of all fuzzy values. The calculated weights of the four dimensions in the patient case are: =0.35、 =0.25、 =0.25、 =0.15, with a total weight of 1, a multi-dimensional comprehensive score, first according to the formula Calculate the average score for each primary dimension, such as the patient's nutritional status dimension. Metabolic function dimension Intestinal tolerance Drug-nutrient interactions Then follow the formula Calculate the comprehensive assessment value, and ultimately the patient's A score of 0.32, close to 0, indicates poor nutritional status requiring strong intervention. By assigning weights to highlight core issues, such as patients with good metabolic function but high nutritional status weights, the overall score can still reflect their malnutrition, avoiding misjudgments caused by traditional egalitarian scoring and helping medical staff accurately identify intervention priorities. By integrating the opinions of gastroenterologists and nutritionists through mathematical tools, a unified assessment conclusion can be formed without repeated communication, improving the efficiency of interdisciplinary collaboration.

[0022] Furthermore, the operational process of the solution development module includes: The decision support system includes constructing a multi-objective optimization model that maximizes nutritional supplementation, minimizes drug side effects, and optimizes treatment efficacy.

[0023] In the formula, This represents the objective function for multi-objective optimization. Based on The calculated nutritional deficiency risk coefficient This is a risk factor for drug side effects calculated based on drug dosage and the patient's liver and kidney function. The efficacy of Crohn's disease treatment is calculated based on the improvement rate of CDAI. To determine the target weights based on clinical data training, Crohn's disease-specific constraints were introduced, including: nutritional intake constraints, nutrient ratio constraints, and drug-nutrient interaction constraints. Specifically, through the objective function This approach quantitatively integrates nutritional, side effect, and therapeutic needs, avoiding the pitfalls of traditional treatment plans that prioritize efficacy over nutrition and supplementation over safety. For example, patient treatment plans reduce energy intake... Furthermore, by controlling the drug dosage, it can be avoided Too high, while ensuring This approach enhances the safety, effectiveness, and nutritional balance of the treatment. Crohn's disease-specific constraints directly mitigate clinical risks, such as drug-nutrient interaction constraints that mandate folic acid supplementation to prevent folic acid deficiency anemia caused by azathioprine. Nutrient ratio constraints control fat intake, reducing intestinal burden and diarrhea, significantly lowering the incidence of adverse events during treatment. The multi-objective optimization model parameters and constraints are tailored to individual patient conditions, such as appropriately increasing drug dosages based on normal liver and kidney function, or selecting lactose-free enteral nutrition formulas based on milk intolerance. This avoids a one-size-fits-all standardized approach and better aligns with the diverse clinical characteristics of adult Crohn's disease patients. Quantifying objectives and constraints through mathematical formulas eliminates the need for repeated manual adjustments by medical staff, improving clinical decision-making efficiency, especially suitable for medical institutions with high outpatient volumes and limited medical resources.

[0024] Furthermore, the operational process of the solution development module also includes: The decision support system includes an AI decision-making model built using an attention mechanism and reinforcement learning. The input features of the AI ​​decision-making model are multi-dimensional comprehensive evaluation data. The patient's age, weight, liver and kidney function, medication history, dietary preferences, and degree of intestinal inflammation were assessed; attention weights were calculated using a feature extraction module based on attention mechanisms. : , In the formula, Indicates the first Attention weights for each feature Indicates the first query vectors With key vector Match score, Indicates the first A query vector, Indicates the first A key vector, This indicates that the query vector is multiplied by the transpose of the key vector. The vector dimension is represented by a reinforcement learning optimization scheme, which defines the patient's current state. Nutritional plan + medication adjustment plan Reward function :

