Method, apparatus, and storage medium for recommending nutritional intake for icu patients

Abstract

This invention relates to the field of data processing and discloses a method, apparatus, device, and storage medium for recommending nutritional intake for ICU patients. The method for recommending nutritional intake for ICU patients includes: acquiring multi-source heterogeneous medical record data of the patient; generating a target nutritional intake plan for the patient based on the multi-source heterogeneous medical record data; if it is determined that the target nutritional intake plan is being implemented, monitoring the patient's real-time nutritional data; dynamically optimizing the target nutritional intake plan based on the real-time nutritional data and a pre-built critical care nutrition knowledge graph, and generating nutritional recommendation information. The critical care nutrition knowledge graph is a pre-constructed knowledge base containing multiple nutritional management strategy information entries. Each nutritional management strategy entry is associated with corresponding applicable conditions, adjustment actions, and decision-making basis information. The knowledge base is periodically updated with evaluated content. This invention improves the completion rate of nutritional plans for ICU patients.

Description

Technical Field

[0001] This invention belongs to the field of data processing, and in particular relates to a method, apparatus, device and storage medium for recommending nutritional intake for ICU patients. Background Technology

[0002] Traditional ICU patient nutrition management methods typically focus on enteral nutrition intake, neglecting parenteral nutrition intake. In the acute phase of illness, especially for patients with severe traumatic brain injury, delayed gastric emptying and paralytic ileus make it difficult to obtain sufficient nutrition through enteral nutrition, necessitating parenteral nutrition supplementation. Furthermore, in addition to common parenteral nutrition preparations, solvents such as glucose solutions and propofol, as well as sedatives, all contain energy substances and must be calculated together to avoid excessive energy intake.

[0003] While traditional methods can prompt adjustments to feeding protocols, they lack personalized and dynamic nutritional strategy recommendations, making it difficult for healthcare professionals to adjust feeding protocols based on evidence. This results in poor completion rates of nutritional plans for ICU patients, necessitating a new technological approach to address these issues. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, apparatus, device and storage medium for recommending nutritional intake for ICU patients, which can solve the problem of the lack of dynamic and precise nutritional programs for critical care patients in clinical practice.

[0005] The first aspect of this invention provides a method for recommending nutritional intake for ICU patients, comprising: Acquire multi-source heterogeneous medical record data of patients, which includes basic physiological indicators, examination data and medication records; Based on multi-source heterogeneous medical record data, a target nutritional intake plan for patients is generated, which includes nutritional routes, intake amounts, and intake timing. If the target nutrition intake plan is determined to be in progress, monitor the patient's real-time nutritional data. The real-time nutritional data includes enteral nutrition intake, parenteral nutrition intake, occurrence of complications, and laboratory indicators, including albumin and prealbumin levels. Based on real-time nutrition data and a pre-built critical care nutrition knowledge graph, the target nutrition intake plan is dynamically optimized and nutrition recommendation information is generated. The critical care nutrition knowledge graph is a pre-built knowledge base containing multiple nutrition management strategies that have been evaluated and uploaded. Each nutrition management strategy is associated with corresponding applicable conditions, adjustment actions, and decision-making basis information. The knowledge base is regularly updated with evaluated content.

[0006] Optionally, in a first implementation of the first aspect of the present invention, generating a target nutritional intake plan for the patient based on multi-source heterogeneous medical record data includes: Based on the height and weight in the multi-source heterogeneous medical record data, the patient's basal energy expenditure was calculated, and the daily target calorie and target protein intake were determined based on the disease course information and test indicators in the multi-source heterogeneous medical record data. The target daily calorie intake and target daily protein intake are used as intake amounts. Based on examination data and medication records from multi-source heterogeneous medical records, the nutritional pathways and intake sequences are determined to obtain a target nutrient intake plan.

[0007] Optionally, in a second implementation of the first aspect of the present invention, the target nutritional intake plan is dynamically optimized based on real-time nutritional data and a preset critical care nutrition knowledge graph, including: Calculate the sum of the actual intake of enteral nutrition and the actual intake of parenteral nutrition to obtain the cumulative intake for the day; When the difference between the cumulative daily intake and the intake in the target nutrition intake plan is lower or higher than the preset threshold, the intake is updated by matching the corresponding adjustment strategy according to the critical care nutrition knowledge graph. When complications occur and the preset conditions are met, the first strategy information is matched according to the critical care nutrition knowledge graph, and the nutritional pathway and / or intake sequence are updated according to the first strategy information. When the test indicators fail to reach the preset expected target, the second strategy information is matched according to the critical care nutrition knowledge graph, and the intake, nutritional route and / or intake sequence are updated according to the second strategy information.

[0008] Optionally, in a third implementation of the first aspect of the present invention, if it is determined that the target nutrition intake plan is being implemented, then the patient's real-time nutritional data is detected, including: If it is determined that the target nutrition intake plan is being implemented, the patient's actual enteral nutrition intake and actual parenteral nutrition intake are obtained in real time. Real-time monitoring of patient complications, including at least one of abdominal distension, diarrhea, refeeding syndrome, constipation, and tolerance score; Real-time monitoring of patients' test results.

