A mechanical ventilation treatment data management method and system

Abstract

The present invention relates to a mechanical ventilation therapy data management method and system, which belongs to the technical field of mechanical ventilation therapy data management and real-time analysis, solves the many difficulties caused by huge heterogeneous data to clinical applications, and the problem of affecting the quality of mechanical ventilation therapy, and improves the data processing efficiency of doctors and nurses. The method includes: placing patient information and real-time received mechanical ventilation therapy data into standard data structure objects respectively to form standardized data; the mechanical ventilation therapy data includes mechanical ventilation data; data cleaning is performed on the standardized data to obtain a medical data set; data annotation is performed on the mechanical ventilation therapy data in the medical data set; the data annotation includes: marking the amplitude, frequency and shape of waveform data, marking the measurement value of numerical data, and marking abnormal events; when a report generation instruction is received, a mechanical ventilation therapy data analysis report is generated based on the medical data set carrying the data annotation.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical ventilation therapy data management and real-time analysis, and in particular to a mechanical ventilation therapy data management method and system. Background Art

[0002] Ventilator-mechanical ventilation treatment technology has been widely used in intensive care units (ICU, NICU, RICU, EICU, etc.), anesthesia departments and other departments in hospitals around the world. Every year, tens of millions of critically ill patients around the world use this treatment technology. If used improperly, it will cause complications such as lung damage, ventilator-associated pneumonia, arrhythmia, acute heart failure, aggravate the primary disease, and endanger the patient's life.

[0003] In a common mechanical ventilation treatment scenario, the patient is equipped with a ventilator and a multi-parameter monitor. Doctors and nurses observe and analyze continuous mechanical ventilation waveform data and numerical data on each ventilator screen to determine the working status of the ventilator, analyze the patient's respiratory mechanics changes and disease evolution, the state of human / machine interaction, and observe changes in physiological parameters such as heart rate, blood pressure, blood oxygen saturation, and body temperature. In addition, they need to observe blood gas analysis, laboratory diagnosis, imaging and other data through the clinical information system (HIS / CIS / EMR system) to fully understand the disease status and evolution trend, adjust ventilator parameters and treatment strategies, and prevent and control various complications.

[0004] However, each ventilator generates up to hundreds of MB of waveform data in 24 hours. Doctors and nurses frequently patrol the screens of various devices every day to observe and analyze various data. At the same time, they also need to understand the patient's condition in the past and retrieve 24 / 48 hours of mechanical ventilation data and physiological parameters during treatment from the device or clinical information system. However, the equipment and clinical information system only store numerical data, lack important waveform data, and contain a large amount of interference data, as well as data analyzed incorrectly by the device itself. Doctors and nurses need to spend a lot of time reading and analyzing each section and manually identifying and judging. Obviously, the clinical application of this data creates heavy work pressure and requires a lot of manpower to deal with. However, ventilators from various manufacturers cannot provide 24-hour mechanical ventilation data analysis reports, and there is a general shortage of doctors and nurses in ICUs in hospitals around the world. So far, this situation has not changed.

[0005] In addition, a hospital is usually equipped with dozens to hundreds of ventilators and multi-parameter monitors from different manufacturers. There are complex issues such as different data communication protocols and data formats for the same type of equipment, and different types of equipment. Clinical information systems also come from different manufacturers. These problems increase the difficulty of processing mechanical ventilation treatment data. Over the past decades, ventilator technology has continued to develop, but the clinical application of mechanical ventilation treatment data has not improved and has major defects. Summary of the invention

[0006] In view of the above analysis, the embodiments of the present invention aim to provide a mechanical ventilation treatment data management method and system to solve the many difficulties caused by huge heterogeneous data in clinical applications, the poor medical quality of mechanical ventilation treatment, and the low efficiency of data processing by doctors and nurses.

