CN112837785A - Clinical nutrition digital diagnosis and treatment method and system - Google Patents

Abstract

The invention discloses a clinical nutrition digital diagnosis and treatment method and a system. The method comprises the following steps: establishing a database, wherein the database stores the nutrition state and disease diagnosis and treatment information of historical patients, the venous energy value and nutrient proportion information in clinical treatment, EN nutrition scheme result information formed by man-machine interaction in the individualized diagnosis and treatment process of clinicians and nutriologists, and corresponding objective effect data; according to the input disease diagnosis and treatment information of the current patient, the database utilizes an artificial intelligence algorithm to independently learn, and an EN nutrition scheme which is consistent with the disease diagnosis and treatment information of the current patient and has the optimal objective effect is matched. The invention can make the clinical nutrition diagnosis and treatment play the role of the due auxiliary clinical treatment, shorten the recovery period of patients, reduce the fatality rate and lighten the family and social burden of the patients.

Description

Clinical nutrition digital diagnosis and treatment method and system

Technical Field

The invention relates to the technical field of nutrition treatment, in particular to a clinical nutrition digital diagnosis and treatment method and a system.

Background

Nutrition is closely related to the occurrence and development of diseases and the functions of the immune system, malnutrition can aggravate the progress of diseases, including immune dysfunction, recovery delay, treatment failure, increased infection and complications, prolonged hospitalization time, increased readmission rate and mortality rate, and nutritional status is in a bidirectional influence relationship with inflammation and immune function. Malnutrition can increase host susceptibility and severity of infection in a number of ways, and nutritional status can significantly affect the response to vaccines or therapeutic drugs.

Malnutrition in most hospitals is often misdiagnosed or missed diagnosis and inappropriate nutritional support is common, and although guidelines suggest early EN in most cases, PN (parenteral nutrition) is still an overwhelming form of nutritional support. Therefore, there is an urgent need for comprehensive research on malnutrition and nutritional support, improving the awareness of malnutrition and improving the nutritional treatment effect. But due to lack of awareness of the importance of nutritional treatment, hospitalized patients are anxious in nutritional treatment.

Disclosure of Invention

The invention aims to provide a clinical nutrition digital diagnosis and treatment method and system aiming at the technical defects in the prior art.

The technical scheme adopted for realizing the purpose of the invention is as follows:

a clinical nutrition digital diagnosis and treatment method comprises the following steps:

the database stores the nutritional state and disease diagnosis and treatment information of historical patients, the venous energy value and nutrient proportion information in clinical treatment, EN nutritional scheme result information formed by man-machine interaction in the individualized diagnosis and treatment process of clinicians and nutriologists, and corresponding objective effect data;

according to the input disease diagnosis and treatment information of the current patient, the database utilizes an artificial intelligence algorithm to independently learn, and an EN nutrition scheme which is consistent with the disease diagnosis and treatment information of the current patient and has the optimal objective effect is matched.

The disease diagnosis and treatment information comprises the weight, sex, age, height, body temperature, body composition analysis indexes, disease information, disease specific index change, stress coefficient and EN contraindications of a patient.

Wherein the EN contraindications include digestive tract tolerance and fluid volume tolerance.

The EN nutrition scheme formed by man-machine interaction is formed by forming a theoretical nutrition prescription through an initial rule of an intelligent nutrition prescription generation system, selecting a target energy value, a nutrition type and various nutrition similarity ratios by a doctor according to disease information, disease specificity index change, stress coefficient, human body component analysis index and EN contraindications of a patient, deducting PN energy value, and confirming or adjusting and modifying the obtained EN energy value within a certain range by a dietician.

Preferably, the certain range is 20-150%.

Wherein the index of nutritional status comprises prealbumin, albumin, C-reactive protein, hemoglobin, lymphocyte count, total white blood cell count, and platelet count.