[0009] Optionally, in a fourth implementation of the first aspect of the present invention, generating nutritional recommendation information includes: The dynamically optimized target nutrient intake plan will be used as the recommended content. Extract decision-making basis information corresponding to this optimization from the critical care nutrition knowledge graph; The recommended content is linked with the information on the basis for decision-making to generate nutritional recommendation information.

[0010] Optionally, in the fifth implementation of the first aspect of the present invention, before the step of dynamically optimizing the target nutrition intake plan based on real-time nutrition data and a preset critical care nutrition knowledge graph, the method further includes: Periodically update the critical care nutrition knowledge map.

[0011] Optionally, in a sixth implementation of the first aspect of the present invention, acquiring the patient's multi-source heterogeneous medical record data includes: Initial multi-source heterogeneous medical record data of patients are collected from the electronic medical record system; Preprocessing is performed on the initial case data to obtain multi-source heterogeneous medical record data. The preprocessing includes data cleaning, data anonymization, data standardization, and data association and integration.

[0012] Secondly, embodiments of the present invention provide a nutritional intake recommendation device for ICU patients, the device comprising: The acquisition module is used to acquire patients' multi-source heterogeneous medical record data, which includes basic physiological indicators, examination and test data, and medication records. The generation module is used to generate a target nutritional intake plan for patients based on multi-source heterogeneous medical record data. The target nutritional intake plan includes nutritional routes, intake amounts, and intake timing. The monitoring module is used to monitor the patient's real-time nutritional data if it is determined that the target nutritional intake plan is being implemented. The real-time nutritional data includes enteral nutrition intake, parenteral nutrition intake, occurrence of complications, and test indicators. The optimization module is used to dynamically optimize the target nutrition intake plan based on real-time nutrition data and a preset critical care nutrition knowledge graph, and generate nutrition recommendation information.

[0013] Thirdly, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for recommending nutritional intake for ICU patients.

[0014] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for recommending nutritional intake for ICU patients.

[0015] Fifthly, embodiments of the present invention provide a computer program product that, when run on an electronic device, causes the electronic device to execute the above-described method for recommending nutritional intake for ICU patients.

[0016] The beneficial effects of this invention compared to existing technologies are as follows: By acquiring multi-source heterogeneous medical record data of patients, a target nutritional intake plan including nutritional routes, intake amounts, and intake timing is generated. During the execution of the plan, real-time nutritional data including enteral nutrition intake, parenteral nutrition intake, complication occurrence, and test indicators are monitored in real time. Based on a preset critical care nutrition knowledge graph, the target nutritional intake plan is dynamically optimized and nutritional recommendation information is generated, achieving comprehensive monitoring and accurate assessment of the nutritional status of ICU patients, overcoming the limitations of traditional technologies that only focus on enteral nutrition intake. By introducing a critical care nutrition knowledge graph for dynamic optimization, the nutritional recommendation strategy can be personalized according to the patient's real-time health profile, and each optimization is associated with corresponding decision-making information, effectively improving the completion rate of nutritional plans for ICU patients. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For medical personnel in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of Embodiment 1 of the method for recommending nutritional intake for ICU patients in this invention. Figure 2 This is a schematic diagram of Embodiment 2 of the method for recommending nutritional intake for ICU patients in this invention; Figure 3 This is a schematic diagram of Embodiment 3 of the method for recommending nutritional intake for ICU patients in this invention; Figure 4 This is a schematic diagram of one embodiment of the nutritional intake recommendation device for ICU patients in this invention. Figure 5 This is a schematic diagram of one embodiment of the electronic device in this invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are protected by this invention.

[0020] It should be noted that the terms "comprising," "including," and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this invention, are intended to cover non-exclusive inclusion. For example, a process, method, terminal, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices. In the claims, specification, and accompanying drawings of this invention, relational terms such as "first" and "second" are used merely to distinguish one entity / operation / object from another entity / operation / object, and do not necessarily require or imply any such immediate relationship or order between these entities / operations / objects.

[0021] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0022] Traditional ICU patient nutrition management methods typically focus on enteral nutrition intake, neglecting parenteral nutrition intake. In the acute phase of illness, especially for patients with severe traumatic brain injury, delayed gastric emptying and paralytic ileus make it difficult to obtain sufficient nutrition through enteral nutrition, necessitating parenteral nutrition supplementation. Furthermore, in addition to common parenteral nutrition preparations, solvents such as glucose solutions and propofol, as well as sedatives, all contain energy substances and must be calculated together to avoid excessive energy intake.

[0023] While traditional methods can prompt adjustments to feeding protocols, they lack personalized and dynamic nutritional strategy recommendations, making it difficult for healthcare professionals to adjust feeding protocols based on evidence. This results in poor completion rates of nutritional plans for ICU patients, necessitating a new technological approach to address these issues.

