Nutrient intake difference analysis and quantitative matching system under complex chronic co-disease scene

Abstract

The invention belongs to the technical field of medical nutrition management, and particularly relates to a nutrition intake difference value analysis and quantitative matching system under a complex chronic common disease scene, which comprises a common disease nutrition demand knowledge base module, a common disease nutrition demand knowledge base module, a common disease nutrition demand knowledge base module, a common disease nutrition demand knowledge base module, a common disease nutrition demand knowledge base module and a common disease nutrition demand knowledge base module, setting a nutritional requirement priority; and the medicine-nutrition interaction adaptation module is used for generating a final nutrition constraint which overlaps the medicine influence by integrating a clinical chronic disease medicine database, automatically synchronizing a medicine use record of a patient, identifying a taboo nutrient corresponding to a medicine and recommending a collaborative nutrient, and performing risk early warning and replacement recommendation on a conflict scheme. The co-disease classification system is constructed according to the organ system and the disease severity, and the multi-dimensional nutrition constraint rule is matched for each type of co-diseases, so that the multi-dimensional nutrition constraint accurate adaptation in a complex co-disease scene is realized, and the limitation of nutrition management of a single disease is broken through.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical nutrition management, in particular to a nutrition intake difference analysis and quantitative matching system in a complex chronic comorbidity scenario. BACKGROUND

[0002] With the aggravation of population aging, the incidence of chronic diseases is increasing year by year, and the proportion of complex chronic comorbidity (i.e. the same patient suffering from 3 or more chronic diseases) has reached more than 45% of the chronic disease population. Nutrition support as the core link of chronic disease management directly affects the incidence of complications, hospitalization period and quality of life - for example, protein intake control of chronic kidney disease patients with diabetes can reduce the risk of 30% kidney damage, and sodium intake management of hypertension patients with heart failure can reduce the probability of 50% acute exacerbation.

[0003] However, the existing nutrition management system cannot adapt to the particularity of complex chronic comorbidity, and there are the following key problems:

[0004] I. Single disease oriented, unable to meet multi-dimensional nutrition constraints

[0005] Existing systems mostly design nutrition plans for single diseases (such as only adapting to the carbohydrate restriction of diabetes and only adapting to the protein restriction of kidney disease), without considering the nutrition constraint conflicts between comorbidities. For example: a 62-year-old female patient, diagnosed with type 2 diabetes (needs to control carbohydrate ≤180g / day), chronic kidney disease stage 3 (needs to control protein ≤0.8g / kg body weight, i.e. 56g / day) and hypertension (needs to control sodium ≤5g / day). The existing diabetes nutrition system only recommends a lunch plan of "mixed grain rice 150g + beef 100g" (containing carbohydrate 110g, protein 20g) according to blood glucose data, but does not superimpose the protein total amount constraint of kidney disease - if the patient executes this plan, the total protein intake is 72g (exceeding the target by 28.6%), leading to the monthly blood creatinine rising from 140μmol / L to 168μmol / L, aggravating kidney damage.

[0006] II. Single difference analysis dimension, lack of dynamic adaptability

[0007] The existing system only calculates the difference in nutrient intake (such as "actual sodium intake-target sodium intake") through "monthly / quarterly total statistics", without adjusting the difference calculation logic in real time combined with the physiological state of the patient (such as acute exacerbation, liver and kidney function fluctuation). For example: a 58-year-old male patient with chronic obstructive pulmonary disease (stable period needs to control fat ≤60g / day) combined with chronic liver disease (needs to control protein ≤1.0g / kg body weight). The existing system calculates fat intake once a month, showing a difference of 5g / day (actual 65g / day), without capturing the metabolic changes of the patient during the acute exacerbation of chronic obstructive pulmonary disease (oxygen saturation 89%) - at this time, the patient's fat metabolism capacity decreases, and the target fat intake needs to be reduced to 45g / day, resulting in an enlarged fat intake difference of 20g / day during the exacerbation period, which induces respiratory function deterioration.

