Postoperative drainage device for clinical nursing and monitoring system

Abstract

The invention discloses a postoperative drainage device for clinical nursing and a monitoring system, and belongs to the technical field of biomedical engineering. The problems that an existing device manually monitors drainage liquid data, consequently, the monitoring precision is insufficient, and the complication probability cannot be predicted are solved, drainage liquid and patient vital sign data are collected in real time through the micro spectrograph, the character multi-parameter sensor, the micro pressure sensor and the intelligent bracelet and transmitted to the remote monitoring platform, and the monitoring accuracy is improved. Medical staff can check real-time data of a patient at any time and adjust a treatment scheme in time, so that the nursing efficiency and quality are improved; multi-parameter data such as color, metering and property of drainage liquid and vital sign data are analyzed through a risk prediction model, intervention measures are dynamically adjusted according to different complication types and probabilities in combination with patient condition changes and real-time monitoring data, accurate intervention on the complication is achieved, and the occurrence rate and severity of the complication are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of biomedical engineering, in particular to a postoperative drainage device and monitoring system for clinical nursing. BACKGROUND

[0002] Postoperative wound surface is prone to accumulate blood, tissue fluid and other liquids, which will lead to wound infection, delayed healing and even abscess if not discharged in time. The negative pressure drainage device can discharge the liquid outside in time through continuous negative pressure suction, keep the wound dry and promote wound healing. In clinical nursing, the postoperative drainage device is an important part of postoperative nursing. However, the traditional drainage device has the following defects: 1. The traditional drainage device cannot monitor the color, property and quantity of the drainage liquid in real time, and the medical staff need to manually check the quantity and color of the drainage liquid at regular intervals, which is not only time-consuming and laborious, but also prone to errors due to subjective judgment.

[0003] 2. The traditional drainage device cannot analyze the drainage liquid data and predict the risk of postoperative complications, so it is difficult for medical staff to take intervention measures in advance.

[0004] Therefore, the existing needs are not met, and for this purpose, we propose a postoperative drainage device and monitoring system for clinical nursing. SUMMARY

[0005] The purpose of the present application is to provide a postoperative drainage device and monitoring system for clinical nursing, which can collect the data of drainage liquid and patient vital signs in real time through a miniature spectrometer, a property multi-parameter sensor, a miniature pressure sensor and a smart bracelet, and transmit them to a remote monitoring platform. Medical staff can view the real-time data of patients at any time, adjust the treatment plan in time, and improve the nursing efficiency and quality. The risk prediction model can analyze the multi-parameter data of the color, quantity and property of the drainage liquid and the vital sign data, and dynamically adjust the intervention measures according to different types and probabilities of complications, combined with the changes of patient's condition and real-time monitoring data, to realize accurate intervention on complications, reduce the incidence and severity of complications, and solve the problems raised in the above background technology.

[0006] To achieve the above purpose, the present application provides the following technical scheme: The postoperative drainage device for clinical nursing comprises a negative pressure drainage device, which is composed of a drainage bottle, a connecting pipe, an adapter and a drainage tube. The top of the drainage bottle is provided with a negative pressure pump for providing negative pressure for the drainage bottle. The top end of the inside of the drainage bottle is provided with a miniature spectrometer for monitoring the color of the drainage liquid and a property multi-parameter sensor for monitoring the property of the drainage liquid. The bottom of the drainage bottle is provided with a miniature pressure sensor for monitoring the quantity of the drainage liquid. The front of the drainage bottle is provided with a radio frequency business card for storing patient information.

[0007] Further, the micro spectrometer, the character multi-parameter sensor and the micro pressure sensor are connected with the remote monitoring platform through wireless transmission technology respectively, and are used for feeding back the color data, the measurement data and the character multi-parameter of the drainage liquid monitored in real time to the remote monitoring platform.

[0008] Further, a connecting pipe is inserted into the top of the drainage bottle, one side of the connecting pipe is connected with a drainage tube through an adapter, one end of the drainage tube is in contact with the postoperative wound, a hanging ring is further arranged on the top of the drainage bottle, and a scale line is arranged on the bottle body of the drainage bottle.

[0009] The postoperative drainage monitoring system for clinical nursing comprises a data acquisition unit, a wireless transmission unit, a cloud big data platform and a remote monitoring platform. The data acquisition unit is composed of a micro spectrometer, a character multi-parameter sensor, a micro pressure sensor and a smart bracelet, and is configured to monitor the color data, the measurement data and the character multi-parameter of the drainage liquid in real time based on the micro spectrometer, the character multi-parameter sensor and the micro pressure sensor, and to monitor the vital sign data of the patient after the operation in real time based on the smart bracelet, including the heart rate, the blood pressure and the body temperature. The wireless transmission unit is configured to transmit the color data, the measurement data and the character multi-parameter of the drainage liquid, and the vital sign data to the cloud big data platform for intelligent analysis in real time based on the wireless transmission technology, to synchronously feed back the analysis result to the remote monitoring platform, and to realize the data interaction of the drainage monitoring system. The cloud big data platform is configured to build a risk prediction model, to extract the feature engineering in the color data, the measurement data and the character multi-parameter of the drainage liquid in the data through the risk prediction model, to predict the probability of the patient's postoperative complications, to develop a dynamic intervention scheme based on different complication types and probabilities, to generate and dynamically optimize the individualized intervention scheme in real time by building a monitoring change curve, identifying abnormal fluctuations, and combining the key degree of the patient data with the historical similar cases, and to set the risk probability threshold based on the clinical experience as the triggering intervention condition. The remote monitoring platform is configured to display the drainage liquid data and the vital sign data of the patient, and the risk prediction result and the corresponding intervention scheme in real time.

