A monitoring and early warning system and method for enteral nutrition interruption in critical patients

Abstract

The application discloses a kind of monitoring and early warning methods of severe patient enteral nutrition interruption, comprising: S1: obtaining patient information data, nutrition pump working data and nutrition pump interruption data;S2: according to nutrition pump interruption data, judge whether interruption event occurs and calibrate nutrition pump interruption data;S3: analyze the cause of interruption event and predict future risk;S4: provide compensatory program to interruption event, provide preventive program to future risk;Advantages are that through data automatic calibration and multidimensional intelligent analysis, fine monitoring and early intervention to enteral nutrition interruption are realized, and the intelligence of enteral nutrition interruption management is improved.

Description

Technical Field

[0001] This invention relates to the field of information processing technology, and in particular to a monitoring and early warning system and method for enteral nutrition interruption in critically ill patients. Background Technology

[0002] Enteral nutrition (EN) is the preferred method of nutritional support for critically ill patients. However, in the complex intensive care environment, enteral nutrition interruption (EFI) is extremely common, with research data showing an incidence rate as high as 12.8% to 68.0%. Enteral nutrition interruption not only directly affects the rate of achieving nutritional goals, but also significantly increases the risk of malnutrition, related complications, and even mortality. Traditional management methods rely heavily on manual records and experience-based judgment by medical staff, resulting in problems such as incomplete records, delayed feedback, and difficulty in quantification, making it impossible to achieve refined and intelligent management of the enteral nutrition process.

[0003] Chinese invention patent application CN118824557A discloses a method for managing enteral nutrition in critically ill patients, including: S1: real-time monitoring and acquisition of the patient's basic data, calculation of nutritional parameters based on the basic data, and fusion into a standard dataset; S2: calculation and assessment of the patient's enteral nutritional status based on the standard dataset; S3: outputting early warning information and corresponding treatment plans based on the enteral nutritional status. Although this invention solves the problem of real-time monitoring of the patient's enteral nutritional status, it lacks a data calibration mechanism in terms of refinement and lacks forward-looking prediction of future interruption risks in terms of intelligence. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a monitoring and early warning system and method for enteral nutrition interruption in critically ill patients. Through automatic data calibration and multi-dimensional intelligent analysis, the system enables refined monitoring and proactive intervention of enteral nutrition interruption, thereby improving the intelligence of enteral nutrition interruption management.

[0005] The technical solution adopted by this invention to solve the above-mentioned technical problem is: a method for monitoring and early warning of enteral nutrition interruption in critically ill patients, characterized in that it includes: S1: Acquire patient information data, nutrition pump operation data, and nutrition pump interruption data; S2: Determine whether an interruption event has occurred based on the nutrient pump interruption data, and calibrate the nutrient pump interruption data; S3: Analyze the causes of the interruption event and predict future risks; S4: Provide compensatory solutions for disruption events and preventative solutions for future risks.

[0006] Furthermore, the patient information data in step S1 includes basic patient information, medical orders, laboratory test results from the hospital information system and / or hospital laboratory information system, and nursing records, KSC diarrhea score, and feeding tolerance score from the nursing information system; the nutrition pump operation data in step S1 includes the motor operating time of the nutrition pump, the cumulative infusion volume of the nutrition pump, the start time of the nutrition pump, and the shutdown time of the nutrition pump; the nutrition pump interruption data in step S1 includes the start time and end time of the nutrition pump pause.

[0007] Furthermore, the method for determining whether an interruption event has occurred in step S2 is as follows: if the nutrient pump pauses for more than 10 minutes, an immediate alarm is issued; if the nutrient pump pauses for 15 minutes or more, it is determined that an interruption event has occurred.