[0008] III. Quantitative matching without priority mechanism, conflict cannot be coordinated

[0009] Patients with comorbidities often have conflicting nutritional needs (such as the need for Omega-3 in coronary heart disease and the need for total fat restriction in liver disease), and the existing system only mechanically matches a single nutrient without setting priority, resulting in an ineffective plan. For example: a 70-year-old male patient with coronary heart disease (Omega-3 ≥2g / day) combined with cirrhosis (total fat ≤50g / day). The existing system directly recommends "deep sea fish oil supplement 2g" (containing fat 1.8g), but does not simultaneously reduce dietary fat intake - on that day, the patient consumes braised pork 100g (containing fat 35g) + fish oil 2g, with a total fat of 51.8g, exceeding the liver disease tolerance threshold, causing transaminase to rise from 80U / L to 120U / L.

[0010] Based on the above, therefore, the invention provides a system for analyzing and quantitatively matching the difference in nutrient intake in complex chronic comorbidity scenarios. SUMMARY

[0011] To solve the above technical problems, the present application provides the following technical solutions:

[0012] The system for analyzing and quantitatively matching the difference in nutrient intake in complex chronic comorbidity scenarios comprises:

[0013] A comorbid nutritional needs knowledge base module for classifying common comorbidity combinations by organ system and disease severity, and matching basic constraints for each type of comorbidity, and setting priority for nutritional needs;

[0014] A drug-nutrient interaction adaptation module for integrating a clinical chronic disease drug database, automatically synchronizing patient medication records, identifying contraindicated nutrients and recommended synergistic nutrients corresponding to the drugs, generating final nutrient constraints superimposed with drug effects, and providing risk warnings and alternative recommendations for conflicting plans;

[0015] A dynamic intake difference analysis module is configured to calculate a nutrient intake difference in real time and cumulatively by collecting multi-dimensional real-time data, combining final nutritional constraints, and attributing the difference to reasons by an AI algorithm, and output a risk level;

[0016] A quantitative matching optimization module is configured to prioritize high-risk constraints by establishing a nutritional demand priority coordination mechanism, generate a basic food material scheme in combination with patient dietary preferences and cooking habits, and support rapid adjustment of the scheme according to real-time physiological data;

[0017] A feedback iteration module is configured to collect operation logs of each module, combine changes in patient physiological indicators, and dynamically iterate to continuously improve the adaptation accuracy of the system to complex comorbidity scenarios.

[0018] As a preferred scheme of the nutritional intake difference analysis and quantitative matching system in the complex chronic comorbidity scenario, the comorbidity nutritional demand knowledge base module comprises:

[0019] A comorbidity combination classification unit is configured to classify comorbidities into multiple categories according to organ systems and disease severity to cover common clinical combinations;

[0020] A nutritional constraint rule unit is configured to match multi-dimensional constraints for each comorbidity category and label conflict relationships;

[0021] A priority assignment unit is configured to assign priorities according to disease acute / chronic properties and nutritional sensitivity.

[0022] As a preferred scheme of the nutritional intake difference analysis and quantitative matching system in the complex chronic comorbidity scenario, the drug-nutrient interaction adaptation module comprises:

[0023] A drug database unit is configured to integrate commonly used chronic disease drugs in clinical practice, label contraindicated nutrients and recommended synergistic nutrients for each drug, and support automatic synchronization of patient medication records from hospital HIS systems without manual input;

[0024] An interaction rule unit is configured to establish a mapping relationship between drug combinations and nutritional constraints;

[0025] A risk warning unit is configured to automatically mark a risk level and push an alternative scheme if there is a drug-nutrient conflict in the initially generated nutritional scheme.

[0026] As a preferred scheme of the nutritional intake difference analysis and quantitative matching system in the complex chronic comorbidity scenario, the dynamic intake difference analysis module comprises:

[0027] A multi-source data collection unit is configured to collect real-time dietary data, physiological data, and state data;

[0028] a difference calculation unit configured to calculate the nutrient difference in two dimensions of real-time difference and cumulative difference;

[0029] a difference attribution unit configured to analyze the difference causes by an AI algorithm and output attribution results.