[0010] Further, the cloud big data platform comprises: The feature extraction module is configured to extract the features from the color data, the measurement data and the character multi-parameter of the drainage liquid. The feature calculation module is configured to build a random forest model as the risk prediction model, to train the model through the historical data, to evaluate the performance of the risk prediction model using cross-validation, and to calculate the AUC value of each feature engineering in the risk prediction model to evaluate the prediction ability of the risk prediction model to the complications.

[0011] Further, the cloud big data platform further comprises: a threshold setting module configured to set different thresholds according to the severity and clinical needs of different types of complications; a threshold triggering module configured to set a default risk probability threshold according to clinical experience and model performance, and trigger an intervention condition when the complication risk probability predicted by the model is greater than or equal to the risk probability threshold.

[0012] Further, the threshold setting module comprises: a first threshold setting unit configured to determine a first threshold of the patient based on each type of complication and the severity of the complication, wherein the severity is determined according to the symptom characteristics of the complication; a second threshold setting unit configured to determine a second threshold of the patient based on the clinical needs of the patient, wherein the clinical needs are determined according to the individual difference data and clinical sign data of the patient; a threshold comprehensive setting unit configured to determine a risk probability threshold of the patient based on the real-time patient needs combined with the first threshold and the second threshold.

[0013] Further, the cloud big data platform further comprises: a model prediction module configured to use the trained risk prediction model to predict the real-time drainage fluid data and vital sign data to obtain the complication risk probability at each time point, and compare the predicted risk probability with the preset threshold; if the risk probability is higher than the threshold, it is judged that the patient may have the corresponding complication; a risk intervention module configured to develop a corresponding intervention scheme according to the judged complication type and probability; a dynamic adjustment module configured to dynamically adjust the intervention scheme according to the patient's condition changes and real-time monitoring data.

[0014] Further, the dynamic adjustment module comprises: a curve construction unit configured to obtain real-time monitoring data of the patient and historical monitoring data within a preset time period, and perform data classification, construct a monitoring change curve of each data type of monitoring data based on the data classification result, and obtain a monitoring change curve group of the patient; a curve extraction unit configured to extract an abnormal fluctuation curve in the monitoring change curve group as a first monitoring change curve group; a monitoring processing unit configured to extract the monitoring data type corresponding to each monitoring change curve in the first monitoring change curve group, and based on the data key degree of the corresponding monitoring data of the monitoring data type, the data key degree is determined based on the influence degree of each monitoring data on the type and probability of the patient's complication; The data sorting unit is configured to sort the monitoring data based on the data criticality, and determine a monitoring priority queue based on the data sorting result; The first scheme determination unit is configured to determine a patient intervention scheme corresponding to each monitoring data in the monitoring priority queue, and obtain a first adjusted intervention scheme of the patient; The calibration comparison unit is configured to extract a historical patient with the highest similarity to the patient's physical sign data and real-time condition from the patient database as a calibration patient; The intervention sorting unit is configured to compare the historical calibration intervention scheme of the calibration patient with the first adjusted intervention scheme; The intervention sorting unit is configured to compare the historical calibration intervention scheme of the calibration patient with the first adjusted intervention scheme; The intervention sorting unit is configured to compare the historical calibration intervention scheme of the calibration patient with the first adjusted intervention scheme; The second scheme determination unit is configured to combine the intervention priority queue with real-time feedback and real-time intervention demand to obtain a second adjusted intervention scheme of the patient; The intervention adjustment unit is configured to dynamically adjust the intervention scheme based on the first adjusted intervention scheme and the second adjusted intervention scheme.

[0015] Further, the cloud big data platform further comprises: The kind distinguishing module is configured to determine the kind of possible complications of the patient according to the model prediction result, and record the risk probability of each kind of complication to provide a basis for subsequent intervention; The label setting module is configured to set a complication label based on historical data, and use the complication label as an input set of the risk prediction model; The quality control module is configured to supplement missing values of historical data by using multiple imputation, analyze abnormal data according to clinical knowledge, and determine whether the abnormal data need to be corrected or deleted; and perform time alignment processing on all data based on the end time of the operation as a base point.

[0016] Compared with the prior art, the present application has the following advantages: 1. In the present application, the micro spectrometer, the character multi-parameter sensor, the micro pressure sensor and the smart bracelet in the drainage device can collect the data of the drainage fluid and the vital signs of the patient in real time, and transmit the data to the remote monitoring platform through wireless transmission technology. Medical staff can check the real-time data and risk prediction results of the patient through the platform at any time and any place, and timely adjust the treatment scheme, thereby improving the nursing efficiency and quality.