[0007] In one aspect, the present invention discloses a method for managing mechanical ventilation treatment data, the method comprising:

[0008] Putting patient information and mechanical ventilation treatment data received in real time into standard data structure objects respectively to form standardized data; the mechanical ventilation treatment data includes mechanical ventilation data;

[0009] Performing data cleaning on the standardized data to obtain a medical data set;

[0010] Data annotation is performed on the mechanical ventilation treatment data in the medical data set; the data annotation includes: marking the amplitude, frequency and shape of the waveform data, marking the measurement value of the numerical data, and marking abnormal events;

[0011] When a report generation instruction is received, a mechanical ventilation treatment data analysis report is generated based on the medical data set carrying the data annotations.

[0012] On the basis of the above scheme, the present invention also makes the following improvements:

[0013] Furthermore, the mechanical ventilation treatment data also includes one or more of the following: physiological parameter data, blood gas analysis data, laboratory diagnosis data, imaging data, clinical diagnosis information, and device ID for outputting mechanical ventilation treatment data; wherein,

[0014] Physiological parameter data are output by a multi-parameter monitor;

[0015] Patient information, blood gas analysis data, laboratory diagnosis data, imaging data, and clinical diagnosis and treatment information are all output by the clinical information system;

[0016] Mechanical ventilation data is output by the ventilator device.

[0017] Furthermore, in the mechanical ventilation data, the waveform data includes: mechanical ventilation pressure waveform data, volume waveform data, flow waveform data and end-tidal carbon dioxide waveform data; the numerical data includes: airway resistance, work of breathing, compliance, ventilation parameters and respiratory rate;

[0018] During the data labeling process of the mechanical ventilation data,

[0019] The marked waveform data include: the start of inspiration, inspiration, end of inspiration, start of expiratory, expiratory, end of expiratory, peak airway pressure, peak expiratory flow rate of each mechanical ventilation waveform, and the inspiration baseline, expiratory plateau height, start and end of expiratory plateau of the end-tidal carbon dioxide waveform;

[0020] The marked abnormal events include: human-machine asynchrony events, airway pressure too high / too low, positive end-expiratory pressure too high / too low, exhaled tidal volume too high / too low, minute ventilation too high / too low, inspired oxygen concentration too high / too low, oxygen concentration too high / too low, respiratory rate too high / too low, end-tidal carbon dioxide partial pressure too high / too low, tidal volume too high / too low, ventilator shutdown, suffocation and increased airway pressure.

[0021] Furthermore, in the physiological parameter data, waveform data includes: electrocardiogram waveform data, respiratory waveform data, volume pulse waveform data and invasive arterial pressure waveform data; numerical data includes: heart rate, blood pressure, blood oxygen saturation and body temperature;

[0022] In the process of labeling the physiological parameter data,

[0023] The morphology of the marked ECG waveform data includes: P wave, QRS complex, T wave, ST segment, PR interval, RR interval and QT interval of each ECG cycle waveform;

[0024] The morphology of the marked respiratory waveform data includes: the peaks and troughs of each respiratory cycle waveform;

[0025] The morphology of the marked volume pulse wave waveform data includes: the peaks and troughs of each pulse cycle waveform;

[0026] The morphology of the marked invasive arterial pressure waveform data includes: the peaks and troughs of each arterial systolic-diastolic cycle waveform;

[0027] Abnormal events marked include: tachycardia, bradycardia, arrhythmia, cardiac arrest, atrial flutter, atrial fibrillation, ventricular flutter, ventricular fibrillation, atrioventricular block, ST segment elevation / depression, high / low heart rate, high / low blood pressure, high / low body temperature and low blood oxygen saturation.

[0028] Further, in the blood gas analysis data, the numerical data include: oxygen partial pressure, carbon dioxide partial pressure, total carbon dioxide, potassium ion concentration, sodium ion concentration, bicarbonate ion concentration and pH value;

[0029] In the process of labeling the blood gas analysis data,

[0030] The marked abnormal events include: oxygen partial pressure too high / too low, carbon dioxide partial pressure too high / too low, total carbon dioxide too high / too low, potassium ion concentration too high / too low, sodium ion concentration too high / too low, and pH too high / too low.