[0024] In view of this, embodiments of the present invention provide a method, apparatus, device, and storage medium for recommending nutritional intake for ICU patients. By acquiring multi-source heterogeneous medical record data of patients, a target nutritional intake plan including nutritional routes, intake amounts, and intake sequences is generated. During the execution of the plan, real-time nutritional data including enteral nutrition intake, parenteral nutrition intake, occurrence of complications, and test indicators are monitored in real time. The target nutritional intake plan is dynamically optimized based on a preset critical care nutrition knowledge graph, and nutritional recommendation information is generated. This achieves comprehensive monitoring and accurate assessment of the nutritional status of ICU patients, overcoming the limitations of traditional technologies that only focus on enteral nutrition intake. By introducing a critical care nutrition knowledge graph for dynamic optimization, the nutritional recommendation strategy can be personalized according to the patient's real-time health profile, and each optimization is associated with corresponding decision-making information, effectively improving the completion rate of nutritional plans for ICU patients.

[0025] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0026] Example 1: Figure 1 This illustration shows a flowchart of a method for recommending nutritional intake for ICU patients, provided by an embodiment of the present invention. This method can be applied to electronic devices, such as servers, service clusters, mobile phones, tablets, laptops, ultra-mobile personal computers (UMPCs), netbooks, etc.

[0027] Specifically, the above-mentioned recommended method for nutritional intake of ICU patients may include the following steps S101 to S104.

[0028] Step S101: Obtain the patient's multi-source heterogeneous medical record data, which includes basic physiological indicators, examination data, and medication records.

[0029] In embodiments of this invention, multi-source heterogeneous data related to patients are collected from various information systems within the hospital. The acquired data includes at least basic physiological indicators, examination data, enteral and parenteral nutrition complications, and medication usage records. This multi-source heterogeneous data originates from different systems; if the formats are inconsistent, they are integrated. Basic physiological indicators may include, for example, height, weight, and disease duration; examination data may include laboratory test results, such as albumin and globulin levels.

[0030] Optionally, when acquiring data, it can be further extracted directly from the electronic medical record system, and the initial data can be preprocessed, such as data cleaning, desensitization, standardization, and data association and integration.

[0031] Step S102: Based on multi-source heterogeneous medical record data, generate a target nutritional intake plan for the patient. The target nutritional intake plan includes the nutritional route, intake amount, and intake sequence.

[0032] In an embodiment of the present invention, an initial target nutritional intake plan for the patient is generated based on multi-source heterogeneous medical record data obtained through multi-source heterogeneous data analysis. This plan includes the nutritional route, daily intake, and intake sequence. The nutritional route may include enteral nutrition, parenteral nutrition, or a combination of both, and the intake sequence includes the timing and rate of feeding.

[0033] Optionally, when generating a plan, the system can calculate basal energy expenditure based on height and weight from the patient's medical record data, and determine the daily target calorie and protein intake by combining disease progression information and test results. Furthermore, it can use examination data and medication records to assist in deciding on nutritional pathways and intake timing, making the plan more tailored to the patient's actual condition.

[0034] Step S103: If it is determined that the target nutrition intake plan is being implemented, monitor the patient's real-time nutrition data. The real-time nutrition data includes enteral nutrition intake, parenteral nutrition intake, occurrence of complications, and test indicators, including albumin and prealbumin levels.

[0035] In an embodiment of the present invention, once the plan is detected to have commenced, the patient's actual nutritional data is continuously monitored. Real-time nutritional data includes at least the actual enteral nutrition intake, the actual parenteral nutrition intake, the occurrence of complications, and laboratory indicators. Complications include abdominal distension, diarrhea, refeeding syndrome, and constipation. Laboratory indicators may include albumin, globulin, and prealbumin.

[0036] Optionally, the monitoring process can be connected to devices such as infusion pumps and feeding pumps to automatically obtain the actual infusion volume; complications and laboratory indicators can be collected in real time by obtaining nursing records, laboratory data, or bedside monitoring devices.

[0037] Step S104: Based on real-time nutrition data and a pre-set critical care nutrition knowledge graph, dynamically optimize the target nutrition intake plan and generate nutrition recommendation information. The critical care nutrition knowledge graph is a pre-built knowledge base containing multiple nutrition management strategy information uploaded after evaluation. Each nutrition management strategy information is associated with corresponding applicable conditions, adjustment actions, and decision-making basis information. The knowledge base is regularly updated with the evaluated content.

[0038] In this embodiment of the invention, real-time monitoring data is merged and analyzed with a pre-generated critical care nutrition knowledge graph. The critical care nutrition knowledge graph is a structured knowledge base that stores information on multiple nutrition management strategies. Each strategy is associated with applicable clinical conditions, specific adjustment actions, and corresponding decision-making criteria. By matching the current patient status with the conditions in the knowledge graph, the original nutrition plan is dynamically adjusted, such as modifying intake, changing nutritional routes, or adjusting feeding sequences. Recommended information containing optimized plans and decision-making criteria is generated for medical staff to refer to.

[0039] Optionally, the optimization process may include calculating the total daily intake; if the deviation from the target amount is too large, adjusting the intake based on the knowledge graph; adjusting the pathway or timing according to the graph when specific complications occur; and comprehensively adjusting multiple parameters if the test indicators do not meet expectations. Furthermore, the knowledge graph can be updated regularly to incorporate the latest clinical guideline information.