[0030] As a preferred scheme of the nutrient intake difference analysis and quantitative matching system in the complex chronic comorbidity scenario, wherein the calculation formula of the nutrient difference is:

[0031] a certain nutrient difference = (target intake amount under the current physiological state) - (actual intake amount).

[0032] As a preferred scheme of the nutrient intake difference analysis and quantitative matching system in the complex chronic comorbidity scenario, wherein the quantitative matching optimization module comprises:

[0033] a conflict coordination unit configured to dynamically adjust the conflicting nutrients according to the priority of the knowledge base;

[0034] a personalized scheme generation unit configured to generate a visual scheme of a list of ingredients, a serving size, and cooking suggestions in combination with the patient's dietary preferences and cooking methods;

[0035] a special dietary disorder adaptation module configured to identify the type of dietary disorder, establish a disorder-ingredient form mapping rule, convert the ingredients in the basic scheme to an adapted form, and provide visual preparation guidance while ensuring that the nutrient content remains unchanged;

[0036] a real-time adjustment unit configured to automatically update the target intake amount and the matching scheme within 10 minutes when the physiological data trigger threshold is reached.

[0037] As a preferred scheme of the nutrient intake difference analysis and quantitative matching system in the complex chronic comorbidity scenario, wherein the special dietary disorder adaptation module comprises:

[0038] a disorder type identification unit configured to first divide the dietary disorder into difficulty swallowing, chewing disorder, food allergy, and digestive absorption disorder through patient self-evaluation, clinical diagnosis, and intelligent device detection, and then match the ingredient form standard and the list of contraindicated ingredients for each type of disorder;

[0039] a scheme form conversion unit configured to automatically convert the ingredients of the basic scheme to an adapted form;

[0040] a preparation guidance unit configured to first generate a visual operation guide suitable for a home cooking scenario, and then link an intelligent cooking device to support one-key generation of adapted ingredient forms.

[0041] As a preferred scheme of the nutritional intake difference analysis and quantitative matching system in the complex chronic comorbidity scenario of the application, wherein the feedback iteration module comprises:

[0042] The acquisition unit is used for acquiring the patient plan execution rate, physiological index change, and optimizing the knowledge base rule through machine learning;

[0043] The generation unit is used for generating a nutritional management effect report every month, comparing the difference change trend and the incidence of complications, and providing a basis for clinical adjustment.

[0044] Compared with the prior art:

[0045] 1. By constructing a comorbidity classification system according to organ system + disease severity, and matching multi-dimensional nutritional constraint rules for each type of comorbidity, the multi-dimensional nutritional constraint precise adaptation in the complex comorbidity scenario is realized, and the limitation of single disease nutritional management is broken through.

[0046] 2. By collecting real-time data of diet, physiology and state, adopting real-time + cumulative two-dimensional difference calculation logic, and dynamically updating the target nutrient intake according to the physiological state of the patient, the difference analysis is synchronized with the physiological change of the patient, and the static statistical limitation is avoided.

[0047] 3. By constructing a nutritional demand priority mechanism based on the acute / chronic properties of the disease and the nutritional sensitivity, and combining with the conflict dynamic coordination rule, the ordered coordination of nutritional conflicts between comorbidities is realized, and the personalized nutritional plan conforming to the actual needs of the patient is generated. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 The figure is a schematic diagram of the overall framework of the application;

[0049] Figure 2 The figure is a schematic diagram of the comorbidity nutritional demand knowledge base module framework of the application;

[0050] Figure 3 The figure is a schematic diagram of the drug-nutrient interaction adaptation module framework of the application;

[0051] Figure 4 The figure is a schematic diagram of the dynamic intake difference analysis module framework of the application;

[0052] Figure 5 The figure is a schematic diagram of the quantitative matching optimization module framework of the application;

[0053] Figure 6 The figure is a schematic diagram of the feedback iteration module framework of the application;

[0054] Figure 7 The figure is a schematic diagram of the special diet disorder adaptation module framework of the application. DETAILED DESCRIPTION

[0055] In order to make the purposes, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the drawings.