[0031] Further, in the laboratory diagnosis data, the numerical data include: white blood cell count, red blood cell count, platelet count, hemoglobin, hematocrit, blood glucose, prothrombin time and activated partial thromboplastin time;

[0032] In the process of data annotation for experimental diagnostic data,

[0033] The marked abnormal events include: white blood cell count is too high / too low, red blood cell count is too high / too low, hemoglobin is too high / too low, hematocrit is too high / too low, blood sugar is too high / too low, platelet count is too high / too low, prothrombin time is too long / too short, activated partial thromboplastin time is too long / too short.

[0034] Furthermore, the report framework of the mechanical ventilation treatment data analysis report includes a report summary page and a report analysis page; wherein the report summary page is divided into a basic information area, a mechanical ventilation data area and an analysis conclusion area;

[0035] The basic information area includes at least patient information and key time node information; and also includes the following optional items: clinical diagnosis information, device ID for outputting mechanical ventilation treatment data, and mechanical ventilation mode and setting information;

[0036] The mechanical ventilation data area at least includes respiratory analysis results, human-machine asynchrony event analysis results, airway resistance change analysis results, compliance change analysis results and lung injury warning analysis results; each analysis result of the mechanical ventilation data area is obtained based on mechanical ventilation data and its data annotation; the respiratory analysis results include shallow rapid breathing index;

[0037] The analysis conclusion area is used to display the analysis description and conclusion of the mechanical ventilation treatment data;

[0038] The report analysis page is used to display detailed analysis reports of one or more mechanical ventilation treatment data in the form of charts.

[0039] Further, the report analysis page includes a trend chart page; the trend chart page includes:

[0040] Pressure, volume, and ventilation trend analysis area for mechanical ventilation data;

[0041] Respiratory rate trend analysis area;

[0042] Inspired oxygen concentration trend analysis area; and,

[0043] The trend analysis description and conclusion area is used to summarize the trend trends of each trend analysis area and give a trend analysis description and corresponding conclusions;

[0044] The contents of each trend analysis area in the trend graph page are obtained based on mechanical ventilation data and data annotations thereof.

[0045] Further, the report analysis page also includes a data page;

[0046] The data page at least includes:

[0047] Mechanical ventilation setting parameter data subpage, including: mechanical ventilation setting parameter record table, mechanical ventilation ventilation mode and setting parameter change trend chart;

[0048] Mechanical ventilation monitoring parameter data subpage, including mechanical ventilation monitoring parameter record table;

[0049] The contents of the mechanical ventilation setting parameter data subpage and the mechanical ventilation monitoring parameter data subpage are obtained based on the mechanical ventilation data and data annotation thereof;

[0050] The data page also includes the following optional data subpages:

[0051] 24-hour physiological parameter data subpage, including 24-hour vital signs monitoring record sheet, heart rate / respiratory rate / blood pressure trend chart; the content of the 24-hour physiological parameter data subpage is obtained based on the physiological parameter data and its data annotation;

[0052] 24-hour blood gas analysis and laboratory diagnosis data subpage, including 24-hour blood gas analysis monitoring record sheet, oxygen partial pressure / carbon dioxide partial pressure / bicarbonate trend chart, and laboratory diagnosis data sheet;

[0053] The content of the 24-hour blood gas analysis and laboratory diagnosis data subpage is obtained based on the blood gas analysis data, laboratory diagnosis data and data annotation thereof;

[0054] Alarm event data subpage, including: alarm event record table, 24-hour alarm event distribution diagram, alarm event comprehensive analysis table;

[0055] The alarm event data subpage is obtained based on abnormal events marked by mechanical ventilation treatment data;

[0056] The extended image data subpage is used to display the image data.

[0057] Furthermore, the report analysis page also includes a graphics page; the graphics page is divided into a mechanical ventilation waveform data area, a mechanical ventilation loop diagram area, and a physiological parameter waveform data area; wherein,

[0058] The mechanical ventilation waveform data area includes continuous normal mechanical ventilation data waveform graphs and abnormal mechanical ventilation event data waveform graphs;

[0059] The mechanical ventilation loop diagram area includes a pressure-volume loop diagram and a flow-volume loop diagram drawn by combining mechanical ventilation pressure, volume, and flow data;

[0060] The contents of the mechanical ventilation waveform data area and the mechanical ventilation ring diagram area are obtained based on the mechanical ventilation data and data annotation thereof;

[0061] The physiological parameter waveform data area includes a continuous physiological parameter data waveform graph synchronized with the mechanical ventilation data waveform graph;

[0062] The content of the physiological parameter waveform data area is obtained based on the physiological parameter data and its data annotation.