Wherein the disease-specific index comprises blood ammonia, blood creatinine, blood urea nitrogen BUN), blood potassium, blood sodium, blood sugar, blood fat, alanine aminotransferase ALT, total bilirubin Bil, prothrombin time PT, alkaline phosphatase che, urine volume, 24-hour urine protein quantification, creatine kinase CK, aspartate aminotransferase AST, lactate dehydrogenase LDH.

The invention also aims to provide a clinical nutrition digital diagnosis and treatment system, which comprises:

the server is provided with a database, and the database stores the nutrition state and disease diagnosis and treatment information of historical patients, the venous energy value and nutrient proportion information in clinical treatment, and the EN nutrition scheme result information and objective effect data formed by man-machine interaction in the individualized diagnosis and treatment process of clinicians and nutriologists;

and the computer is in communication connection with the server and is used for inputting the current disease diagnosis and treatment information of the patient, receiving the database of the server, autonomously learning by utilizing an artificial intelligence algorithm according to the current disease diagnosis and treatment information of the patient, matching the output EN nutrition scheme which is consistent with the current disease diagnosis and treatment information of the patient and has the optimal objective effect, and displaying the EN nutrition scheme on a display interface.

Preferably, the computer is connected with a wireless-controlled online weighing device, a human body composition analysis device and a hospital medical record system.

The invention can make the clinical nutrition diagnosis and treatment play the role of the due auxiliary clinical treatment, shorten the recovery period of patients, reduce the fatality rate and lighten the family and social burden of the patients. By building an intelligent platform with cross fusion of clinical medicine and nutriology, the system can play a role with high efficiency no matter in a special period (epidemic situation) or daily.

Drawings

FIG. 1 is a process diagram of the clinical nutrition digital diagnosis and treatment method of the present invention;

fig. 2 is a schematic construction diagram of the clinical nutrition digital diagnosis and treatment system of the invention.

Detailed Description

The invention is described in further detail below with reference to the figures and specific examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The invention aims to carry out accurate individualized nutrition diagnosis and treatment and objective evaluation of clinical effects of malnutrition patients, automatically store relevant information such as clinical nutrition diagnosis and treatment processes and effects in a database or a server, upload the information to the cloud after information security filtering, accumulate the data into big data of the clinical nutrition diagnosis and treatment and the effect evaluation, carry out automatic operation, induction and summarization by a set artificial intelligence algorithm, and finally form an accurate individualized clinical nutrition diagnosis and treatment scheme for the malnutrition patients.

As shown in fig. 1, the clinical nutrition digital diagnosis and treatment method of the present invention comprises the following steps:

the database stores the nutritional state and disease diagnosis and treatment information of historical patients, the venous energy value and nutrient proportion information in clinical treatment, EN nutritional scheme result information formed by man-machine interaction in the individualized diagnosis and treatment process of clinicians and nutriologists, and corresponding objective effect data;

according to the input disease diagnosis and treatment information of the current patient, the database utilizes an artificial intelligence algorithm to independently learn, and an EN nutrition scheme which is consistent with the disease diagnosis and treatment information of the current patient and has the optimal objective effect is matched.

The disease diagnosis and treatment information comprises the weight, sex, age, height, body temperature, body composition analysis indexes, disease information, disease specific index change, stress coefficient and EN contraindications of a patient.

Wherein the EN contraindications include digestive tract tolerance and fluid volume tolerance.

The EN nutrition scheme formed by man-machine interaction is formed by forming a theoretical nutrition prescription through an initial rule of an intelligent nutrition prescription generation system, selecting a target energy value, a nutrition type and various nutrition similarity ratios by a doctor according to disease information, disease specificity index change, stress coefficient, human body component analysis index and EN contraindications of a patient, deducting PN energy value, and confirming or adjusting and modifying the obtained EN energy value within a certain range by a dietician.