[0017] 2.In the present application, the color, quantity, properties and other multi-parameter data of the drainage fluid and vital sign data are analyzed by the risk prediction model, which can predict the probability of postoperative complications in advance, provide a scientific basis for clinical decision-making, and help medical staff better understand the patient's condition; and according to different types and probabilities of complications, combined with patient condition changes and real-time monitoring data, dynamically adjust intervention measures to realize personalized nursing of patients, improve the pertinence and effectiveness of treatment, so as to realize precise intervention on complications, reduce the incidence and severity of complications, and improve the postoperative rehabilitation quality of patients; at the same time, the system combines big data, artificial intelligence technology and clinical nursing, promotes the informatization and intelligent development of medical care field, and provides a new idea and direction for innovation in biomedical engineering field. BRIEF DESCRIPTION OF DRAWINGS

[0018] Fig. 1 is the application diagram of the postoperative drainage device for clinical nursing of the present application; Fig. 2 is a structural diagram of the postoperative drainage device for clinical nursing of the present application; Fig. 3 is a flow chart of the postoperative drainage monitoring system for clinical nursing of the present application.

[0019] In the figure: 1, negative pressure drainage device; 2, drainage bottle; 21, hanging ring; 22, scale line; 3, connecting pipe; 4, adapter; 5, drainage tube; 6, negative pressure pump; 7, miniature spectrometer; 8, multi-parameter sensor; 9, miniature pressure sensor; 10, radio frequency business card; 11, smart bracelet. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0021] In order to solve the technical problems in the prior art that the traditional drainage device cannot monitor the color, properties and quantity of the drainage fluid in real time, medical staff need to manually check the quantity and color of the drainage fluid at regular intervals, which is not only time-consuming and laborious, but also prone to errors due to subjective judgment; and the drainage fluid data cannot be analyzed and the risk of postoperative complications cannot be predicted, which makes it difficult for medical staff to take intervention measures in advance, please refer to Figs. 1-3 The present embodiment provides the following technical solutions: Please refer to Figs. 1-2: The postoperative drainage device for clinical nursing comprises a negative pressure drainage device 1, wherein the negative pressure drainage device 1 is composed of a drainage bottle 2, a connecting pipe 3, an adapter 4 and a drainage pipe 5; the connecting pipe 3 is inserted into the top of the drainage bottle 2; one side of the connecting pipe 3 is connected with the drainage pipe 5 through the adapter 4; one end of the drainage pipe 5 is in contact with a postoperative wound surface; the postoperative wound surface is also in contact with a cleaning liquid equipment A; a hanging ring 21 is arranged on the top of the drainage bottle 2; a scale line 22 is arranged on the bottle body of the drainage bottle 2; a negative pressure pump 6 for providing negative pressure for the drainage bottle 2 is arranged on the top of the drainage bottle 2; a micro spectrometer 7 for monitoring the color of drainage liquid and a multi-parameter sensor 8 for monitoring the properties of the drainage liquid are arranged at the top end in the drainage bottle 2; a micro pressure sensor 9 for monitoring the measurement of the drainage liquid is arranged at the bottom of the drainage bottle 2; a radio frequency business card 10 for storing patient information is arranged on the front of the drainage bottle 2; wherein the micro spectrometer 7, the multi-parameter sensor 8 and the micro pressure sensor 9 are connected with a remote monitoring platform through wireless transmission technology, and are used for feeding back the color data, the measurement data and the multi-parameter properties of the drainage liquid monitored in real time to the remote monitoring platform.

[0022] Specifically, the liquid medicine is guided to the postoperative wound surface for debridement through the cleaning liquid equipment A; the drainage pipe 5 is arranged in the postoperative wound surface; the negative pressure pump 6 is opened to generate negative pressure; wound exudate is sucked into the drainage pipe 5 and enters the drainage bottle 2 through the connecting pipe 3, so as to form a closed drainage environment; in this way, the risk of infection is reduced, and the wound healing is accelerated; the color data, the multi-parameter properties and the measurement data of the drainage liquid are collected in real time through the micro spectrometer 7, the multi-parameter sensor 8 and the micro pressure sensor 9; the potential complication risk is analyzed combined with artificial intelligence and big data; personalized nursing suggestions are provided; medical staff can discover problems in time and intervene; secondly, the patient's personal information is recorded in the radio frequency business card 10; before the intervention scheme is implemented on the patient, the bar code of the radio frequency business card 10 is identified through a scanning device; whether the patient and the recorded information are consistent is obtained and verified; medical errors are prevented; medical disputes are reduced.

[0023] The above-mentioned beneficial effects are achieved: the drainage liquid data are collected in real time through the micro spectrometer 7, the multi-parameter sensor 8 and the micro pressure sensor 9 in the negative pressure drainage device 1; the data are transmitted to the remote monitoring platform through wireless transmission technology; medical staff can check the real-time data and risk prediction results of the patient through the platform at any time and any place; the treatment scheme can be adjusted in time; the nursing efficiency and quality are improved.

[0024] Please refer to Figs. 1-3 : The postoperative drainage monitoring system for clinical nursing comprises a data acquisition unit, a wireless transmission unit, a cloud big data platform and a remote monitoring platform.

[0025] The data acquisition unit is composed of the miniature spectrometer 7, the character multi-parameter sensor 8, the miniature pressure sensor 9 and the smart bracelet 11, and is configured to monitor the color data, the metering data and the property multi-parameters of the drainage liquid in real time based on the miniature spectrometer 7, the character multi-parameter sensor 8 and the miniature pressure sensor 9, for example, the color R, G and B quantitative values, the HSL values, the character labels, the metering mL / 2h, mL / 8h and mL / 24h, the property viscosity, the turbidity, the pH, the amylase, the bilirubin and the cell count; the smart bracelet 11 is worn on the wrist of the patient, and the vital sign data of the patient after the operation is monitored in real time based on the smart bracelet 11, including the heart rate, the blood pressure and the body temperature.