[0063] Further, the report generation instruction is a customized report generation instruction;

[0064] The customized report generation instruction includes one or more customized parameters for generating a report summary page and a report analysis page matching the customized parameters;

[0065] The customized parameters include a trend graph page, a data page or a graphic page.

[0066] Furthermore, the method further comprises:

[0067] During the data labeling process, when one or more abnormal events such as ventilator cessation, suffocation, increased airway pressure, cardiac arrest, tachycardia, bradycardia, ventricular flutter, ventricular fibrillation and hypotension occur, an emergency instruction is issued, and the emergency instruction carries information about the abnormal event that occurred.

[0068] Furthermore, the method further comprises:

[0069] During the data labeling process, when one or more abnormal events occur, including human-machine asynchrony events, airway pressure too high, airway pressure too low, positive end-expiratory pressure too high, positive end-expiratory pressure too low, minute ventilation too high, minute ventilation too low, tidal volume too high, tidal volume too low, respiratory rate too high, respiratory rate too low, oxygen concentration too high, and oxygen concentration too low, a ventilator parameter adjustment instruction is issued, and the ventilator parameter adjustment instruction carries information about the abnormal events that have occurred.

[0070] Furthermore, in the standard data structure object, a corresponding data structure object is configured for each mechanical ventilation treatment data and patient information;

[0071] Each decoupled mechanical ventilation treatment data and patient information are placed into corresponding data structure objects to form standardized data.

[0072] Furthermore, while receiving the mechanical ventilation treatment data in real time, it also receives the streaming media data of the treatment scene in real time;

[0073] Live broadcast of waveform data, numerical data, treatment scenarios and mechanical ventilation treatment data analysis reports in the mechanical ventilation treatment data of some or all patients in one or more wards.

[0074] On the other hand, the present invention also provides a mechanical ventilation treatment data management system, the system includes an application software system, a clinical data terminal device and a database server; wherein the application software system includes:

[0075] A data acquisition and standardization module, used to place patient information and mechanical ventilation treatment data received in real time into standard data structure objects to form standardized data; the mechanical ventilation treatment data includes mechanical ventilation data;

[0076] A data cleaning module, used for cleaning the standardized data to obtain a medical data set;

[0077] A data annotation module is used to annotate the mechanical ventilation treatment data in the medical data set; the data annotation includes: marking the amplitude, frequency and shape of the waveform data, marking the measurement value of the numerical data, and marking abnormal events;

[0078] A report generation module, for generating a mechanical ventilation treatment data analysis report based on a medical data set carrying data annotations when a report generation instruction is received;

[0079] A clinical data terminal device, used to issue the report generation instruction, and also used to receive and display the mechanical ventilation treatment data analysis report;

[0080] The database server is used to store the standardized data, medical data set, and mechanical ventilation treatment data analysis report in partitions.

[0081] On the basis of the above scheme, the present invention also makes the following improvements:

[0082] Furthermore, the mechanical ventilation treatment data also includes one or more of the following: physiological parameter data, blood gas analysis data, laboratory diagnosis data, imaging data, clinical diagnosis and treatment information, and the device ID that outputs the mechanical ventilation treatment data; wherein,

[0083] Physiological parameter data are output by a multi-parameter monitor;

[0084] Patient information, blood gas analysis data, laboratory diagnosis data, imaging data, and clinical diagnosis and treatment information are all output by the clinical information system;

[0085] Mechanical ventilation data is output by the ventilator device.