[0008] Furthermore, the calibration method for the nutrient pump interruption data in step S2 is as follows: first calculate the interruption error, and then perform hierarchical calibration on the interruption error; The interruption error is calculated as follows: the actual interruption time is obtained by comparing the start-up time of the nutrient pump, the shutdown time of the nutrient pump, and the working time of the motor of the nutrient pump; the interruption error is obtained by comparing the actual interruption time with the sum of all interruption data. The hierarchical calibration method is as follows: If only one pause occurs between the start-up and shutdown times of the nutrient pump, all interruption errors are included in this pause to obtain the calibration time. Finally, the calibration time is used to determine whether this pause is an interruption event. If multiple pauses occur between the start-up and shutdown times of the nutrient pump, the interruption data in the 15-minute critical pause time is corrected first. When the interruption error is greater than 0, the increment required to calibrate to exactly 15 minutes is calculated for each interruption data in the critical pause time. Then, the interruption data in the critical pause time is compensated sequentially according to the absolute value of the increment from smallest to largest. When the interruption error is less than 0, the decrement required to calibrate to exactly less than 15 minutes is calculated for each interruption data in the critical pause time. Then, the interruption data in the critical pause time is compensated sequentially according to the absolute value of the decrement from smallest to largest. If there is still a remaining interruption error after correcting all the interruption data in the critical pause time, the remaining interruption error is distributed to the uncorrected interruption data according to the weight ratio.

[0009] Furthermore, the method for analyzing the cause of the interruption event in step S3 includes intelligent analysis and manual calibration. Intelligent analysis involves determining the cause of the interruption based on patient information data within the interruption period, and then modifying, supplementing, and confirming the cause determined by intelligent analysis through manual calibration.

[0010] Furthermore, the method for predicting future risks in step S3 is as follows: patient information data and nutrition pump operation data are input into a pre-trained learning model, the learning model outputs the probability of enteral nutrition interruption events occurring in the future, and the risk is divided into three levels: "low-medium-high" according to the probability value.

[0011] Furthermore, the compensatory scheme for the interruption event in step S4 is: to calculate the nutritional deficit and recommend a subsequent nutritional supply plan based on the cause of the interruption event and the nutritional pump interruption data. The preventive scheme for future risks in step S4 is: a nutritional supply plan given by combining the risk level predicted by the learning model and the patient information data.

[0012] A monitoring and early warning system for enteral nutrition interruption in critically ill patients includes: The data acquisition module is used to collect patient information data, nutrition pump operation data, and nutrition pump interruption data. A judgment module, which is connected to the data acquisition module, is used to determine whether an interruption event has occurred based on the nutrient pump interruption data. The calibration module is connected to both the data acquisition module and the judgment module. The calibration module is used to calculate the interruption error and calibrate the nutrient pump interruption data. An analysis module, connected to the calibration module, is used to analyze the cause of the interruption event. The analysis module determines the cause of the interruption based on patient information data within the interruption period. A human-computer interaction module is connected to the analysis module. The human-computer interaction module is used to display the interruption reason output by the analysis module to medical staff and receive modification, supplementation or confirmation instructions from medical staff on the interruption reason. A prediction module, which is connected to the data acquisition module, is used to predict future risks and classify risk levels. The result generation module is connected to both the human-computer interaction module and the prediction module. The result generation module includes a first generation module and a second generation module. The first generation module is used to output compensatory solutions for interruption events that have occurred, and the second generation module is used to output preventive solutions for predicted risks in the future.

[0013] Furthermore, the judgment module is connected to an alarm module, which is used to issue an alarm prompt.

[0014] Compared with existing technologies, the advantages of this invention are that by using nutrition pump operating data to calculate the actual feeding volume, the interruption data of the nutrition pump can be cross-validated and calibrated, overcoming the problem of raw data deviation caused by device communication delays and signal false alarms, improving the accuracy and reliability of the entire data source, and achieving a more refined monitoring effect. In addition, this invention can not only accurately monitor and record interruption events, but also integrate patient information data and nutrition pump operating data through a learning model, realizing the prospective prediction of future interruption risks and improving the intelligence of treatment. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the system framework of the present invention. Detailed Implementation