[0040] The beneficial effects of this invention compared to existing technologies are as follows: By acquiring multi-source heterogeneous medical record data of patients, a target nutritional intake plan including nutritional routes, intake amounts, and intake timing is generated. During the execution of the plan, real-time nutritional data including enteral nutrition intake, parenteral nutrition intake, complication occurrence, and test indicators are monitored in real time. Based on a preset critical care nutrition knowledge graph, the target nutritional intake plan is dynamically optimized and nutritional recommendation information is generated, achieving comprehensive monitoring and accurate assessment of the nutritional status of ICU patients, overcoming the limitations of traditional technologies that only focus on enteral nutrition intake. By introducing a critical care nutrition knowledge graph for dynamic optimization, the nutritional recommendation strategy can be personalized according to the patient's real-time health profile, and each optimization is associated with corresponding decision-making information, effectively improving the completion rate of nutritional plans for ICU patients.

[0041] Example 2: In an optional specific embodiment of the present invention, refer to Figure 2 , Figure 2 This is a schematic diagram of Embodiment 2 of the method for recommending nutritional intake for ICU patients in this invention. Based on multi-source heterogeneous medical record data, a target nutritional intake plan for the patient is generated, which may specifically include steps S201 to S202.

[0042] Step S201: Calculate the patient's basal energy expenditure based on the height and weight in the multi-source heterogeneous medical record data, and determine the daily target calorie and target protein intake based on the disease course information and test indicators in the multi-source heterogeneous medical record data.

[0043] In an embodiment of the present invention, the patient's height and weight are extracted from the acquired multi-source heterogeneous medical record data, and the patient's energy and protein intake targets are calculated by calling a simple calculation formula based on body mass.

[0044] In an embodiment of the present invention, the target daily calorie intake required by the patient is calculated based on a calculation formula and the patient's disease progression information. This target calorie intake serves as a component of the overall intake in the subsequent nutrition plan.

[0045] Optionally, if the patient has a special metabolic condition, the calorie target can be adjusted using an additional correction factor, where special metabolic conditions include hyperthermia or burns.

[0046] In this embodiment of the invention, the daily target protein intake is calculated based on a calculation formula and patient disease information, according to the patient's weight and a preset protein requirement coefficient. The protein requirement coefficient can be dynamically adjusted according to the patient's condition. The protein requirement coefficient can be the number of grams of protein required per kilogram of body weight per day. The condition can be information such as the degree of catabolism and kidney function.

[0047] Optionally, the protein requirement coefficient can be obtained from clinical guidelines or customized based on the patient's test results.

[0048] Step S202: The daily target calorie intake and daily target protein intake are taken as intake amounts, and the nutritional pathways and intake sequences are determined based on the examination data and medication records in the multi-source heterogeneous medical record data to obtain the target nutritional intake plan.

[0049] In embodiments of this invention, calorie intake and protein intake are used as intake indicators. Further analysis of examination data and medication records in the medical record, combined with clinical rules, determines the appropriate nutritional route and intake sequence, ultimately forming a personalized target nutritional intake plan for the patient. Examination data may include albumin, globulin, blood glucose, and blood lipids. Medication records may include insulin use, propofol, and other energy-containing medications. Nutritional routes may include enteral nutrition, parenteral nutrition, or mixed feeding. Intake sequence may include continuous infusion, intermittent feeding, and initial infusion rate, where the initial infusion rate, expressed in milliliters per hour, is the initial infusion rate set for the nutrient solution or feeding when the intake sequence is determined.

[0050] Optionally, the choice of nutritional route can be optimized based on the results of gastrointestinal function assessment (such as intra-abdominal pressure and gastric residual volume); the timing of intake can be adjusted according to patient tolerance and daily care arrangements.

[0051] In this embodiment of the invention, individualized nutritional goals for ICU patients are achieved by using body weight and simple formula calculation methods, combined with disease information, examination data and medication records. This ensures the scientific and rational nature of calorie and protein intake, while effectively avoiding problems such as energy excess or deficiency and inappropriate route selection caused by empirical feeding.

[0052] Example 3: In an optional specific embodiment of the present invention, refer to Figure 3 , Figure 3 This is a schematic diagram of Embodiment 3 of the nutritional intake recommendation method for ICU patients in this invention. Based on real-time nutritional data and a preset critical care nutrition knowledge graph, the target nutritional intake plan is dynamically optimized, which may specifically include steps S301 to S304.

[0053] Step S301: Calculate the sum of the actual intake of enteral nutrition and the actual intake of parenteral nutrition to obtain the cumulative intake for the day.

[0054] In an embodiment of the present invention, during the execution of the target nutrition intake plan, the patient's actual enteral nutrition intake and actual parenteral nutrition intake are obtained in real time, and the two are added together to calculate the patient's total cumulative intake for the day.

[0055] Optionally, the actual intake of enteral nutrition can be obtained through automatic reporting by the feeding pump or manual entry; the actual intake of parenteral nutrition can be obtained through infusion pump data or medical order execution records. For solvents and sedatives containing energy substances, the system can convert their energy and include it in the calculation of parenteral nutrition intake.

[0056] Step S302: When the difference between the cumulative daily intake and the intake in the target nutrition intake plan is lower than or higher than a preset threshold, the intake is updated by matching the corresponding adjustment strategy according to the critical care nutrition knowledge graph.

[0057] In an embodiment of the present invention, the cumulative daily intake is compared with the intake specified in the target nutritional intake plan. When the difference between the two exceeds a preset threshold range, an optimization mechanism is triggered, and an appropriate adjustment strategy is matched according to the critical care nutrition knowledge graph, thereby updating the intake in the plan.