[0086] Furthermore, in the mechanical ventilation data, the waveform data includes: mechanical ventilation pressure waveform data, volume waveform data, flow waveform data and end-tidal carbon dioxide waveform data; the numerical data includes: airway resistance, work of breathing, compliance, ventilation parameters and respiratory rate;

[0087] During the data labeling process of the mechanical ventilation data,

[0088] The marked waveform data include: the start of inspiration, inspiration, end of inspiration, start of expiratory, expiratory, end of expiratory, peak airway pressure, peak expiratory flow rate of each mechanical ventilation waveform, and the inspiration baseline, expiratory plateau height, start and end of expiratory plateau of the end-tidal carbon dioxide waveform;

[0089] The marked abnormal events include: human-machine asynchrony events, airway pressure too high / too low, positive end-expiratory pressure too high / too low, exhaled tidal volume too high / too low, minute ventilation too high / too low, inspired oxygen concentration too high / too low, oxygen concentration too high / too low, respiratory rate too high / too low, end-tidal carbon dioxide partial pressure too high / too low, tidal volume too high / too low, ventilator shutdown, suffocation and increased airway pressure.

[0090] Furthermore, in the physiological parameter data, waveform data includes: electrocardiogram waveform data, respiratory waveform data, volume pulse waveform data and invasive arterial pressure waveform data; numerical data includes: heart rate, blood pressure, blood oxygen saturation and body temperature;

[0091] In the process of labeling the physiological parameter data,

[0092] The morphology of the marked ECG waveform data includes: P wave, QRS complex, T wave, ST segment, PR interval, RR interval and QT interval of each ECG cycle waveform;

[0093] The morphology of the marked respiratory waveform data includes: the peaks and troughs of each respiratory cycle waveform;

[0094] The morphology of the marked volume pulse wave waveform data includes: the peaks and troughs of each pulse cycle waveform;

[0095] The morphology of the marked invasive arterial pressure waveform data includes: the peaks and troughs of each arterial systolic-diastolic cycle waveform;

[0096] Abnormal events marked include: tachycardia, bradycardia, arrhythmia, cardiac arrest, atrial flutter, atrial fibrillation, ventricular flutter, ventricular fibrillation, atrioventricular block, ST segment elevation / depression, high / low heart rate, high / low blood pressure, high / low body temperature and low blood oxygen saturation.

[0097] Further, in the blood gas analysis data, the numerical data include: oxygen partial pressure, carbon dioxide partial pressure, total carbon dioxide, potassium ion concentration, sodium ion concentration, bicarbonate ion concentration and pH value;

[0098] In the process of labeling the blood gas analysis data,

[0099] The marked abnormal events include: oxygen partial pressure too high / too low, carbon dioxide partial pressure too high / too low, total carbon dioxide too high / too low, potassium ion concentration too high / too low, sodium ion concentration too high / too low, and pH too high / too low.

[0100] Further, in the laboratory diagnosis data, the numerical data include: white blood cell count, red blood cell count, platelet count, hemoglobin, hematocrit, blood glucose, prothrombin time and activated partial thromboplastin time;

[0101] In the process of data annotation for experimental diagnostic data,

[0102] The marked abnormal events include: white blood cell count is too high / too low, red blood cell count is too high / too low, hemoglobin is too high / too low, hematocrit is too high / too low, blood sugar is too high / too low, platelet count is too high / too low, prothrombin time is too long / too short, activated partial thromboplastin time is too long / too short.

[0103] Furthermore, the report framework of the mechanical ventilation treatment data analysis report includes a report summary page and a report analysis page; wherein the report summary page is divided into a basic information area, a mechanical ventilation data area and an analysis conclusion area;

[0104] The basic information area includes at least patient information and key time node information; and also includes the following optional items: clinical diagnosis information, device ID for outputting mechanical ventilation treatment data, and mechanical ventilation mode and setting information;

[0105] The mechanical ventilation data area at least includes respiratory analysis results, human-machine asynchrony event analysis results, airway resistance change analysis results, compliance change analysis results and lung injury warning analysis results; each analysis result of the mechanical ventilation data area is obtained based on mechanical ventilation data and its data annotation; the respiratory analysis results include shallow rapid breathing index;

[0106] The analysis conclusion area is used to display the analysis description and conclusion of the mechanical ventilation treatment data;

[0107] The report analysis page is used to display detailed analysis reports of one or more mechanical ventilation treatment data in the form of charts.