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0017] like Figures 1 to 2 As shown, a method for monitoring and early warning of enteral nutrition interruption in critically ill patients includes: S1: Acquire patient information data, feeding pump operation data, and feeding pump interruption data. Patient information data includes basic patient information, medical orders, and laboratory test results from the hospital information system and / or hospital laboratory information system, as well as nursing records, KSC diarrhea scores, and feeding tolerance scores from the nursing information system. Feeding pump operation data includes the motor operating time of the feeding pump, the cumulative infusion volume of the feeding pump, the start time of the feeding pump, and the shutdown time of the feeding pump. Feeding pump interruption data includes the start and end times of feeding pump interruption. S2: Determine whether an interruption event has occurred based on the nutrient pump interruption data and calibrate the nutrient pump interruption data; when the nutrient pump changes from "running" state to "paused" state, record it as the "interruption start time"; when the nutrient pump changes from "paused" state back to "running" state, record it as the "interruption end time" and calculate the pause duration. If the pause time of the nutrient pump exceeds 10 minutes, issue an immediate alarm prompt; if the pause time of the nutrient pump is 15 minutes or more, it is determined that an interruption event has occurred. The calibration method for nutrient pump interruption data is as follows: first calculate the interruption error, and then perform stratified calibration on the interruption error; The interruption error is calculated as follows: the actual interruption time is obtained by comparing the start-up time, shutdown time, and motor operating time of the nutrient pump ((shutdown time - start-up time) - motor operating time of the nutrient pump = actual interruption time). The actual interruption time is then compared with the sum of all interruption data to obtain the interruption error (actual interruption time - sum of all interruption data = interruption error). The tiered calibration method is as follows: If only one pause occurs between the start-up and shutdown times of the nutrition pump, all interruption errors are included in this pause to obtain the calibration time. Finally, the calibration time is used to determine whether this pause constitutes an interruption event. If multiple pauses occur between the start-up and shutdown times of the nutrition pump, priority is given to correcting interruption data within the 15-minute critical zone (15min + 2s). When the interruption error is greater than 0 (e.g., the nutrition pump crashes due to a system error at 09:00 and is not turned on by the nurse until 09:05, during which time the motor is not working, but the "pause start time" and "pause end time" are not recorded due to the system error, in which case the interruption error is greater than 0), the increment required to calibrate to exactly 15 minutes is calculated for each interruption data in the critical zone. Then, the interruptions in the critical zone are corrected sequentially according to the absolute value of the increment from smallest to largest. The data is compensated. When the interruption error is less than 0 (e.g., if a medical staff member presses the pause button at 10:00, but the nutrition pump motor actually stops completely at 10:00:10 due to inertia or system delay, and the motor responds immediately when the nurse presses the start button at 10:11; the system records the pause time as 10:00-10:11 (recorded as 11 minutes); the actual motor stop time is 10:00:10-10:11 (the actual time without motor operation is 10 minutes and 50 seconds), and the interruption error is less than 0), the reduction required to calibrate to just below 15 minutes is first calculated for each interruption data in the critical zone. Then, the interruption data in the critical zone is compensated sequentially according to the absolute value of the reduction from small to large. If there is still a remaining interruption error after correcting all the interruption data in the critical zone, the remaining interruption error is distributed to the uncorrected interruption data according to the weight ratio. S3: Analyze the causes of interruption events and predict future risks; the analysis methods for the causes of interruption events include intelligent analysis and manual calibration. Intelligent analysis determines the cause of the interruption based on patient information data during the interruption period. For example, if the nursing records captured before and after the interruption event show a "tolerance score" > 5 points or a KSC diarrhea score of C, the cause of the interruption is initially identified as "feeding intolerance interruption". If the time of treatment orders such as "CT scan" coincides with the time of the interruption event, the cause of the interruption is initially identified as "examination and testing interruption". Then, manual calibration modifies, supplements and confirms the cause identified by intelligent analysis to ensure the final accuracy of the cause record. The method for predicting future risks is to input patient information data and nutrition pump operation data into a pre-trained learning model (such as LSTM, XGBoost, etc.). The learning model outputs the probability of an enteral nutrition interruption event occurring within a future period (6 hours) and classifies the risk into three levels: "low-medium-high" based on the probability value. S4: Provide compensatory solutions for interruption events and preventative solutions for future risks. Compensatory solutions for interruption events calculate the nutritional deficit and recommend subsequent nutritional supply plans based on the cause of the interruption and nutritional pump interruption data. For example, for "examination and testing interruption" lasting 3 hours, calculate the nutritional deficit and recommend appropriately increasing the infusion rate in the following 24 hours to complete compensatory feeding. Preventative solutions for future risks combine the risk level predicted by the learning model with the nutritional supply plan given by the patient information data. For example, if the risk prediction is "high risk" and the keyword captured by the learning model is "persistently elevated KSC diarrhea score", it will push a suggestion to "suggest switching to short peptide or dietary fiber-containing nutritional solutions and closely monitoring intake and output" for proactive intervention.