[0058] Optionally, the preset threshold can be dynamically adjusted according to the patient's condition, such as using a more lenient threshold during the acute phase and a stricter threshold during the stable phase. Adjustment strategies may include specific actions such as increasing the feeding rate, extending the feeding time, and supplementing with parenteral nutrition.

[0059] Step S303: When the occurrence of complications meets the preset conditions, match the first strategy information according to the critical care nutrition knowledge graph, and update the nutritional pathway and / or intake sequence according to the first strategy information.

[0060] In embodiments of the present invention, the occurrence of complications in patients is monitored in real time. When a complication is detected and meets preset triggering conditions, first strategy information is matched according to a critical care nutrition knowledge graph, and the nutritional pathway and / or intake sequence in the target nutrition intake plan are updated according to this strategy information (e.g., reducing the initial feeding rate, or restarting feeding after a pause). Updating the nutritional pathway in the target nutrition intake plan includes switching from enteral nutrition to partial parenteral nutrition. Updating the intake sequence in the target nutrition intake plan may include reducing the initial feeding rate or restarting feeding after a pause.

[0061] Optionally, the first strategy information may include a multi-level treatment plan for specific complications, allowing for different levels of adjustment based on the severity of the complication. For example, mild abdominal distension may require slowing the infusion rate, while severe abdominal distension may require suspending enteral nutrition and initiating parenteral nutrition.

[0062] Step S304: When the test indicators do not reach the preset expected target, match the second strategy information according to the critical care nutrition knowledge graph, and update the intake, nutritional route and / or intake sequence according to the second strategy information.

[0063] In an embodiment of the present invention, the patient's test indicators are monitored in real time. When a test indicator is detected to have failed to meet the preset target, a second strategy is matched based on a critical care nutrition knowledge graph, and the intake amount, nutritional route, and / or intake sequence in the target nutrition intake plan are comprehensively updated based on this strategy information.

[0064] Optionally, the second strategy information may include comprehensive intervention plans for delayed recovery, such as increasing protein intake, changing the type of nutritional preparation, and adjusting the feeding sequence to suit the patient's circadian rhythm. Multiple alternative plans can be output simultaneously for healthcare professionals to choose from.

[0065] In this embodiment of the invention, by setting intake deviation, complication occurrence, and testing indicators, and matching corresponding adjustment strategies based on a critical care nutrition knowledge graph, dynamic closed-loop optimization of nutritional support for ICU patients is achieved. This allows for timely correction of intake deviations, management of complications, and promotion of recovery based on the patient's real-time condition, effectively improving the effectiveness of nutritional therapy.

[0066] Example 4: In an optional embodiment of the present invention, if it is determined that the target nutrition intake plan is being implemented, the patient's real-time nutrition data is monitored, which may specifically include steps S401 to S403.

[0067] Step S401: If it is determined that the target nutrition intake plan is being implemented, the actual enteral nutrition intake and the actual parenteral nutrition intake of the patient are obtained in real time.

[0068] In embodiments of the present invention, during the execution of the target nutritional intake plan, the actual nutritional intake of the patient via enteral and parenteral routes is continuously collected. The actual enteral nutritional intake can be obtained automatically by connecting a feeding pump or manually entered by nursing staff; the actual parenteral nutritional intake can be obtained through infusion pump data, medical order execution records, or medication preparation information.

[0069] Optionally, for non-nutrition preparations containing energy substances, their energy components can be identified and converted into parenteral nutrition intake.

[0070] Step S402: Monitor the patient's complications in real time. Complications include at least one of the following: abdominal distension, diarrhea, refeeding syndrome, constipation, and tolerance score.

[0071] In embodiments of the present invention, during nutritional support, patients are monitored for complications related to nutritional therapy. The monitored complications include at least one or more of the following: abdominal distension, diarrhea, refeeding syndrome, and constipation; these are quantitatively assessed based on tolerance scores. Complication data can be obtained through multiple channels, including bedside monitoring devices, nursing records, and laboratory test results.

[0072] Optionally, trigger thresholds for complications can be set, such as intra-abdominal pressure exceeding a certain value being considered a risk of abdominal distension, and abnormal bowel movement frequency or characteristics being considered diarrhea, and a tolerance score can be calculated according to the scoring rules.

[0073] Step S403: Monitor the patient's test results in real time.

[0074] In embodiments of this invention, laboratory indicators reflecting a patient's nutritional status and disease recovery progress are monitored. These indicators may include nutrition-related indicators, disease severity scores, and functional status indicators. Data sources include electronic medical record systems, laboratory systems, and nursing records. Specifically, laboratory indicators may include albumin, globulin, and prealbumin; disease severity scores may include changes in APACHE II and SOFA scores; and functional status indicators may include muscle strength and progress on weaning from mechanical ventilation.

[0075] Optionally, indicator values ​​can be automatically captured according to a preset monitoring frequency, and trend changes can be recorded.

[0076] In this embodiment of the invention, by separately acquiring and calculating the actual intake of nutrients in the enteral and parenteral systems, and monitoring complications and test indicators, multi-dimensional real-time nutritional data perception is achieved. This reflects the patient's implementation of nutritional therapy, tolerance status, and therapeutic response, providing a precise data foundation for subsequent dynamic optimization based on knowledge graphs.