In the above technical solution, the human body composition analysis device outputs parameters by detecting a patient, the parameters including: body FAT percentage (FAT%), body FAT mass, muscle mass, body water mass (TBW), protein, inorganic salt, body weight, target body weight, body weight to be increased or decreased, body FAT mass and muscle mass, Basal Metabolic Rate (BMR), Body Mass Index (BMI), FAT mass per site, FAT percentage, muscle mass (right leg, left leg, right arm, left arm, torso), and the like.

In the above technical solution, the database information includes:

1. patient information: the affected area, the number of the bed, the age, the sex, the height, the weight, the body temperature and the stress coefficient.

2. Nutritional status and disease information: clinical and nutritional diagnosis, and screening of examination results related thereto; wherein the nutritional indicators (commonalities) include: prealbumin, albumin, C-reactive protein, hemoglobin, lymphocyte count, total white blood cells, platelet count.

Disease-specific markers include blood ammonia, blood creatinine, Blood Urea Nitrogen (BUN), blood potassium, blood sodium, blood glucose, blood lipids, digital alanine Aminotransferase (ALT), total bilirubin (Bil), Prothrombin Time (PT), alkaline phosphatase (che), urine volume, 24 hour urine protein quantification, Creatine Kinase (CK), aspartate Aminotransferase (AST), Lactate Dehydrogenase (LDH), and the like. Scoring information for each different disease category.

3. Venous energy value and nutrient proportion information in clinical treatment. PN (parenteral nutrition) information includes sugars: 10% of glucose, 5% of glucose, 10% of compound electrolyte glucose MG3, 5% of glucose sodium chloride, 1% of sodium potassium magnesium calcium glucose, 10% of fructose (Fenghai Neng), 10% of fructose (Prikang). Amino acids (proteins): 20% alanyl glutamine, 40% low molecular dextran amino acid, 10% compound amino acid (Anping), 11.4% compound amino acid 18 AA-II (11.4), 8.5% compound amino acid 18 AA-II (8.5), 4.26% compound amino acid 3AA, 5.59% compound amino acid 9AA, and 6.74% compound amino acid 18AA for children. Fat: fat milk amino acid glucose [ Kavin ], 20% medium/long-chain fat milk (C6-24), 20% medium/long-chain fat milk (C8-24v), etc.

4. The human-computer interaction track in the individual diagnosis and treatment process of the clinician and the nutritionist and the result information formed by the clinical and nutritional subject fusion comprise the nutrition treatment guidelines of nearly 3 years at home and abroad according to which the preliminary rules of the human-computer exchange intelligent nutrition prescription generation system approved by the clinician and the nutritionist are formed through repeated discussion and communication of nutrition and clinical (ICU) experts. According to the Harris-Benedict formula: male BEE: 66.4730+13.751 body weight (kg) +5.0033 height (cm) -6.7550 age (years) ═ Kcal. Female BEE: 655.0955+9.463 body weight (kg) +1.8496 height (cm) -4.6755 age (years) ═ Kcal. The average required value of the Chinese is 12.5 percent lower than the result value of the formula, and the theoretical value of the energy requirement of the patient is obtained.

And starting a man-machine fusion exchange mechanism, and adjusting by a doctor within the range of 20-150% of a theoretical value according to the acquired accurate information and by combining the illness state and the digestion bearing capacity of the patient. The method comprises the following steps:

s1, starting comprehensive operation according to a preliminary rule and displaying the theoretical energy value of a patient and the total energy and types referred by a doctor, namely the ratio (%) of three nutrients including carbohydrate, protein and fat.

S2, the doctor screens nutrition risks and determines whether EN contraindications exist according to the disease conditions, wherein the EN contraindications include digestive tract bearing capacity, liquid quantity bearing capacity and the like, and target energy values and types (the proportion of the three nutrients) are selected.

And S3, the system comprehensively calculates the total energy (PN + EN) and the type with the PN energy value and the naturally formed type in clinical treatment, and displays the EN energy value and the type to be executed (man-machine interaction).