[0026] The wireless transmission unit is configured to transmit the color data, the metering data and the property multi-parameters of the drainage liquid and the vital sign data to the cloud big data platform for intelligent analysis based on the wireless transmission technology, and synchronously feed back the analysis results to the remote monitoring platform, so as to realize the data interaction of the drainage monitoring system.

[0027] The cloud big data platform is configured to build a risk prediction model, extract the feature engineering in the color data, the metering data and the property multi-parameters of the drainage liquid in the data through the risk prediction model, and predict the probability of the patient after the operation to have complications; based on different complication types and probabilities, a dynamic intervention scheme is formulated; the cloud big data platform includes: The label setting module is configured to set the complication labels based on the historical data, and take the complication labels as the input set of the risk prediction model; for example, bleeding, infection, anastomotic fistula, pancreatic fistula, chylous fistula and abdominal abscess, which are double-blindly labeled by two specialist doctors according to the ISGPS and CDC definitions; in the specific implementation, the postoperative patient data of a hospital in the past several years, for example, 5 years, is collected, including the color, metering and properties such as turbidity and viscosity of the drainage liquid, and the vital signs such as heart rate, blood pressure and body temperature; at the same time, whether the patient after the operation has complications such as bleeding, infection and anastomotic fistula is recorded.

[0028] The quality control module is configured to supplement the missing values of the historical data by using multiple imputation, and analyze the abnormal data according to the clinical knowledge to judge whether it needs to be corrected or deleted; for example, <10% is imputed by multiple imputation; >10% is taken as a separate missing indicator variable; based on the end time of the operation as the base point, all the data are time-aligned; for example, the end time of the operation is taken as t=0, and the data are recorded once an hour.

[0029] The feature extraction module is configured to extract features from the color data, metering data, and property multi-parameters of the drainage fluid; for example: color: convert color descriptions such as bright red, dark red, and milk white into numerical features. Metering: calculate the average value, maximum value, and change rate of the drainage fluid per hour, per 8 hours, and per 24 hours. Property: convert property descriptions such as turbidity and viscosity into binary features.

[0030] The feature calculation module is configured to build a random forest model as a risk prediction model, train it using historical data, and evaluate its performance using cross-validation. It also calculates the AUC values of each feature engineering in the risk prediction model to assess its predictive ability for complications.

[0031] The threshold setting module is configured to set different thresholds for different types of complications based on their severity and clinical needs; for example: bleeding: risk threshold set to 10%; infection: risk threshold set to 20%; chylous leakage: risk threshold set to 30%.

[0032] The threshold triggering module is configured to set a default risk probability threshold based on clinical experience and model performance, such as 20%; when the model predicts a complication risk probability ≥ the risk probability threshold, the intervention condition is triggered.

[0033] The category differentiation module is configured to determine the type of complication that the patient may have based on the model's prediction results; for example, if the risk probability of bleeding is higher than 10%, it is determined that the patient may have bleeding; and the risk probability of each complication is recorded to provide a basis for subsequent intervention; for example, the risk probability of infection is 25%, which is recorded as "infection risk 25%".

[0034] The model prediction module is configured to use the trained risk prediction model to predict real-time drainage fluid data and vital sign data to obtain the complication risk probability at each time point; compare the predicted risk probability with the preset threshold; if the risk probability is higher than the threshold, it is determined that the patient may have the corresponding complication.

[0035] The risk intervention module is configured to develop appropriate intervention strategies based on the determined complication type and probability; for example: bleeding risk ≥10%: immediately notify the doctor, review blood routine and coagulation function, and perform imaging examination such as CT angiography if necessary. Infection risk ≥20%: collect blood culture and drainage fluid culture, and adjust the antibiotic treatment plan according to the results. Pancreatic fistula risk ≥20%: fasting, use somatostatin analogues, and monitor the amylase level of the drainage fluid.

[0036] The dynamic adjustment module is configured to dynamically adjust the intervention plan based on changes in the patient's condition and real-time monitoring data; for example, if the drainage fluid volume suddenly increases, the bleeding risk is re-evaluated and the intervention measures are adjusted.

[0037] The remote monitoring platform is configured to display the drainage fluid data and vital sign data of the patient in real time, and the risk prediction result and the corresponding intervention scheme; and feedback is made to the model prediction result according to the actual intervention effect, and the system updates the model according to the feedback data to optimize the prediction accuracy.

[0038] The above-mentioned beneficial effects are achieved: through the risk prediction model, the color, metering, properties and other multi-parameter data of the drainage fluid and the vital sign data are analyzed, the probability of occurrence of postoperative complications can be predicted in advance, a scientific basis is provided for clinical decision-making, and medical personnel can better understand the patient's condition; and according to different types and probabilities of complications, the intervention measures are dynamically adjusted in combination with the patient's condition changes and real-time monitoring data, personalized nursing of the patient is realized, the treatment is more targeted and effective, thus precise intervention of complications is realized, the incidence and severity of complications are reduced, and the postoperative rehabilitation quality of the patient is improved; at the same time, the system combines big data, artificial intelligence technology and clinical nursing, promotes the informatization and intelligentization development of the medical care field, and provides a new idea and direction for innovation in the biomedical engineering field.