[0077] Example 5: In an optional embodiment of the present invention, generating nutritional recommendation information may specifically include steps S501 to S502.

[0078] Step S501: The dynamically optimized target nutrient intake plan is used as the recommended content.

[0079] In an embodiment of the present invention, after the dynamic optimization of the target nutrient intake plan is completed, the optimized target nutrient intake plan is used as the recommended content to be pushed. The optimized plan has been adjusted accordingly based on the patient's real-time nutritional data and critical care nutrition knowledge graph, including updated nutritional pathways, intake amounts, and intake sequences.

[0080] Optionally, multiple candidate optimization schemes can be generated, and the optimal scheme can be selected as the recommended content according to priority rules.

[0081] Step S502: Extract decision-making basis information corresponding to this optimization from the critical care nutrition knowledge graph.

[0082] In an embodiment of the present invention, the triggering conditions and matching process of the current optimization are traced back, and decision-making basis information corresponding to the current optimization is extracted from the critical care nutrition knowledge graph. The decision-making basis information includes the clinical conditions on which the optimization was triggered, the strategy content of the knowledge graph matching, and the source of guideline evidence, clinical research, or expert consensus supporting the strategy. The clinical conditions include specific complication types and the magnitude of intake deviation; the strategy content includes adjustment actions.

[0083] Optionally, the decision-making basis information can be presented as structured data or natural language description. The structured data may include guideline clause number and level of evidence, while the natural language description may be "According to the 202X Critical Care Nutrition Guidelines, enteral nutrition should be suspended and parenteral nutrition should be initiated for patients with intra-abdominal pressure >20 mmHg."

[0084] Step S502: Associate the recommended content with the decision-making basis information to generate nutritional recommendation information.

[0085] In embodiments of the present invention, recommended content is linked and integrated with decision-making basis information to generate nutritional recommendation information. The recommendation information may simultaneously include what to recommend and why it is recommended in this way.

[0086] Optionally, nutritional recommendations can be presented differently based on user roles, such as showing doctors detailed evidence sources and adjustment logic, and nurses specific implementation guidelines. The recommendations can also be pushed to clinical workstations, mobile devices, or printed out.

[0087] Optionally, the critical care nutrition knowledge graph can be updated periodically. By introducing a mechanism for periodically updating the critical care nutrition knowledge graph before the dynamic optimization step, the knowledge base can include the latest clinical guidelines, research evidence, and clinical experience. This enables the matching and optimization of nutritional strategies based on cutting-edge knowledge, effectively improving the timeliness of recommendation results.

[0088] In this embodiment of the invention, by linking the dynamically optimized nutrition plan with the corresponding decision-making information, medical staff can not only obtain personalized nutrition adjustment suggestions, but also simultaneously understand the clinical evidence behind the suggestions.

[0089] Example 6: In an optional embodiment of the present invention, obtaining multi-source heterogeneous medical record data of patients may specifically include steps S601 to S602.

[0090] Step S601: Collect the patient's initial multi-source heterogeneous medical record data from the electronic medical record system.

[0091] In an embodiment of the present invention, a data interface is established with an electronic medical record system to automatically acquire the patient's original multi-source heterogeneous medical record data. The original multi-source heterogeneous medical record data exists in its original format and includes various types such as basic physiological indicators, examination data, and medication records. The data is from scattered sources, has different structures, and has not been processed.

[0092] Optionally, the data collection method can be timed batch extraction or real-time event-triggered push; the collection scope can be filtered according to patient identification to obtain only the data of currently hospitalized ICU patients.

[0093] Step S602: Perform preprocessing on the initial case data to obtain multi-source heterogeneous medical record data. The preprocessing includes data cleaning, data anonymization, data standardization, and data association and integration.

[0094] In embodiments of the present invention, preprocessing may include data cleaning, data anonymization, data standardization, and data association and integration. Specifically, it involves identifying and correcting errors, duplicates, missing items, or inconsistencies in the data. Sensitive information involving patient privacy is anonymized or encrypted. Heterogeneous data from different systems is converted into a unified format, units, and encoding standard. Multi-source data is associated according to patient and time dimensions to obtain a complete, time-series view of patient data.

[0095] Optionally, the preprocessing process can dynamically adjust the cleaning rules according to the data quality; the degree of desensitization can be set in stages according to the application scenario.

[0096] In this embodiment of the invention, by acquiring multi-source heterogeneous medical record data of patients, a target nutritional intake plan including nutritional routes, intake amounts, and intake timing is generated. During the execution of the plan, real-time nutritional data including enteral nutrition intake, parenteral nutrition intake, occurrence of complications, and test indicators are monitored. The target nutritional intake plan is dynamically optimized based on a preset critical care nutrition knowledge graph, and nutritional recommendation information is generated. This achieves comprehensive monitoring and accurate assessment of the nutritional status of ICU patients, overcoming the limitations of traditional technologies that only focus on enteral nutrition intake. By introducing a critical care nutrition knowledge graph for dynamic optimization, the nutritional recommendation strategy can be personalized according to the patient's real-time health profile, and each optimization is associated with corresponding decision-making information, effectively improving the completion rate of nutritional plans for ICU patients.