[0018] A monitoring and early warning system for enteral nutrition interruption in critically ill patients includes: The data acquisition module is used to collect patient information data, nutrition pump operation data, and nutrition pump interruption data. The judgment module is connected to the data acquisition module. The judgment module is used to determine whether an interruption event has occurred based on the nutrient pump interruption data. The judgment module is also connected to an alarm module, which is used to issue an alarm prompt. The calibration module is connected to both the data acquisition module and the judgment module. The calibration module is used to calculate the interruption error and calibrate the nutrient pump interruption data. An analysis module, connected to the calibration module, is used to analyze the cause of the interruption event. The analysis module determines the cause of the interruption based on patient information data within the interruption period. A human-computer interaction module is connected to the analysis module. The human-computer interaction module is used to display the interruption reason output by the analysis module to medical staff and receive modification, supplementation or confirmation instructions from medical staff on the interruption reason. A prediction module, which is connected to the data acquisition module, is used to predict future risks and classify risk levels. The result generation module is connected to both the human-computer interaction module and the prediction module. The result generation module includes a first generation module and a second generation module. The first generation module is used to output compensatory solutions for interruption events that have occurred, and the second generation module is used to output preventive solutions for predicted risks in the future.

[0019] The scope of protection of this invention includes, but is not limited to, the above embodiments. The scope of protection of this invention is defined by the claims. Any substitutions, modifications, or improvements to this technology that are easily conceived by those skilled in the art fall within the scope of protection of this invention.

Claims

1. A method for monitoring and early warning of an interruption of enteral nutrition in a critically ill patient, characterized in that, include: Step S1: Obtain patient information data, nutrition pump operation data, and nutrition pump interruption data; Step S2: Determine whether an interruption event has occurred based on the nutrient pump interruption data, and calibrate the nutrient pump interruption data; Step S3: Analyze the causes of the interruption event and predict future risks; Step S4: Provide compensatory solutions for interruption events and preventative solutions for future risks.

2. The monitoring and early warning method for enteral nutrition interruption in critically ill patients as described in claim 1, characterized in that, The patient information data in step S1 includes basic patient information, medical orders, laboratory test results from the hospital information system and / or hospital laboratory information system, and nursing records, KSC diarrhea score, and feeding tolerance score from the nursing information system; the nutrition pump operation data in step S1 includes the motor operating time of the nutrition pump, the cumulative infusion volume of the nutrition pump, the start time of the nutrition pump, and the shutdown time of the nutrition pump; the nutrition pump interruption data in step S1 includes the start time and end time of the nutrition pump pause.

3. The monitoring and early warning method for enteral nutrition interruption in critically ill patients as described in claim 2, characterized in that, The method for determining whether an interruption event has occurred in step S2 is as follows: if the nutrient pump pauses for more than 10 minutes, an immediate alarm is issued; if the nutrient pump pauses for 15 minutes or more, it is determined that an interruption event has occurred.