[0039] Working principle: the color, metering and property multi-parameter data of the drainage fluid are monitored in real time through the integrated miniature spectrometer 7, the property multi-parameter sensor 8 and the miniature pressure sensor 9, the postoperative vital signs of the patient are monitored through the smart bracelet 11; the real-time transmission is realized through wireless transmission technology to the cloud big data platform, the key features are extracted from the drainage fluid data through the risk prediction model in the cloud big data platform, and the AUC value is calculated; and the risk probability threshold is compared to determine whether the intervention condition is triggered; if triggered, the corresponding intervention scheme is obtained according to the prediction result, and the intervention scheme is dynamically adjusted in combination with the patient's condition changes and real-time data; medical personnel can view the drainage fluid data, vital sign data, risk prediction result and intervention scheme of the patient in real time through the remote monitoring platform, and realize intelligent and precise nursing and complication management of postoperative patients.

[0040] The above-mentioned threshold setting module comprises: The first threshold setting unit is configured to determine the first threshold of the patient based on each type of complication and the severity of the complication, wherein the severity is determined according to the symptom characteristics of the complication; The second threshold setting unit is configured to determine the second threshold of the patient based on the clinical needs of the patient, wherein the clinical needs are determined according to the individual difference data and the clinical sign data of the patient; The threshold comprehensive setting unit is configured to determine the risk probability threshold of the patient based on the real-time patient needs in combination with the first threshold and the second threshold.

[0041] In this embodiment, the risk probability threshold is Y; ; ; wherein Y is a risk probability threshold value of the patient, is a first threshold value, is a second threshold value, is a real-time threshold influence weight of the first threshold value, is a real-time threshold influence weight of the second threshold value, is a standard symptom characteristic value of a kth symptom characteristic in the symptom characteristics of the complication, is patient characteristic data of a feature type corresponding to the standard symptom characteristic value of the kth complication, is a feature conversion coefficient of the patient characteristic data, is a i th patient individual data of the patient, is a standard individual data corresponding to the i th patient individual data, is a data conversion coefficient of the patient individual data, Q represents a risk threshold value of the patient clinical sign data, is a j th clinical sign data of the patient, is an upper limit of the standard sign data corresponding to the j th clinical sign data, is a lower limit of the standard sign data corresponding to the j th clinical sign data, is a data conversion coefficient of the patient clinical sign data, 、 respectively are data influence weights of the individual data and the clinical sign data of the patient, n is a feature quantity of the symptom characteristics of the patient complication with a complication occurrence probability, p is a data quantity of the individual difference data of the patient, and m is a data quantity of the clinical sign data of the patient, wherein, 、 the sum of the weights of is 1, 、 the sum of the weights of is 1.

[0042] The beneficial effects achieved by the above are: by combining the comprehensive judgment of the severity of different complications and the clinical needs of the patient, the corresponding threshold value is determined, and the influence weight is determined by combining the influence factors of the threshold value corresponding to the severity and clinical needs of different complications, the risk probability threshold value of the patient is determined, which can be adjusted in time in combination with the real-time patient needs, realizes the precise intervention of the complication, and improves the postoperative rehabilitation quality of the patient.

[0043] The above dynamic adjustment module comprises: The curve construction unit is configured to acquire real-time monitoring data and historical monitoring data in a preset time period of a patient, perform data classification, construct a monitoring change curve of monitoring data of each data type based on a data classification result, and obtain a monitoring change curve group of the patient; The curve extraction unit is configured to extract an abnormal fluctuation curve in the monitoring change curve group as a first monitoring change curve group. The monitoring processing unit is configured to extract a monitoring data type corresponding to each monitoring change curve in the first monitoring change curve group, and determine a data key degree of corresponding monitoring data of the monitoring data type based on the data key degree, which is determined based on an influence degree of each monitoring data on a type and a probability of a complication of the patient. The data sorting unit is configured to perform data sorting on the monitoring data based on the data key degree, and determine a monitoring priority queue based on a data sorting result. The first scheme determination unit is configured to determine a patient intervention scheme corresponding to each monitoring data in the monitoring priority queue, and obtain a first adjustment intervention scheme of the patient. The calibration comparison unit is configured to extract a historical patient with the highest similarity to the patient sign data and the patient real-time condition from a patient database as a calibration patient. The intervention sorting unit is configured to compare a historical calibration intervention scheme of the calibration patient with the first adjustment intervention scheme. The intervention sorting unit is configured to compare a historical calibration intervention scheme of the calibration patient with the first adjustment intervention scheme. The intervention sorting unit is configured to compare a historical calibration intervention scheme of the calibration patient with the first adjustment intervention scheme. The second scheme determination unit is configured to combine the intervention priority queue with real-time feedback and real-time intervention demand to obtain a second adjustment intervention scheme of the patient. The intervention adjustment unit is configured to dynamically adjust the intervention scheme based on the first adjustment intervention scheme and the second adjustment intervention scheme.