[0097] Figure 4 This diagram illustrates a structural schematic of a nutritional intake recommendation device for ICU patients according to an embodiment of the present invention. The nutritional intake recommendation device 700 for ICU patients can be configured on an electronic device. Specifically, the nutritional intake recommendation device 700 for ICU patients may include: The acquisition module 701 is used to acquire the patient's multi-source heterogeneous medical record data, which includes basic physiological indicators, examination data and medication records. The generation module 702 is used to generate a target nutritional intake plan for patients based on multi-source heterogeneous medical record data. The target nutritional intake plan includes nutritional routes, intake amounts, and intake timing. The monitoring module 703 is used to detect the patient's real-time nutritional data if it is determined that the target nutritional intake plan is being implemented. The real-time nutritional data includes enteral nutrition intake, parenteral nutrition intake, occurrence of complications, and test indicators. The optimization module 704 is used to dynamically optimize the target nutrition intake plan based on real-time nutrition data and a preset critical care nutrition knowledge graph, and generate nutrition recommendation information.

[0098] The beneficial effects of the embodiments of the present invention compared with the prior art are: beneficial effects.

[0099] The generation module 702 can also be specifically used to: generate a patient's target nutritional intake plan based on multi-source heterogeneous medical record data. The target nutritional intake plan includes nutritional routes, intake amounts, and intake timing.

[0100] The optimization module 704 can also be specifically used for: calculating the sum of the actual intake of enteral nutrition and the actual intake of parenteral nutrition to obtain the cumulative intake for the day; when the difference between the cumulative intake for the day and the intake in the target nutrition intake plan exceeds a preset threshold, updating the intake by matching the corresponding adjustment strategy according to the critical care nutrition knowledge graph; when the occurrence of complications meets preset conditions, updating the nutritional route and / or intake sequence according to the first strategy information by matching the first strategy information; when the test indicators do not reach the preset expected target, updating the intake, nutritional route and / or intake sequence according to the second strategy information by matching the second strategy information.

[0101] The monitoring module 703 can also be specifically used for: if it is determined that the target nutrition intake plan is being implemented, to obtain the patient's actual enteral nutrition intake and actual parenteral nutrition intake in real time; to monitor the occurrence of complications in real time, including at least one of abdominal distension, diarrhea, refeeding syndrome, constipation and tolerance score; and to monitor the patient's laboratory indicators in real time.

[0102] The optimization module 704 can also be specifically used to: use the dynamically optimized target nutrient intake plan as recommended content; extract decision-making basis information corresponding to this optimization from the critical care nutrition knowledge graph; and associate the recommended content with the decision-making basis information to generate nutritional recommendation information.

[0103] Optimization module 704 can also be specifically used for: periodically updating the critical care nutrition knowledge graph.

[0104] The acquisition module 701 can also be specifically used for: collecting initial multi-source heterogeneous medical record data of patients from the electronic medical record system; performing preprocessing on the initial case data to obtain multi-source heterogeneous medical record data, wherein the preprocessing includes data cleaning, data desensitization, data standardization and data association integration.

[0105] like Figure 5 The diagram illustrates an electronic device according to an embodiment of the present invention. The electronic device 800 may include a processor 801, a memory 802, and a computer program 803 stored in the memory 802 and executable on the processor 801, such as a nutritional intake recommendation program for ICU patients. When the processor 801 executes the computer program 803, it implements the steps described in the various embodiments for nutritional intake recommendations for ICU patients.

[0106] A computer program can be divided into one or more modules / units. One or more modules / units are stored in memory 802 and executed by processor 801 to complete the present invention. One or more modules / units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in an electronic device.

[0107] The electronic device may include, but is not limited to, a processor 801 and a memory 802. Those skilled in the art will understand that... Figure 5 This is merely an example of an electronic device and does not constitute a limitation on the electronic device. It may include more or fewer components than shown, or combine certain components, or different components. For example, an electronic device may also include input / output devices, network access devices, buses, etc.

[0108] The processor 801 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0109] The memory 802 can be an internal storage unit of an electronic device, such as a hard drive or RAM. The memory 802 can also be an external storage device of the electronic device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 802 can include both internal and external storage units. The memory 802 is used to store computer programs and other programs and data required by the electronic device. The memory 802 can also be used to temporarily store data that has been output or will be output.

[0110] It should be noted that, for the sake of convenience and brevity, the structure of the above-mentioned electronic device can also be referred to the specific description of the structure in the method embodiment, which will not be repeated here.

[0111] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps in the above-described method for recommending nutritional intake for ICU patients.

[0112] This invention provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps in the above-described method for recommending nutritional intake for ICU patients.

[0113] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0114] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for various specific applications, but such implementations should not be considered beyond the scope of this invention.

[0115] In the embodiments provided by this invention, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the electronic device embodiments described above are merely illustrative. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0116] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0117] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0118] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0119] The embodiments described above are merely illustrative of 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. These modifications or substitutions do 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, and should all be included within the protection scope of the present invention.