[0108] Further, the report analysis page includes a trend chart page; the trend chart page includes:

[0109] Pressure, volume, and ventilation trend analysis area for mechanical ventilation data;

[0110] Respiratory rate trend analysis area;

[0111] Inspired oxygen concentration trend analysis area; and,

[0112] The trend analysis description and conclusion area is used to summarize the trend trends of each trend analysis area and give a trend analysis description and corresponding conclusions;

[0113] The contents of each trend analysis area in the trend graph page are obtained based on mechanical ventilation data and data annotations thereof.

[0114] Further, the report analysis page also includes a data page;

[0115] The data page at least includes:

[0116] Mechanical ventilation setting parameter data subpage, including: mechanical ventilation setting parameter record table, mechanical ventilation ventilation mode and setting parameter change trend chart;

[0117] Mechanical ventilation monitoring parameter data subpage, including mechanical ventilation monitoring parameter record table;

[0118] The contents of the mechanical ventilation setting parameter data subpage and the mechanical ventilation monitoring parameter data subpage are obtained based on the mechanical ventilation data and data annotation thereof;

[0119] The data page also includes the following optional data subpages:

[0120] 24-hour physiological parameter data subpage, including 24-hour vital signs monitoring record sheet, heart rate / respiratory rate / blood pressure trend chart; the content of the 24-hour physiological parameter data subpage is obtained based on the physiological parameter data and its data annotation;

[0121] 24-hour blood gas analysis and laboratory diagnosis data subpage, including 24-hour blood gas analysis monitoring record sheet, oxygen partial pressure / carbon dioxide partial pressure / bicarbonate trend chart, and laboratory diagnosis data sheet;

[0122] The content of the 24-hour blood gas analysis and laboratory diagnosis data subpage is obtained based on the blood gas analysis data, laboratory diagnosis data and data annotation thereof;

[0123] Alarm event data subpage, including: alarm event record table, 24-hour alarm event distribution diagram, alarm event comprehensive analysis table;

[0124] The alarm event data subpage is obtained based on abnormal events marked by mechanical ventilation treatment data;

[0125] The extended image data subpage is used to display the image data.

[0126] Furthermore, the report analysis page also includes a graphics page; the graphics page is divided into a mechanical ventilation waveform data area, a mechanical ventilation loop diagram area, and a physiological parameter waveform data area; wherein,

[0127] The mechanical ventilation waveform data area includes continuous normal mechanical ventilation data waveform graphs and abnormal mechanical ventilation event data waveform graphs;

[0128] The mechanical ventilation loop diagram area includes a pressure-volume loop diagram and a flow-volume loop diagram drawn by combining mechanical ventilation pressure, volume, and flow data;

[0129] The contents of the mechanical ventilation waveform data area and the mechanical ventilation ring diagram area are obtained based on the mechanical ventilation data and data annotation thereof;

[0130] The physiological parameter waveform data area includes a continuous physiological parameter data waveform graph synchronized with the mechanical ventilation data waveform graph;

[0131] The content of the physiological parameter waveform data area is obtained based on the physiological parameter data and its data annotation.

[0132] Further, the report generation instruction is a customized report generation instruction;

[0133] The customized report generation instruction includes one or more customized parameters for generating a report summary page and a report analysis page matching the customized parameters;

[0134] The customized parameters include a trend graph page, a data page or a graphic page.

[0135] Furthermore, during the data annotation process of the data annotation module, when one or more abnormal events of ventilator stopping working, asphyxia, airway pressure increase, cardiac arrest, tachycardia, bradycardia, ventricular flutter, ventricular fibrillation and hypotension occur, an emergency instruction is sent to the clinical data terminal device, and the emergency instruction carries information of the abnormal event that occurred;

[0136] The clinical data terminal device is also used to receive and respond to the emergency instruction.