[0044] In this embodiment, the real-time monitoring data refers to various data reflecting the physical condition of the patient acquired in real time during the postoperative treatment or monitoring process, which can reflect the current physiological state of the patient in real time. For example, the drainage volume is recorded once an hour through the scale on the drainage device; the temperature and humidity of the skin around the drainage site of the patient are continuously monitored by using an electronic monitor; and the pressure value in the drainage tube is acquired in real time by using a pressure sensor.

[0045] In this embodiment, the historical monitoring data refers to the relevant data of the patient's physical condition obtained within a preset time period. For example, the first three days after the operation are set as the preset time period, and the drainage volume data recorded every 4 hours per day is collected to form the historical monitoring data of the drainage volume at different time points within the three days.

[0046] In this embodiment, data classification is to divide the obtained monitoring data according to data rules or data characteristics. For example, the monitoring data is divided into drainage volume related data (such as hourly drainage volume, 24-hour total drainage volume, etc.), drainage fluid property related data (such as color, transparency, odor, whether containing blood clots, etc.), and drainage site related data (such as whether the skin of the drainage site is red and swollen, the degree of pain, etc.).

[0047] In this embodiment, one monitoring change curve corresponds to one monitoring data, wherein the horizontal axis of the monitoring change curve is time, and the vertical axis is the numerical value of the monitoring data. Through the monitoring curve, the data fluctuation of the monitoring data can be more intuitively judged, and the abnormal fluctuation curve can be extracted. For example, the change curve of the hourly drainage volume within 72 hours after the operation is drawn, the horizontal axis is time (hour), and the vertical axis is drainage volume (milliliter).

[0048] In this embodiment, the monitoring change curve group is a set composed of the monitoring change curves of each data type of monitoring data based on different data classification results. For example, the curve set of the patient's drainage volume change curve, the drainage fluid color change curve, and the skin temperature change curve of the drainage site is the monitoring change curve group.

[0049] In this embodiment, the abnormal fluctuation curve refers to a curve that does not match the normal change range or trend in the monitoring change curve group and appears abnormal fluctuation. For example, under normal circumstances, the postoperative drainage volume will gradually decrease, and if the drainage volume change curve suddenly rises sharply within a certain period of time, it is far beyond the normal decreasing trend.

[0050] In this embodiment, the data criticality is a measurement index determined based on the influence degree of each postoperative drainage monitoring data on the types and probabilities of postoperative complications. The higher the data criticality, the greater the influence on the patient's condition. For example, the criticality of the drainage volume data is higher than that of the skin humidity of the drainage site, because excessive drainage volume may indicate bleeding, infection, etc., which is easy to cause severe complications such as hemorrhagic shock and abdominal infection.

[0051] In this embodiment, the monitoring priority queue is obtained by sorting the postoperative drainage monitoring data according to the data criticality, for example, the postoperative drainage monitoring data of a patient is sorted according to the data criticality as follows: drainage volume, drainage fluid color, skin temperature of drainage site, skin humidity of drainage site, and the monitoring priority queue is arranged as follows: drainage volume, drainage fluid color, skin temperature of drainage site, skin humidity of drainage site.

[0052] In this embodiment, the patient intervention scheme is determined according to the type and probability of complications determined by the risk prediction model, for example, the patient intervention scheme includes collecting drainage fluid for bacterial culture, using sensitive antibiotics according to the culture results, and strengthening the cleaning and nursing of the skin of the drainage site.

[0053] In this embodiment, the first adjustment intervention scheme is a preliminary adjusted intervention scheme for the current postoperative drainage condition of the patient, which is formulated according to the patient intervention scheme corresponding to each postoperative drainage monitoring data in the monitoring priority queue, for example, the first adjustment intervention scheme may be to increase the monitoring frequency of the drainage volume on the basis of the original nursing measures.

[0054] In this embodiment, the patient database is a database storing a large amount of postoperative drainage related information of patients, including basic information of patients, surgical information, postoperative drainage monitoring data, intervention schemes and effects, etc.

[0055] In this embodiment, the calibration patient is a historical patient with the highest similarity to the current patient in terms of surgical type, postoperative drainage condition and real-time condition, for example, a patient with similar age, same surgical type and similar drainage fluid color is determined as the calibration patient of the current patient.

[0056] In this embodiment, the historical calibration intervention scheme is the intervention scheme implemented by the calibration patient, for example, the intervention scheme includes limiting patient activity, using hemostatic drugs, keeping the drainage tube unobstructed, etc.

[0057] In this embodiment, the sub-intervention scheme is a specific execution scheme in the intervention scheme, for example, in the intervention scheme for a patient with hypertension, limiting patient activity is a sub-intervention scheme, and using hemostatic drugs is a sub-intervention scheme.

[0058] In this embodiment, the scheme priority refers to the order determined by sorting the importance and urgency of each sub-intervention scheme in the intervention scheme, the higher the scheme priority, the more the sub-intervention scheme needs to be prioritized for scheme execution, for example, for a patient with postoperative drainage tube blockage and excessive drainage volume, the sub-intervention scheme of immediately unblocking the drainage tube has the highest priority in the intervention scheme.

[0059] In this embodiment, the first scheme priority is the scheme priority of the sub-intervention scheme in the historical calibration intervention scheme, which is directly taken as the priority of the corresponding sub-intervention scheme in the first adjusted intervention scheme.