S4, confirming or modifying by a nutriologist (the range is less than or equal to 10%), and confirming again by a clinician (human-to-human interaction).

S5, the two parties confirm to enter a manual or automatic handler. The clinical nutrition digital diagnosis and treatment system integrates respective emphasis points of a clinician and a dietician in a nutrition prescription forming mode to form interdisciplinary study.

Therefore, a new mode of subject fusion, man-machine interaction and intelligent generation of EN accurate nutrition prescriptions is realized, and the audience quantity and efficiency of nutrition diagnosis and treatment are improved.

5. The information of the patient, the clinical nutrition digital diagnosis and treatment process and the outcome condition thereof are stored to form a database, the database comprises the weight of the patient, the analysis change of human body components, nutrition indexes, disease specificity index change, the difference between the actual nutrition prescription and theoretical and target energy values and proportions, and the database is stored in association with the nutrition and disease outcome of the patient.

6. The method comprises the steps of screening clinical nutrition digital diagnosis and treatment processes and relevant information of outcome thereof according to clinical needs, setting keywords observed from different angles, circularly operating to obtain regular data, and forming a continuously updated nutrition prescription generation rule through autonomous learning.

The artificial intelligence algorithm can correlate the nutritional common indexes before and after the patients and the specific indexes of the whole diseases with the nutritional diagnosis and treatment process and effect, thereby obtaining the optimal nutritional diagnosis and treatment scheme of the patients with different disease types, disease periods, disease conditions and different ages and sexes.

For example, the AdaBoost algorithm and the decision tree algorithm are combined to train learning together. The AdaBoost algorithm calls classifiers with poor classification effect as weak classifiers and the classifiers with good classification effect as strong classifiers. The basic principle of the Adaboost algorithm is to reasonably combine a plurality of weak classifiers (the weak classifiers generally adopt a single-layer decision tree) to form a strong classifier. Adaboost adopts the idea of iteration, only one weak classifier is trained in each iteration, and the trained weak classifier participates in the use of the next iteration. That is, in the nth iteration, there are N weak classifiers, of which N-1 are trained before, and various parameters are not changed, and the nth classifier is trained this time. The weak classifiers are related in such a way that the Nth weak classifier is more likely to classify the data which is not classified by the first N-1 weak classifiers, and the final classification output needs to see the comprehensive effect of the N classifiers.

As shown in fig. 2, the present invention also provides a clinical nutrition digital diagnosis and treatment system, comprising:

the server is provided with a database, and the database stores the nutrition state and disease diagnosis and treatment information of historical patients, the venous energy value and nutrient proportion information in clinical treatment, and the EN nutrition scheme result information and objective effect data formed by man-machine interaction in the individualized diagnosis and treatment process of clinicians and nutriologists;

and the computer is in communication connection with the server and is used for inputting the current disease diagnosis and treatment information of the patient, receiving the database of the server, autonomously learning by utilizing an artificial intelligence algorithm according to the current disease diagnosis and treatment information of the patient, matching the output EN nutrition scheme which is consistent with the current disease diagnosis and treatment information of the patient and has the optimal objective effect, and displaying the EN nutrition scheme on a display interface.

The computer comprises an input device which comprises a display terminal, a keyboard or a mouse or a touch screen interface input device, is internally provided with corresponding processing program software and is used for inputting corresponding retrieval parameters or keywords on a configured program retrieval interface and receiving corresponding information for retrieval input by external equipment, such as the weight of a patient, body composition information and the like, and filling the corresponding position for input, can be configured into a selection interface with corresponding information for input, such as a pull-down menu, and is convenient to use when a user selects the corresponding input information on the program interface, and can be configured with a result display module so as to display a corresponding matched EN nutrition scheme with the optimal objective effect in one area of the display terminal.

Preferably, the computer is connected with a wireless-controlled online weighing device, a human body composition analysis device and a nutrition processing device. Wherein, the database can be connected with a medical record system of a hospital to collect and update the data information quantity thereof.