[0056] The present application provides a nutrition intake difference analysis and quantitative matching system in a complex chronic comorbidity scenario, please refer to Figures 1-7 , comprising:

[0057] A comorbidity nutrition demand knowledge base module is used to classify common clinical comorbidity combinations by organ system and disease severity, and match basic constraints for each type of comorbidity, and set the priority of nutrition demand;

[0058] A drug-nutrition interaction adaptation module is used to integrate a clinical chronic disease drug database, automatically synchronize patient medication records, identify the corresponding contraindicated nutrients and recommended synergistic nutrients of the drugs, generate the final nutrition constraints superimposed with the effects of the drugs, and perform risk warning and alternative recommendation on the conflict scheme;

[0059] A dynamic intake difference analysis module is used to collect multi-dimensional real-time data, combine the final nutrition constraints, calculate the nutrient intake difference in real time and cumulative dimensions, and attribute the difference to the cause through an AI algorithm, while outputting the risk level;

[0060] A quantitative matching optimization module is used to establish a nutrition demand priority coordination mechanism, prioritize high-risk constraints, and generate a basic food material scheme in combination with patient dietary preferences and cooking habits, while supporting rapid adjustment of the scheme according to real-time physiological data;

[0061] A feedback iteration module is used to collect operation logs of each module, combine changes in patient physiological indicators, and perform dynamic iteration to continuously improve the adaptation accuracy of the system to complex comorbidity scenarios.

[0062] The comorbidity nutrition demand knowledge base module comprises:

[0063] A comorbidity combination classification unit is used to classify comorbidities into multiple categories (20+ major categories, 100+ subcategories) according to organ systems (such as kidney-metabolism-circulatory system) and disease severity (such as kidney disease stages 1-5) to cover common clinical combinations (such as "diabetes + kidney disease + hypertension" and "chronic obstructive pulmonary disease + heart failure + diabetes");

[0064] A nutrition constraint rule unit is used to match multi-dimensional constraints (such as nutrient range, contraindicated ingredients) for each type of comorbidity, and label conflict relationships (such as "kidney disease protein restriction" and "tumor protein demand" conflict);

[0065] A priority assignment unit is configured to assign priority according to disease acute / chronic property and nutrition sensitivity (for example, acute heart failure sodium restriction priority = 9, chronic hypertension sodium restriction priority = 6; kidney disease protein restriction priority = 8, diabetes carbohydrate restriction priority = 5).

[0066] The drug-nutrition interaction adaptation module comprises:

[0067] A drug database unit is configured to integrate clinical commonly used chronic disease drugs (such as hypoglycemic drugs, antihypertensive drugs, anticoagulant drugs, etc.), cover 500+ drugs, and mark the contraindicated nutrients (such as warfarin contraindicated high vitamin K food, metformin contraindicated high lactic acid food) and recommended synergistic nutrients (such as statins requiring coenzyme Q10 supplementation) of each drug; at the same time, the patient's medication record (such as “insulin 30R + clopidogrel + furosemide”) of the hospital HIS system is automatically synchronized, without manual input;

[0068] An interaction rule unit is configured to establish a drug combination-nutrition constraint mapping relationship, for example:

[0069] When the patient takes “insulin + spironolactone” (hypoglycemic + potassium-sparing diuretic) at the same time, an additional constraint of “potassium intake ≤1.5g / day” is automatically added (the original kidney potassium constraint may be 2g / day, which needs to be superimposed with the influence of drugs);

[0070] When the patient takes “methotrexate” (immunosuppressive, commonly used in patients with rheumatoid arthritis combined with diabetes), the supplement requirement of “folic acid ≥0.4mg / day” is automatically added;

[0071] A risk warning unit is configured to, if there is a drug-nutrition conflict in the initially generated nutrition scheme (for example, the scheme for a patient taking warfarin contains “spinach 200g” (vitamin K 200μg)), the risk level (for example, “high risk: vitamin K exceeds 3 times”) is automatically marked, and the alternative scheme (for example, spinach is replaced by lettuce 50g (vitamin K 15μg)) is pushed.