[0060] In this embodiment, the intervention priority queue is a queue obtained by arranging the corresponding sub-intervention schemes in the first adjusted intervention scheme according to the priority order in combination with the first scheme priority, which clearly defines the execution order of each sub-intervention scheme.

[0061] In this embodiment, the real-time feedback refers to information about changes in the patient's condition, treatment effect, etc. obtained during the patient's treatment or monitoring; and the real-time intervention requirement refers to the requirement for immediate intervention measures or adjustment of the intervention scheme according to the real-time feedback, such as increasing the infusion speed of the patient, adjusting the dose of anesthetic drugs, etc. when the patient's blood pressure suddenly drops is the real-time intervention requirement.

[0062] In this embodiment, the second adjusted intervention scheme is an intervention scheme obtained by further adjusting the first adjusted intervention scheme in combination with the intervention priority queue, the real-time feedback and the real-time intervention requirement, such as the intervention priority queue of the first adjusted intervention scheme is to limit patient activity, use hemostatic drugs, keep the drainage tube unobstructed, and strengthen skin care at the drainage site, but it is found in real-time monitoring that the patient's drainage fluid color suddenly becomes dark and has an odor (real-time feedback), which suggests that there may be signs of infection. At this time, antibiotic treatment is needed immediately (real-time intervention requirement), so the second adjusted intervention scheme will improve the intervention priority of using antibiotics, thereby adjusting the order of the intervention priority queue.

[0063] In this embodiment, the dynamic adjustment continuously modifies the intervention scheme according to the first adjusted intervention scheme and the second adjusted intervention scheme, and the new situation of the patient continuously appearing during the treatment, wherein the modification of the scheme includes increasing or decreasing the dose of drugs, replacing the treatment drugs, adjusting the nursing measures, etc.

[0064] The above-mentioned beneficial effects are achieved: by obtaining the monitoring data and constructing the monitoring change curve, the abnormal monitoring data can be accurately identified, so as to determine the monitoring priority queue according to the data key degree of the abnormal monitoring data, and determine the intervention priority queue by referring to the historical intervention scheme of the calibration patient, so as to formulate the second adjustment scheme in combination with the real-time feedback and dynamically adjust, thereby dynamically adjusting the intervention scheme, which can more timely and accurately respond to the changes in the patient's condition, reduce the risk of complications in patients, and improve the quality of postoperative care.

[0065] Working principle: first, collect the patient drainage data, construct the monitoring change curve group, so as to identify the abnormal fluctuation curve; then, based on the influence degree of monitoring data on patient complications, the key degree of monitoring data with abnormal fluctuation is determined, the monitoring priority queue is formed, and the first adjustment intervention scheme is obtained; at the same time, similar calibration patients are found out from the patient database, the historical calibration intervention scheme of the calibration patient is referred to, the priority of the sub intervention scheme is determined, the intervention priority queue is formed combined with the first adjustment intervention scheme, and the second adjustment intervention scheme is determined combined with real-time feedback and real-time intervention demand, so as to dynamically adjust the intervention scheme of the patient according to the first adjustment intervention scheme and the second adjustment intervention scheme, and ensure timely and accurate response to the patient's condition change.

[0066] It should be noted that in this paper, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0067] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A post-operative drainage device for clinical care, comprising: The negative pressure drainage device (1) is characterized in that: the negative pressure drainage device (1) is composed of a drainage bottle (2), a connecting pipe (3), an adapter (4) and a drainage pipe (5), the top of the drainage bottle (2) is provided with a negative pressure pump (6) for providing negative pressure for the drainage bottle (2), the top end of the inside of the drainage bottle (2) is provided with a miniature spectrometer (7) for monitoring the color of the drainage liquid, and a character multi-parameter sensor (8) for monitoring the properties of the drainage liquid, the bottom of the drainage bottle (2) is provided with a miniature pressure sensor (9) for monitoring the drainage liquid metering, and the front of the drainage bottle (2) is provided with a radio frequency business card (10) for storing patient information.

2. The post-operative drainage device for clinical care of claim 1, wherein, The miniature spectrometer (7), the character multi-parameter sensor (8) and the miniature pressure sensor (9) are respectively connected with a remote monitoring platform through wireless transmission technology, and are used to feed back the color data, the metering data and the property multi-parameters of the real-time monitored drainage liquid to the remote monitoring platform.

3. The post-operative drainage device for clinical care of claim 1, wherein, The top of the drainage bottle (2) is inserted with the connecting pipe (3), one side of the connecting pipe (3) is connected with the adapter (4) and the drainage pipe (5), one end of the drainage pipe (5) is in contact with the postoperative wound surface, and the top of the drainage bottle (2) is further provided with a hanging ring (21), and the bottle body of the drainage bottle (2) is provided with a scale line (22).