Claims

1. A method for recommending nutritional intake for ICU patients, characterized in that, include: Acquire multi-source heterogeneous medical record data of patients, including basic physiological indicators, examination data and medication records; Based on the multi-source heterogeneous medical record data, a target nutritional intake plan is generated for the patient, which includes nutritional routes, intake amounts, and intake timing. If it is determined that the target nutrition intake plan is being implemented, the patient's real-time nutrition data is monitored. The real-time nutrition data includes enteral nutrition intake, parenteral nutrition intake, occurrence of complications, and test indicators, including albumin and prealbumin levels. Based on the real-time nutrition data and the preset critical care nutrition knowledge graph, the target nutrition intake plan is dynamically optimized, and nutrition recommendation information is generated. The critical care nutrition knowledge graph is a pre-built knowledge base containing multiple nutrition management strategy information uploaded after evaluation. Each nutrition management strategy information is associated with corresponding applicable conditions, adjustment actions, and decision-making basis information. The knowledge base is regularly updated with the evaluated content.

2. The method for recommending nutritional intake for ICU patients as described in claim 1, characterized in that, The step of generating the patient's target nutritional intake plan based on the multi-source heterogeneous medical record data includes: Based on the height and weight in the multi-source heterogeneous medical record data, the patient's basal energy expenditure is calculated, and the daily target calorie and target protein intake is determined based on the disease course information and test indicators in the multi-source heterogeneous medical record data. The target daily calorie intake and the target daily protein intake are used as the intake amounts. Based on the examination data and drug use records in the multi-source heterogeneous medical record data, the nutritional pathway and the intake sequence are determined to obtain the target nutritional intake plan.

3. The method for recommending nutritional intake for ICU patients as described in claim 1, characterized in that, The step of dynamically optimizing the target nutrition intake plan based on the real-time nutrition data and a preset critical care nutrition knowledge graph includes: Calculate the sum of the actual intake of enteral nutrition and the actual intake of parenteral nutrition to obtain the cumulative intake for the day; When the difference between the cumulative daily intake and the intake in the target nutrition intake plan is lower than or higher than a preset threshold, the intake is updated according to the corresponding adjustment strategy matched by the critical care nutrition knowledge graph. When the occurrence of the complication meets the preset conditions, the first strategy information is matched according to the critical care nutrition knowledge graph, and the nutritional pathway and / or the intake sequence is updated according to the first strategy information. When the test indicators fail to reach the preset expected target, the second strategy information is matched according to the critical care nutrition knowledge graph, and the intake, the nutritional route and / or the intake sequence are updated according to the second strategy information.

4. The method for recommending nutritional intake for ICU patients as described in claim 1, characterized in that, If it is determined that the target nutrition intake plan is being implemented, then the patient's real-time nutritional data is detected, including: If it is determined that the target nutrition intake plan is being implemented, the actual enteral nutrition intake and the actual parenteral nutrition intake of the patient are obtained in real time. The patient's complications are monitored in real time, including at least one of abdominal distension, diarrhea, refeeding syndrome, constipation, and tolerance score. The patient's test results are monitored in real time.

5. The method for recommending nutritional intake for ICU patients as described in claim 1, characterized in that, The generated nutrition recommendation information includes: The dynamically optimized target nutrient intake plan will be used as the recommended content. Extract decision-making basis information corresponding to this optimization from the critical care nutrition knowledge graph; The recommended content is associated with the decision-making basis information to generate the nutritional recommendation information.

6. The method for recommending nutritional intake for ICU patients as described in claim 5, characterized in that, Before the step of dynamically optimizing the target nutrient intake plan based on the real-time nutrient data and a preset critical care nutrition knowledge graph, the method further includes: The critical care nutrition knowledge graph is updated periodically.

7. The method for recommending nutritional intake for ICU patients as described in claim 1, characterized in that, The acquisition of patients' multi-source heterogeneous medical record data includes: The patient's initial multi-source heterogeneous medical record data were collected from the electronic medical record system; Preprocessing is performed on the initial case data to obtain the multi-source heterogeneous medical record data, wherein the preprocessing includes data cleaning, data anonymization, data standardization, and data association and integration.

8. A nutritional intake recommendation device for ICU patients, characterized in that, The device includes: The acquisition module acquires multi-source heterogeneous medical record data of patients, including basic physiological indicators, examination data and medication records; The generation module generates a target nutritional intake plan for the patient based on the multi-source heterogeneous medical record data. The target nutritional intake plan includes nutritional routes, intake amounts, and intake timing. The monitoring module, if it determines that the target nutrition intake plan is being implemented, monitors the patient's real-time nutrition data, including enteral nutrition intake, parenteral nutrition intake, occurrence of complications, and test indicators, including albumin and prealbumin levels. The optimization module dynamically optimizes the target nutrition intake plan based on the real-time nutrition data and the preset critical care nutrition knowledge graph, and generates nutrition recommendation information. The critical care nutrition knowledge graph is a pre-built knowledge base containing multiple nutrition management strategy information uploaded after evaluation. Each nutrition management strategy information is associated with corresponding applicable conditions, adjustment actions, and decision-making basis information. The knowledge base is regularly updated with the evaluated content.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for recommending nutritional intake for ICU patients as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the nutritional intake recommendation method for ICU patients as described in any one of claims 1 to 7.

Images