4. The monitoring and early warning method for enteral nutrition interruption in critically ill patients as described in claim 3, characterized in that, The calibration method for the nutrient pump interruption data in step S2 is as follows: first calculate the interruption error, and then perform stratified calibration on the interruption error; The interruption error is calculated as follows: the actual interruption time is obtained by comparing the start-up time of the nutrient pump, the shutdown time of the nutrient pump, and the working time of the motor of the nutrient pump; the interruption error is obtained by comparing the actual interruption time with the sum of all interruption data. The hierarchical calibration method is as follows: If only one pause occurs between the start-up and shutdown times of the nutrient pump, all interruption errors are included in this pause to obtain the calibration time. Finally, the calibration time is used to determine whether this pause is an interruption event. If multiple pauses occur between the start-up and shutdown times of the nutrient pump, the interruption data in the 15-minute critical pause time is corrected first. When the interruption error is greater than 0, the increment required to calibrate to exactly 15 minutes is calculated for each interruption data in the critical pause time. Then, the interruption data in the critical pause time is compensated sequentially according to the absolute value of the increment from smallest to largest. When the interruption error is less than 0, the decrement required to calibrate to exactly less than 15 minutes is calculated for each interruption data in the critical pause time. Then, the interruption data in the critical pause time is compensated sequentially according to the absolute value of the decrement from smallest to largest. If there is still a remaining interruption error after correcting all the interruption data in the critical pause time, the remaining interruption error is distributed to the uncorrected interruption data according to the weight ratio.

5. The monitoring and early warning method for enteral nutrition interruption in critically ill patients as described in claim 1, characterized in that, The method for analyzing the cause of the interruption event in step S3 includes intelligent analysis and manual calibration. Intelligent analysis determines the cause of the interruption based on patient information data within the interruption period, and then manual calibration modifies, supplements and confirms the cause identified by intelligent analysis.

6. The monitoring and early warning method for enteral nutrition interruption in critically ill patients as described in claim 1, characterized in that, The method for predicting future risks in step S3 is as follows: patient information data and nutrition pump operation data are input into a pre-trained learning model. The learning model outputs the probability of enteral nutrition interruption events occurring in the future, and the risk is divided into three levels: "low-medium-high" based on the probability value.

7. The monitoring and early warning method for enteral nutrition interruption in critically ill patients as described in claim 6, characterized in that, The compensatory scheme for the interruption event in step S4 is: to calculate the nutritional deficit and recommend a subsequent nutritional supply plan based on the cause of the interruption event and the nutritional pump interruption data. The preventive scheme for future risks in step S4 is: a nutritional supply plan given by combining the risk level predicted by the learning model and patient information data.

8. A monitoring and early warning system for enteral nutrition interruption in critically ill patients, characterized in that, include: The data acquisition module is used to collect patient information data, nutrition pump operation data, and nutrition pump interruption data. A judgment module, which is connected to the data acquisition module, is used to determine whether an interruption event has occurred based on the nutrient pump interruption data. The calibration module is connected to both the data acquisition module and the judgment module. The calibration module is used to calculate the interruption error and calibrate the nutrient pump interruption data. An analysis module, connected to the calibration module, is used to analyze the cause of the interruption event. The analysis module determines the cause of the interruption based on patient information data within the interruption period. A human-computer interaction module is connected to the analysis module. The human-computer interaction module is used to display the interruption reason output by the analysis module to medical staff and receive modification, supplementation or confirmation instructions from medical staff on the interruption reason. A prediction module, which is connected to the data acquisition module, is used to predict future risks and classify risk levels. The result generation module is connected to both the human-computer interaction module and the prediction module. The result generation module includes a first generation module and a second generation module. The first generation module is used to output compensatory solutions for interruption events that have occurred, and the second generation module is used to output preventive solutions for predicted risks in the future.

9. The monitoring and early warning system for enteral nutrition interruption in critically ill patients as described in claim 8, characterized in that, The judgment module is connected to an alarm module, which is used to issue an alarm prompt.

Images