[0025] In the formula, This represents the reward function for reinforcement learning. , , This represents the reward weighting coefficient. This indicates the improvement value of the Crohn's disease activity index. This indicates the increase in the comprehensive evaluation value across multiple dimensions. This represents the probability of side effects calculated based on drug dosage and patient nutritional status parameters. The AI ​​decision-making model comprehensively evaluates data from multiple dimensions. The output of nutritional and drug synergistic solutions includes nutritional and drug solutions; Specifically, an AI decision-making model is built using attention mechanisms and reinforcement learning to generate personalized nutrition and medication synergy plans for high-risk Crohn's disease patients, collecting seven core input features from patients: multi-dimensional comprehensive assessment data. The model input feature set is formed by considering the following data: age 32, weight 52kg, liver and kidney function, medication history, dietary preferences, and degree of intestinal inflammation. The attention weight of each input feature is calculated using the feature extraction module based on the attention mechanism. First, define the query vector. and key vector This is used to reflect the correlation between characteristics and the synergistic effect of nutrition and drugs, and then calculated using the formula... Calculate the matching score, and finally use the softmax function. ,in, Weights are assigned to the total number of features. The calculated weights for key features in patient cases are: =0.32、 =0.28、 =0.21, while age ,weight With relatively low weighting, priority is given to nutritional status, inflammation level, and drug-related factors; define the current state. - Nutrition plan + medication adjustment plan -Reward function Current state The patient's current input characteristics, such as CRP = 12 mg / L, nutritional plan + medication adjustment plan Nutritional plans to be optimized, such as energy intake, type of nutritional supplements, and medication adjustments, such as azathioprine dosage; reward function. , Weighting coefficients determined for clinical training; initial protocol generated: nutrition includes short-peptide enteral nutrition formula, 1800 kcal daily energy, lactose-free + medications including azathioprine 50 mg / day, vitamin D 800 IU / day; protocol execution effect simulated again: , , Substituting into the reward function, we get ; Adjust the nutrition plan and medication plan through iterative adjustments The reward value is recalculated, and finally... At its maximum, the final nutritional and medication regimen is output: short-peptide lactose-free enteral nutrition + azathioprine 75mg / d + vitamin D 1000IU / d + folic acid 5mg / d; by focusing on core features through an attention mechanism, ineffective interventions can be avoided, making the regimen more aligned with the patient's core needs. Through reinforcement learning, the regimen is iterated in real time using a reward function, which can be adapted to changes in the patient's condition and avoid a one-size-fits-all fixed regimen.

[0026] Furthermore, the operational procedures for the monitoring and intervention module include: Real-time feedback data is collected, including: diet adherence, gastrointestinal symptoms, performance status score, weekly weight, bi-weekly serum markers, monthly CDAI score, and degree of intestinal inflammation; the status of high-risk patients is updated based on Kalman filtering, and the rate of change of the indicator data is calculated. : , In the formula, Indicates the first The rate of change of the indicators at each monitoring time Indicates the first The indicator values ​​at the time of the next monitoring It is the abbreviation for serum albumin. It is an abbreviation for Crohn's Disease Activity Index. It is a multi-dimensional comprehensive evaluation data. Indicates the first The indicator values ​​at the time of the next monitoring; Specifically, through a closed-loop process of multi-frequency real-time data acquisition → Kalman filter state optimization → quantification of key indicator change rates, the effectiveness of patient protocol execution is dynamically tracked. Dietary execution is collected through a patient's mobile app, such as: actual intake of 1980kcal out of the 2212kcal target, execution rate of 89%; gastrointestinal symptoms: diarrhea frequency decreased from 4 times / day to 2 times / day; physical condition score: ECOG score decreased from 2 points to 1 point; weight is automatically uploaded through a smart scale: 58kg in week 1 → 60kg in week 2. Laboratory data every two weeks: Hospital lab interface synchronized serum indicators: ALB from 31.5g / L → 33.2g / L, Vitamin D from 15ng / mL → 18ng / mL; Doctor's offline assessment of CDAI: 275 points → 230 points, Intestinal inflammation level: CRP from 0.85 → 0.62. Taking weight as an example, Kalman filtering was used to update patient status, resolving data fluctuations caused by weighing time and clothing differences: Based on the first week's weight of 58kg and the expected weight gain of 0.5kg / week, the predicted weight for the second week was calculated. covariance matrix The actual weight collected in week 2 was 60kg, and the final optimized weight calculated using Kalman gain was 59.4kg. Four core indicators were selected: weight, ALB, CDAI, and Z, and substituted into the formula. calculate: The effectiveness of the treatment plan can be intuitively judged by quantifying the rate of change. Differentiated data collection cycles are designed for adult Crohn's disease patients, ranging from daily to weekly to bi-weekly to monthly, to ensure high-frequency tracking of subjective symptoms while avoiding excessive testing of laboratory indicators and reducing the testing burden on patients.

[0027] Furthermore, the operational procedures for the monitoring and intervention module also include: Set the intervention trigger threshold, expression:

[0028] In the formula, Indicates the intervention trigger threshold. The average rate of change for the population. Standard deviation For the present Multi-dimensional comprehensive evaluation data at any given time; Furthermore, the operational procedures for the monitoring and intervention module also include: exist When the indicator deteriorates, it triggers emergency intervention, and the system re-enters the nutrition assessment module, based on the updated data. Adjust the aforementioned synergistic nutrition and medication regimen; in When the indicator is stable, the current plan will be maintained, with minor adjustments to the nutrient ratios; When the indicator improves, the intensity of intervention should be gradually reduced; Specifically, based on monitoring data from 1200 adult Crohn's disease patients, weight indicators... ALB index CDAI indicator Standard deviation The corresponding indicators are as follows: (weight), (ALB) (CDAI); Current multi-dimensional comprehensive assessment value Updated 2 weeks after implementation of the plan. This is an improvement from the initial 0.409. Taking weight as an example, the threshold is calculated as follows: Based on the rate of change of core indicators within 2 weeks, the following calculation was performed: weight =2.41%, ALB =5.40%, CDAI =-16.36%, compared to threshold classification processing: weight index: And it is a positive change, belonging to "- < < "This indicates that the indicators are stable, maintaining a daily energy intake of 2212 kcal, with a slight adjustment to reduce the protein ratio from 25% to 23%; ALB indicator:" The ALB threshold is calculated as follows: Approaching the threshold indicates improvement, thus allowing for a gradual reduction in the hydrolyzed protein supplement dosage; CDAI indicator: The CDAI threshold is calculated as follows: The negative threshold was -0.23%, indicating a significant improvement in the indicator. As planned, the infliximab infusion interval was restored from 6 weeks to 8 weeks to reduce the intervention intensity. ≤- If this occurs, emergency intervention will be triggered: re-enter the nutrition assessment module and update... The nutritional and drug synergy program can be adjusted to overcome the technical limitations of traditional fixed thresholds.

[0029] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A comprehensive nutrition management system for adult Crohn's disease patients, characterized in that: include: The risk screening module integrates multiple patient data points and uses standardized scales to identify high-risk patients. The nutrition assessment module, based on a multi-dimensional assessment mechanism and combined with the actual data of the high-risk patients, outputs multi-dimensional comprehensive assessment data; The solution formulation module uses an artificial intelligence technology to build a decision support system and formulates a synergistic nutrition and drug treatment plan based on the multi-dimensional comprehensive evaluation data. The monitoring and intervention module tracks the feedback data after the implementation of the nutrition and drug synergy program in real time, updates the multi-dimensional comprehensive evaluation data, and then dynamically adjusts the nutrition and drug synergy program.

2. The nutritional management system for adult Crohn's disease patients according to claim 1, characterized in that, The operation process of the risk screening module includes: The data include 12 indicators: age, sex, disease duration, Crohn's disease activity index, degree of intestinal inflammation, history of bowel resection, BMI, serum albumin, hemoglobin, diet adherence, history of malnutrition, and diabetic / nephropathy complications; the data are standardized to eliminate dimensional differences. Objective weights are calculated using the entropy weight method. A judgment matrix is ​​constructed using the 1-9 scale method. Gastroenterology experts judge the importance of the indicator data. Subjective weights are obtained by combining the analytic hierarchy process (AHP) calculation and consistency test. Finally, the objective weights and subjective weights are merged to output the comprehensive weight.

3. The adult Crohn's disease patient nutrition management system according to claim 2, characterized in that, The operation process of the risk screening module also includes: The method uses standardized scales to identify high-risk patients and calculates their comprehensive nutritional risk score. : In the formula, Indicates the first The comprehensive nutritional risk score of each patient Indicates the first The weight of each screening indicator Indicates the first The patient The standardized value of the indicator Represents the coefficient of the interaction term. Indicates the first Crohn's disease activity index of the patients Indicates the first Patients with a history of bowel resection were used to determine screening thresholds based on ROC curves. : Obtain historical data to construct an ROC curve and calculate the Youden index: ， In the formula, Indicates the Yoden Index, This refers to the proportion of patients who are actually malnourished but are correctly identified as high-risk. This refers to the proportion of patients who are actually not malnourished but are correctly identified as low-risk. corresponding As a high-risk assessment threshold, At that time, among them, This indicates the high-risk threshold. If a patient is identified as high-risk, they will proceed to the nutrition assessment module; otherwise, they will be identified as low-risk and a basic nutrition guidance plan will be output.

4. The adult Crohn's disease patient nutrition management system according to claim 3, characterized in that, The operation process of the nutrition assessment module includes: For the high-risk patients, an assessment indicator system was constructed, consisting of 4 primary assessment dimensions and 16 secondary indicators. ,in, As a first-level dimension, The indicators are secondary indicators. Dimension 1 is nutritional status, including: BMI standardized, ALB standardized value, lymphocyte count, and PG-SGA assessment; Dimension 2 is metabolic function, including: basal metabolic rate, energy expenditure coefficient, glucose metabolism abnormality coefficient, and lipid metabolism abnormality coefficient; Dimension 3 is intestinal tolerance, including: number of food intolerances, frequency of diarrhea, erythrocyte sedimentation rate, C-reactive protein, and procalcitonin; Dimension 4 is drug-nutrient interaction, including: dosage of immunosuppressants, duration of hormone use, degree of vitamin D deficiency, degree of vitamin B12 deficiency, degree of folic acid deficiency, degree of iron deficiency, and degree of calcium deficiency.