The online weighing device can be connected with a sickbed and arranged on the sickbed, so that the patient can be weighed online, the obtained information is transmitted to a computer for storage or processing, and the human body composition analysis device can analyze the human body composition of the patient online and transmit the analysis result to the computer for processing.

Wherein, the human composition analysis device outputs parameters including, by detecting a patient: body FAT percentage (FAT%), body FAT mass, muscle mass, body water mass (TBW), protein, inorganic salt, body weight, target body weight, body weight to be increased or decreased, body FAT mass and muscle mass, Basal Metabolic Rate (BMR), Body Mass Index (BMI), FAT mass per site, FAT percentage, muscle mass (right leg, left leg, right arm, left arm, torso), and the like.

Wherein, the nutrition disposal device is a nutrition disposal machine which is used for automatically configuring corresponding nutrients according to the output EN nutrition scheme under the control of the computer, automatically packaging and providing the nutrients for patients to use. The nutrient treatment device can be configured to send a treatment signal to the nutrient treatment device by the computer according to the output EN nutrient scheme, and the nutrient treatment device can automatically package according to the EN nutrient scheme, or can be manually controlled and operated according to the EN nutrient scheme to package nutrients.

According to the invention, by establishing a clinical nutrition digital diagnosis and treatment equipment system, a doctor starts a clinical nutrition treatment basic mode of interdisciplinary and fusion of man-machine interaction and dietitian interaction, and provides a simple and accurate individual nutrition prescription for comprehensively considering theories, targets, PN and EN energy values, carbohydrate, protein and fat proportions, patient energy and liquid bearing capacity for critically ill patients, so that the nutrition diagnosis and treatment public quantity is efficiently expanded.

The invention utilizes a clinical nutrition digital diagnosis and treatment equipment system to divide a clinician and a dietician into a plurality of parts in clinical nutrition diagnosis and treatment of severe patients, and stimulates the mutual learning and scientific and technological exploration interests of the two parts. The method explores the possibility of changing the current situation that all basic indexes of nutriology used at present use foreign standards including the principles of special medical food formula and the like, stores a database to form a shared platform, and obtains the optimal nutritional diagnosis and treatment schemes of different disease types, disease stages and conditions and patients of different ages and sexes through the cyclic learning and correction of an artificial intelligent algorithm, thereby laying a foundation for independently establishing the nutritional diagnosis and treatment standards of China.

The invention can finish the accurate nutrition diagnosis and treatment outside the ward by utilizing the characteristic of wireless control, so that the accurate nutrition treatment effect can be efficiently exerted no matter in special period (epidemic situation) and daily life, and the overall treatment effect of the patient is improved.

The invention solves the problem that the clinical nutrition development is influenced by the deficiency of clinical nutrilites and the serious shortage of talents: the function of introducing a high-tech nutrition diagnosis and treatment system is exerted, so that a basic mode of a clinical nutrition digital diagnosis and treatment equipment system with interdisciplinary and fusion of human-computer interaction and dietician interaction is established by a doctor, a simple and accurate individualized nutrition prescription comprehensively considering theories, targets, PN and EN energy values, carbohydrate, protein, fat proportion, patient energy and liquid bearing capacity is provided for critically ill patients, and the nutrition diagnosis and treatment audience quantity is efficiently expanded on the premise of not increasing personnel.

The invention solves the problems that the clinical nutrition diagnosis and treatment research data is incomplete and the evaluation index of the nutrition treatment effect is questioned: the complete database of the accurate individualized clinical nutrition diagnosis and treatment process and the objective effect is established, the nutrition diagnosis and treatment effect is objectively evaluated from the aspects of dynamically recorded nutrition indexes, disease clinical indexes, lean body mass and the like, and the information related to patients is stored in the server, so that the storage of the related nutrition and disease diagnosis and treatment information, the nutrition diagnosis and treatment process and the objective effect of patients suffering from malnutrition and severe patients is realized, and related shared big data is formed.