[0137] Further, in the data annotation process of the data annotation module, when one or more abnormal events of human-machine asynchrony event, too high airway pressure, too low airway pressure, too high positive end-expiratory pressure, too low positive end-expiratory pressure, too high minute ventilation, too low minute ventilation, too high tidal volume, too low tidal volume, too high respiratory rate, too low respiratory rate, too high oxygen concentration, too low oxygen concentration occur, a ventilator parameter adjustment instruction is sent to the clinical data terminal device, and the ventilator parameter adjustment instruction carries information of the abnormal event that occurred;

[0138] The clinical data terminal device is also used to receive and respond to ventilator parameter adjustment instructions.

[0139] Furthermore, in the standard data structure object, a corresponding data structure object is configured for each mechanical ventilation treatment data and patient information;

[0140] Each decoupled mechanical ventilation treatment data and patient information are placed into corresponding data structure objects to form standardized data.

[0141] Furthermore, the application software system further includes a streaming media management module, which is used to receive streaming media data of the treatment scene in real time; the streaming media data is output by the bedside audio and video equipment;

[0142] The database server is further used to partition and store the streaming media data, and associate the streaming media data with the medical data set and the mechanical ventilation treatment data analysis report;

[0143] The clinical data terminal device is also used to broadcast in real time the waveform data, numerical data, treatment scenarios and mechanical ventilation treatment data analysis report in the mechanical ventilation treatment data of the medical data set of some or all patients in one or more wards.

[0144] Furthermore, the system also includes a multi-mode gateway and a master control server; wherein,

[0145] The multi-mode gateway is used to connect ventilator equipment, multi-parameter monitors, bedside audio and video equipment, clinical information systems and clinical data terminal equipment from different manufacturers in various wards of the hospital through the network; the multi-mode gateway is embedded with multiple communication protocols to realize data communication between ventilator equipment, multi-parameter monitors, bedside audio and video equipment, clinical information systems and data acquisition and standardization modules, and to realize data communication between application software systems and clinical data terminal equipment where patients are located;

[0146] The main control server is embedded with a multi-task operating system and an application software system running on the multi-task operating system, and is used to control the operation of the application software system.

[0147] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0148] The mechanical ventilation treatment data management method and system provided by the present invention have the following advantages:

[0149] 1. The present invention provides doctors and nurses with a technical means for centrally managing mechanical ventilation treatment data. Doctors and nurses use clinical data terminal equipment to centrally manage the mechanical ventilation treatment data services of patients in multiple wards in a visual manner over a long distance, provide real-time data monitoring and mechanical ventilation data analysis reports, improve data work efficiency and mechanical ventilation treatment quality, reduce the labor intensity of doctors and nurses, alleviate the problem of ICU human resources shortage, solve the defects of existing mechanical ventilation treatment services, and have wide applicability.

[0150] 2. The present invention decouples the different data communication protocols of equipment from different manufacturers from the data contained therein, separates multi-source heterogeneous data, analyzes and filters the huge multi-source heterogeneous data separately, and reconstructs it into standardized data, effectively reducing the complexity of the analysis process, improving the analysis speed and stability, and solving the problems of the existing medical equipment data communication protocols and data formats being confusing and difficult to process in a unified manner, and has wide applicability.

[0151] 3. The present invention aggregates and analyzes mechanical ventilation treatment data to generate a mechanical ventilation treatment data analysis report. Doctors and nurses can easily obtain summary data sets, quickly and accurately judge the current and past mechanical ventilation treatment status and disease evolution trends, and take timely intervention measures. It solves the difficulties caused by huge multi-source heterogeneous data in clinical applications, reduces the work pressure of doctors and nurses, improves medical quality and efficiency, and has wide applicability.

[0152] 4. The data labeling method provided by the present invention can effectively remove interfering data and erroneous data analyzed by the equipment itself, effectively reducing the data work pressure of doctors and nurses, while greatly improving the data quality, solving the defects of low data quality, poor availability, and lack of waveform data stored in existing equipment and clinical information systems, and has wide applicability.

[0153] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0154] The drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like components throughout the drawings.