[0072] The dynamic intake difference analysis module comprises:

[0073] A multi-source data acquisition unit is configured to acquire diet data, physiological data, and state data in real time;

[0074] The diet data comprises weighing the weight of food materials by using an intelligent plate and recording snacks / supplements by using a diet APP;

[0075] The physiological data comprises detecting blood pressure, blood glucose, blood oxygen, etc. by using a wearable device, and acquiring liver and kidney function, blood lipids, etc. by using a hospital LIS system;

[0076] The state data includes obtaining acute exacerbation markers by using electronic medical records, obtaining appetite and digestion by using patient self-evaluation, etc.

[0077] The difference calculation unit is configured to calculate the nutrient difference in two dimensions of real-time difference (single meal / single day) and cumulative difference (week / month).

[0078] The difference attribution unit is configured to analyze the difference causes (such as “protein difference exceeds the standard = diet record error? Disease deterioration? Unreasonable scheme?”) by using an AI algorithm, and output the attribution results (such as “80% probability of scheme not adapting to kidney disease progression”).

[0079] The calculation formula of the nutrient difference is:

[0080] A certain nutrient difference = (target intake amount under the current physiological state) - (actual intake amount)

[0081] The target intake amount is dynamically updated according to physiological data, such as reducing the target protein amount by 15% when the serum creatinine of a kidney disease patient increases by 10%.

[0082] The quantitative matching optimization module includes:

[0083] The conflict coordination unit is configured to dynamically adjust the conflicting nutrients according to the priority of the knowledge base, and the rule is that high-priority constraints are satisfied first, and low-priority constraints are adapted within the allowed range. (For example, for a patient with coronary heart disease and liver disease, first satisfy “total fat ≤ 50 g / day”, and then allocate “Omega-3 ≥ 2 g / day” within the range.)

[0084] The personalized scheme generation unit is configured to generate a visual scheme of a food list, a portion, and cooking suggestions by combining the patient's dietary preferences (such as not eating seafood and liking spicy food), cooking methods (such as steaming, boiling, and frying), and replace “deep sea fish oil” with “15 g of flaxseed” to adapt to seafood allergy patients.

[0085] The special dietary disorder adaptation module is configured to identify the type of dietary disorder, establish a disorder-food form mapping rule, convert the food in the basic scheme to an adapted form, and provide visual preparation guidance while ensuring the nutrient content remains unchanged.

[0086] The real-time adjustment unit is configured to automatically update the target intake amount and the matching scheme within 10 minutes when the physiological data triggers a threshold (such as blood glucose > 8.3 mmol / L and serum creatinine increase > 10%).

[0087] The special dietary disorder adaptation module includes:

[0088] The disorder type identification unit is used to first divide the eating disorder into dysphagia (mild / moderate / severe), mastication disorder, food allergy (such as seafood / gluten allergy), and digestive absorption disorder (such as irritable bowel syndrome) by patient self-evaluation (such as “difficult to swallow” and “loose teeth”), clinical diagnosis (such as “post-cerebral infarction sequelae” and “oral mucosa lesion”), and intelligent device detection (such as swallowing function detector data); and then match the food material form standard (such as the need for paste for severe dysphagia and the need for paste for moderate dysphagia) and the list of contraindicated food materials (such as the need to avoid allergens for allergic patients) for each type of disorder.