4. A post-operative drainage monitoring system for clinical care for monitoring a post-operative drainage device according to any one of claims 1 to 3, characterized in that It comprises: a data acquisition unit, a wireless transmission unit, a cloud big data platform and a remote monitoring platform; the data acquisition unit is composed of a miniature spectrometer (7), a character multi-parameter sensor (8), a miniature pressure sensor (9) and a smart bracelet (11), and is configured to monitor the color data, the metering data and the property multi-parameters of the drainage liquid in real time based on the miniature spectrometer (7), the character multi-parameter sensor (8) and the miniature pressure sensor (9), and to monitor the vital sign data of the patient after the operation in real time based on the smart bracelet (11), including heart rate, blood pressure and body temperature; the wireless transmission unit is configured to transmit the color data, the metering data and the property multi-parameters of the drainage liquid, and the vital sign data to the cloud big data platform for intelligent analysis in real time based on wireless transmission technology; and the analysis results are synchronously fed back to the remote monitoring platform, realizing the data interaction of the drainage monitoring system; the cloud big data platform is configured to build a risk prediction model, extract feature engineering from the color data, the metering data and the property multi-parameters of the drainage liquid in the data through the risk prediction model, predict the probability of the patient developing complications after the operation, develop a dynamic intervention scheme based on different types and probabilities of complications, generate and dynamically optimize a personalized intervention scheme in real time by building a monitoring change curve, identifying abnormal fluctuations, and combining the key degree of patient data with historical similar cases for calibration, and preset a risk probability threshold based on clinical experience as a trigger for intervention; the remote monitoring platform is configured to display the drainage liquid data and the vital sign data of the patient, and the risk prediction results and the corresponding intervention scheme in real time.

5. The post-operative drainage monitoring system for clinical care of claim 4, wherein, The cloud big data platform comprises: a feature extraction module configured to extract features from the color data, the metering data and the property multi-parameters of the drainage liquid; The feature calculation module is configured to build a random forest model as a risk prediction model, train the risk prediction model through historical data, and evaluate the performance of the risk prediction model using cross-validation.

6. The post-operative drainage monitoring system for clinical care of claim 5, wherein, The cloud big data platform further comprises: The threshold setting module is configured to set different thresholds for different types of complications according to their severity and clinical needs. The threshold triggering module is configured to set a default risk probability threshold according to clinical experience and model performance, and trigger an intervention condition when the predicted risk probability of the complication is greater than or equal to the risk probability threshold.

7. The post-operative drainage monitoring system for clinical care of claim 6, wherein, The threshold setting module comprises: The first threshold setting unit is configured to determine the first threshold of the patient based on each type of complication and the severity of the complication, wherein the severity is determined according to the symptom characteristics of the complication. The second threshold setting unit is configured to determine the second threshold of the patient based on the clinical needs of the patient, wherein the clinical needs are determined according to the individual difference data and clinical sign data of the patient. The threshold comprehensive setting unit is configured to determine the risk probability threshold of the patient based on the real-time patient needs, the first threshold, and the second threshold.

8. The post-operative drainage monitoring system for clinical care of claim 6, wherein, The cloud big data platform further comprises: The model prediction module is configured to use the trained risk prediction model to predict the real-time drainage fluid data and vital sign data to obtain the risk probability of the complication at each time point, and compare the predicted risk probability with the preset threshold. The risk intervention module is configured to develop a corresponding intervention scheme according to the type and probability of the complication determined. The dynamic adjustment module is configured to dynamically adjust the intervention scheme according to the changes in the patient's condition and real-time monitoring data.

9. The post-operative drainage monitoring system for clinical care of claim 8, wherein, The dynamic adjustment module comprises: The curve construction unit is configured to obtain real-time monitoring data of the patient and historical monitoring data within a preset time period, classify the data, construct a monitoring change curve for each data type based on the data classification result, and obtain a monitoring change curve group of the patient. The curve extraction unit is configured to extract an abnormal fluctuation curve in the monitoring change curve group as a first monitoring change curve group. The monitoring processing unit is configured to extract the monitoring data type corresponding to each monitoring change curve in the first monitoring change curve group, and obtain the data key degree of the monitoring data corresponding to the monitoring data type. The data sorting unit is configured to sort the monitoring data based on the data key degree, and determine a monitoring priority queue based on the data sorting result. The first scheme determination unit is configured to determine the patient intervention scheme corresponding to each monitoring data in the monitoring priority queue to obtain a first adjusted intervention scheme for the patient. The calibration comparison unit is configured to extract a historical patient with the highest similarity to the patient's sign data and real-time condition from the patient database as a calibration patient. The intervention sequencing unit is configured to compare the historical calibration intervention scheme of the calibration patient with the first adjusted intervention scheme based on the calibration; The first adjusted intervention scheme is extracted from the historical calibration intervention scheme, and the scheme priority of the historical calibration intervention scheme is used as the first scheme priority of the corresponding intervention scheme in the first adjusted intervention scheme; The first scheme priority and the corresponding intervention scheme in the first adjusted intervention scheme are combined to obtain an intervention priority queue; The second scheme determination unit is configured to combine the intervention priority queue with real-time feedback and real-time intervention demand to obtain a second adjusted intervention scheme of the patient; The intervention adjustment unit is configured to dynamically adjust the intervention scheme based on the first adjusted intervention scheme and the second adjusted intervention scheme.

10. The post-operative drainage monitoring system for clinical care of claim 8, wherein, The cloud big data platform further comprises: The category distinguishing module is configured to determine the types of possible complications of the patient according to the model prediction result, and record the risk probability of each type of complication to provide a basis for subsequent intervention; The label setting module is configured to set a complication label based on historical data, and use the complication label as an input set of the risk prediction model; The quality control module is configured to supplement missing values of the historical data by using multiple imputations, analyze abnormal data according to clinical knowledge, and determine whether the abnormal data need to be corrected or deleted; and align all data in time based on the end time of the surgery as a base point.

Images