According to the invention, by establishing an artificial intelligence algorithm system, the accumulated clinical nutrition diagnosis and treatment process and objective effect information are stored to form big data, and artificial intelligence algorithm treatment in different angles is carried out, and a brand new clinical nutrition diagnosis and treatment mode is created by means of the artificial intelligence algorithm, including autonomous learning, inductive summarization and the like, so that nutrition diagnosis and treatment rules or schemes of severe patients with malnutrition are obtained, and a foundation is laid for comprehensively developing clinical nutrition research in China.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims (9)

1. A clinical nutrition digital diagnosis and treatment method is characterized by comprising the following steps:

establishing a database, wherein the database stores the nutrition state and disease diagnosis and treatment information of historical patients, the venous energy value and nutrient proportion information in clinical treatment, EN nutrition scheme result information formed by man-machine interaction in the individualized diagnosis and treatment process of clinicians and nutriologists, and corresponding objective effect data;

according to the input disease diagnosis and treatment information of the current patient, the database utilizes an artificial intelligence algorithm to independently learn, and an EN nutrition scheme which is consistent with the disease diagnosis and treatment information of the current patient and has the optimal objective effect is matched.

2. The clinical nutritional digital diagnosis and treatment method according to claim 1, wherein the disease diagnosis and treatment information includes weight, sex, age, height, body temperature, body composition analysis index, disease information, disease specific index change, stress coefficient and EN contraindication of the patient.

3. The clinical nutritional digital diagnosis and treatment method according to claim 2, wherein the EN contraindications include digestive tract tolerance and liquid volume tolerance.

4. The clinical nutritional digital diagnosis and treatment method according to claim 1, wherein the EN nutrition plan formed by human-computer interaction is formed by forming a theoretical nutrition prescription through an initial rule of an intelligent nutrition prescription generation system, selecting target energy values, nutrition types and various nutrition similarity ratios by doctors in combination with disease information, disease specificity index changes, stress coefficients, body composition analysis indexes and EN contraindications of patients, and then confirming or adjusting and modifying the obtained EN energy values within a certain range after subtracting the PN energy values.

5. The clinical nutritional digital diagnosis and treatment method according to claim 4, wherein the certain range is 20-150%.

6. The digital clinical nutritional diagnosis and treatment method according to claim 1, wherein the index of nutritional status includes blood prealbumin, albumin, C-reactive protein, hemoglobin, lymphocyte count, total white blood cell count, and platelet count.

7. The clinical nutritional digital diagnosis and treatment method according to claim 2, wherein the disease specificity index comprises blood ammonia, blood creatinine, blood urea nitrogen BUN), blood potassium, blood sodium, blood sugar, blood fat, alanine aminotransferase ALT, total bilirubin Bil, prothrombin time PT, alkaline phosphatase che, urine volume, 24-hour urine protein quantification, creatine kinase CK, aspartate aminotransferase AST, lactate dehydrogenase LDH.

8. A clinical nutrition digital diagnosis and treatment system, comprising:

the server is provided with a database, and the database stores the nutrition state and disease diagnosis and treatment information of historical patients, the venous energy value and nutrient proportion information in clinical treatment, and the EN nutrition scheme result information and objective effect data formed by man-machine interaction in the individualized diagnosis and treatment process of clinicians and nutriologists;

and the computer is in communication connection with the server and is used for inputting the current disease diagnosis and treatment information of the patient, receiving the database of the server, autonomously learning by utilizing an artificial intelligence algorithm according to the current disease diagnosis and treatment information of the patient, matching the output EN nutrition scheme which is consistent with the current disease diagnosis and treatment information of the patient and has the optimal objective effect, and displaying the EN nutrition scheme on a display interface.

9. The clinical nutritional digital diagnosis and treatment system according to claim 8, wherein the computer is connected with a wireless-controlled online weighing device, a human body composition analysis device and a hospital medical record system.

Images