[0089] The scheme form conversion unit is used to automatically convert the food materials of the basic scheme into an adapted form, for example:

[0090] For patients with severe dysphagia, “steamed fish 100g + rice 50g” is converted into “fish paste 100g (thickened by adding lotus root powder) + rice paste 50g”, while ensuring that the protein and carbohydrate contents remain unchanged;

[0091] For patients with gluten allergy, “buckwheat noodles 50g” is replaced by “corn noodles 50g”, and the dietary fiber intake is adjusted synchronously (to avoid the difference fluctuation caused by the replacement of food materials);

[0092] The preparation guidance unit is used to first generate visual operation guidelines (such as “fish paste preparation: remove the spine after steaming, add 5ml of warm water + 2g of corn starch, and beat in a homogenizer until smooth and particle-free”) to adapt to the home cooking scene; and then link the intelligent cooking equipment (such as a homogenizer with a “chronic disease mode”) to support one-key generation of adapted food material forms.

[0093] The feedback iteration module includes:

[0094] The collection unit is used to collect the patient's scheme execution rate (such as “whether to eat according to the recommended serving size”) and physiological index changes (such as blood pressure and liver and kidney function), and optimize the knowledge base rules through machine learning (such as adjusting the priority weight of certain comorbidities);

[0095] The generation unit is used to generate a nutrition management effect report every month, compare the difference change trend, and the incidence of complications, to provide a basis for clinical adjustment;

[0096] In specific use, the specific steps are as follows:

[0097] S1: The comorbidity nutrition demand knowledge base module is used to classify the common clinical comorbidity combinations according to organ systems and disease severity, and match the basic constraints for each type of comorbidity, and set the nutrition demand priority;

[0098] S2: Integrate the clinical slow disease drug database through the drug-nutrition interaction adaptation module, automatically synchronize the patient's medication record, identify the corresponding contraindicated nutrients and recommended synergistic nutrients of the drug, generate the final nutrient constraints superimposed by the drug effect, and prewarn the risk of the conflict scheme and make alternative recommendations;

[0099] S3: Collect multi-dimensional real-time data through the dynamic intake difference analysis module, combine the final nutrient constraints, calculate the nutrient intake difference in real time and cumulative dimensions, and attribute the difference to the cause through the AI algorithm, and output the risk level;

[0100] S4: Establish a nutrition demand priority coordination mechanism through the quantitative matching optimization module, prioritize high-risk constraints, and generate a basic food material scheme based on the patient's dietary preferences and cooking habits, while supporting rapid adjustment of the scheme based on real-time physiological data;

[0101] S5: Collect operation logs of each module through the feedback iteration module, combine the changes in patient physiological indicators, and dynamically iterate to continuously improve the system's adaptation accuracy in complex comorbidity scenarios.

[0102] Although the present application has been described with reference to the embodiments above, various improvements can be made thereto and equivalents can be substituted therefor without departing from the scope of the present application. In particular, features of the disclosed embodiments can be combined together in any manner provided that there is no structural conflict. Such combinations are not exhaustively described in the specification only for the sake of brevity and resource conservation. Therefore, the present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A system for analyzing and quantitatively matching nutritional intake differences in complex chronic comorbidity scenarios, characterized in that, Comorbidity nutritional requirement knowledge base module for classifying common clinical comorbidity combinations by organ system and disease severity, matching basic constraints for each comorbidity category, and setting nutritional requirement priorities; Drug-nutrient interaction adaptation module for integrating a clinical chronic disease drug database, automatically synchronizing patient medication records, identifying contraindicated nutrients and recommended synergistic nutrients corresponding to the drugs, generating final nutritional constraints superimposed with drug effects, and providing risk warnings and alternative recommendations for conflicting solutions; Dynamic intake difference analysis module for collecting multi-dimensional real-time data, combining final nutritional constraints, calculating nutritional intake differences in real time and cumulative dimensions, and attributing the differences to causes through AI algorithms while outputting risk levels; Quantitative matching optimization module for establishing a nutritional requirement priority coordination mechanism, prioritizing high-risk constraints, generating basic food material solutions in combination with patient dietary preferences and cooking habits, and supporting rapid adjustments to the solutions based on real-time physiological data; Feedback iteration module for collecting operation logs of each module, combining changes in patient physiological indicators, and dynamically iterating to continuously improve the system's adaptation accuracy for complex comorbidity scenarios. The comorbidity nutritional requirement knowledge base module includes:

2. The system for analysis and quantitative matching of nutritional intake differences in the context of complex chronic comorbidities according to claim 1, characterized in that, Comorbidity combination classification unit for classifying comorbidities into multiple categories according to organ systems and disease severity to cover common clinical combinations; Nutritional constraint rule unit for matching multi-dimensional constraints for each comorbidity category and labeling conflict relationships; Priority assignment unit for assigning priorities based on disease acute / chronic properties and nutritional sensitivity. The drug-nutrient interaction adaptation module includes:

3. The system for nutritional intake difference analysis and quantitative matching in complex chronic comorbidity scenarios according to claim 1, characterized in that, Drug database unit for integrating commonly used clinical chronic disease drugs and labeling the contraindicated nutrients and recommended synergistic nutrients for each drug, while supporting automatic synchronization of patient medication records from hospital HIS systems without manual input; Interaction rule unit for establishing a mapping relationship between drug combinations and nutritional constraints; Risk warning unit for automatically labeling risk levels and providing alternative solutions if there is a drug-nutrient conflict in the initially generated nutritional solution. The dynamic intake difference analysis module includes:

4. The nutritional intake difference analysis and quantitative matching system in a complex chronic comorbidity scenario according to claim 1, characterized in that, Multi-source data collection unit for real-time collection of dietary data, physiological data, and state data; Difference calculation unit for calculating nutritional differences in real-time and cumulative dimensions; Difference attribution unit for analyzing difference causes through AI algorithms and outputting attribution results. The calculation formula for nutritional differences is:

5. The system for analysis and quantitative matching of nutritional intake differences in the context of complex chronic comorbidities according to claim 4, characterized in that, Nutrient difference = (target intake under current physiological state) - (actual intake). The quantitative matching optimization module includes:

6. The nutritional intake difference analysis and quantitative matching system in a complex chronic comorbidity scenario according to claim 1, characterized in that, Conflict coordination unit for dynamically adjusting conflicting nutrients based on knowledge base priorities; Individualized solution generation unit for generating visual solutions for food material lists, portion sizes, and cooking suggestions in combination with patient dietary preferences and cooking methods; Special dietary disorder adaptation module for identifying dietary disorder types, establishing disorder-food material form mapping rules, converting food materials in the basic solution to adapted forms, providing visual preparation guidance, and ensuring constant nutrient content; ​ A real-time adjustment unit is configured to automatically update the target intake and the matching plan within 10 minutes when the physiological data triggers the threshold.

7. The system for analysis and quantitative matching of nutritional intake differences in the context of complex chronic comorbidities according to claim 6, characterized in that, The special dietary disorder adaptation module comprises: A disorder type identification unit is configured to first divide the dietary disorder into dysphagia, mastication disorder, food allergy, and digestive absorption disorder through patient self-evaluation, clinical diagnosis, and intelligent device detection, and then match food material form standards and a list of taboo food materials for each type of disorder. A plan form conversion unit is configured to automatically convert the food materials of the basic plan into an adapted form. A preparation guidance unit is configured to first generate a visual operation guide suitable for a home cooking scene, and then link an intelligent cooking device to support one-key generation of adapted food material forms.

8. The nutritional intake difference analysis and quantitative matching system in a complex chronic comorbidity scenario according to claim 1, characterized in that, The feedback iteration module comprises: A collection unit is configured to collect the patient's plan execution rate and physiological index changes, and optimize the knowledge base rules through machine learning. A generation unit is configured to generate a nutrition management effect report every month, compare the difference change trend and the incidence of complications, and provide a basis for clinical adjustment.

Images