5. The nutritional management system for adult Crohn's disease patients according to claim 4, characterized in that, The operational procedures for the nutrition assessment module also include: The evaluation was quantified based on four primary assessment dimensions. A fuzzy judgment matrix was constructed using fuzzy evaluations of the importance of each dimension from gastroenterologists and nutritionists, along with triangular fuzzy numbers. ,in, Represents the fuzzy judgment matrix. Represents the first in the matrix The assessment dimension and the first fuzzy importance relationships among the evaluation dimensions, and calculate the weights of each dimension. ,expression: ， In the formula, Indicates the first The weights of each evaluation dimension are determined based on the weights of each dimension. Then calculate the multi-dimensional comprehensive evaluation value : , In the formula, This indicates data from a multi-dimensional comprehensive evaluation. Indicates the first Scores for each primary evaluation dimension Indicates the first The first primary evaluation dimension The quantitative value of each secondary indicator, , The closer to 0, the worse the nutritional status, and the stronger the need for intervention.

6. The nutritional management system for adult Crohn's disease patients according to claim 5, characterized in that, The operation process of the solution formulation module includes: The decision support system includes constructing a multi-objective optimization model that maximizes nutritional supplementation, minimizes drug side effects, and optimizes treatment efficacy. In the formula, This represents the objective function for multi-objective optimization. Based on The calculated nutritional deficiency risk coefficient This is a risk factor for drug side effects calculated based on drug dosage and the patient's liver and kidney function. The efficacy of Crohn's disease treatment is calculated based on the improvement rate of CDAI. To determine the target weights based on clinical data training, Crohn's disease-specific constraints were introduced, including: nutrient intake constraints, nutrient ratio constraints, and drug-nutrient interaction constraints.

7. The adult Crohn's disease patient nutrition management system according to claim 6, characterized in that, The operation process of the solution formulation module also includes: The decision support system includes an AI decision-making model built using an attention mechanism and reinforcement learning. The input features of the AI ​​decision-making model are multi-dimensional comprehensive evaluation data. The patient's age, weight, liver and kidney function, medication history, dietary preferences, and degree of intestinal inflammation were assessed; attention weights were calculated using a feature extraction module based on attention mechanisms. : , In the formula, Indicates the first Attention weights for each feature Indicates the first query vectors With key vector Match score, Indicates the first A query vector, Indicates the first A key vector, This indicates that the query vector is multiplied by the transpose of the key vector. The vector dimension is represented by a reinforcement learning optimization scheme, which defines the patient's current state. Nutritional plan + medication adjustment plan Reward function : In the formula, This represents the reward function for reinforcement learning. , , This represents the reward weighting coefficient. This indicates the improvement value of the Crohn's disease activity index. This indicates the increase in the comprehensive evaluation value across multiple dimensions. This represents the probability of side effects calculated based on drug dosage and patient nutritional status parameters. The AI ​​decision-making model comprehensively evaluates data from multiple dimensions. The output of nutritional and drug synergistic solutions includes nutritional and drug solutions.

8. The nutritional management system for adult Crohn's disease patients according to claim 7, characterized in that, The operation process of the monitoring and intervention module includes: Real-time feedback data is collected, including: diet adherence, gastrointestinal symptoms, performance status score, weekly weight, bi-weekly serum markers, monthly CDAI score, and degree of intestinal inflammation; the status of high-risk patients is updated based on Kalman filtering, and the rate of change of the indicator data is calculated. : ， In the formula, Indicates the first The rate of change of the indicators at each monitoring time Indicates the first The indicator values ​​at the time of the next monitoring It is the abbreviation for serum albumin. It is an abbreviation for Crohn's Disease Activity Index. It is a multi-dimensional comprehensive evaluation data. Indicates the first The indicator values ​​at the time of the next monitoring.

9. The nutritional management system for adult Crohn's disease patients according to claim 8, characterized in that, The operation process of the monitoring and intervention module also includes: Set the intervention trigger threshold, expression: In the formula, Indicates the intervention trigger threshold. The average rate of change for the population. Standard deviation For the present Multi-dimensional comprehensive evaluation data at any given time.

10. The nutritional management system for adult Crohn's disease patients according to claim 9, characterized in that, The operation process of the monitoring and intervention module also includes: exist When the indicator deteriorates, it triggers emergency intervention, and the system re-enters the nutrition assessment module, based on the updated data. Adjust the aforementioned synergistic nutrition and medication regimen; in When the indicator is stable, the current plan will be maintained, with minor adjustments to the nutrient ratios; When the indicator improves, the intensity of intervention should be gradually reduced.

Images