[0155] Figure 1 A flow chart of a method for managing mechanical ventilation treatment data provided in Example 1 of the present invention;

[0156] Figure 2 This is an example diagram of a standard data structure object provided in Example 1 of the present invention;

[0157] Figure 3 A schematic diagram of a model input processing process in the process of determining a human-machine asynchronous event provided in Embodiment 1 of the present invention; wherein: Figure 3 (a) shows the schematic diagram of two sliding window processing; Figure 3 (b) Schematic diagram of the time series after two sliding window processing;

[0158] Figure 4 A summary page of the mechanical ventilation treatment data analysis report provided in Example 1 of the present invention;

[0159] Figure 5 The trend page of the mechanical ventilation treatment data analysis report provided in Example 1 of the present invention;

[0160] FIG6 is a data page of a mechanical ventilation treatment data analysis report provided in Example 1 of the present invention; wherein FIG6(a) represents a mechanical ventilation setting parameter data subpage; FIG6(b) represents a mechanical ventilation monitoring parameter data subpage; FIG6(c) represents a 24-hour physiological parameter data subpage; FIG6(d) represents a 24-hour blood gas analysis and laboratory diagnosis data subpage; FIG6(e) represents an alarm event data subpage; and FIG6(f) represents an imaging data subpage;

[0161] Figure 7 A graphic page of a mechanical ventilation treatment data analysis report provided in Example 1 of the present invention;

[0162] Figure 8 This is an architecture diagram of a mechanical ventilation therapy data management system that aggregates data from multiple devices provided in Example 2 of the present invention;

[0163] Fig. 9 A visualization method for displaying a patient implementation example provided in Embodiment 2 of the present invention;

[0164] Fig.10 An implementation example of a ward management panel of a clinical data terminal device provided in Embodiment 2 of the present invention;

[0165] Fig.11 An implementation example of a data analysis report panel of a clinical data terminal device provided in Embodiment 2 of the present invention;

[0166] Fig.12 This is an architecture diagram of a mechanical ventilation therapy data management system in the form of an edge computing node provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0167] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0168] Example 1

[0169] Embodiment 1 of the present invention provides a method for managing mechanical ventilation treatment data, the flow chart of which is as follows: Figure 1 As shown, the following steps are included:

[0170] Step S1: placing patient information and mechanical ventilation treatment data received in real time into standard data structure objects respectively to form standardized data; the mechanical ventilation treatment data includes mechanical ventilation data;

[0171] In step S1, the mechanical ventilation treatment data also includes one or more of the following: physiological parameter data, blood gas analysis data, laboratory diagnosis data, imaging data, clinical diagnosis information, and device ID for outputting mechanical ventilation treatment data; wherein the physiological parameter data is output by a multi-parameter monitor; patient information, blood gas analysis data, laboratory diagnosis data, imaging data, and clinical diagnosis information are all output by a clinical information system; and mechanical ventilation data is output by a ventilator device. Exemplarily, the clinical information system includes at least one of a HIS, a CIS, and an EMR system.

[0172] In the specific implementation process, in order to solve the problem that the data output by similar equipment from different manufacturers and different types of mechanical ventilation treatment data acquisition equipment are not unified, which causes the subsequent data processing process to be cumbersome, this embodiment sets a standard data structure object to standardize the multi-source heterogeneous data output by the above-mentioned equipment. Specifically, in this embodiment, different data communication protocols of equipment from different manufacturers are received and processed in parallel, and data messages from ventilators, multi-parameter monitors and clinical information systems are synchronously collected, the head and tail structures and data parts in the data messages are decoupled and separated, the data parts are parsed in parallel, and valid data is extracted from them, and filled into the corresponding positions of the preset standard data structure objects to form standardized data, which is bundled and packaged with patient information, set a unique identifier, and stored in the database; the standard data structure object is an abstraction of mechanical ventilation treatment data and patient information, which simplifies complex multi-source heterogeneous data, thereby improving the efficiency of data analysis, processing and utilization, and reducing data application costs. Therefore, this embodiment can complete the standardization process of multi-source heterogeneous data by performing data analysis and standardization processing, reduce the complexity of multi-source heterogeneous data analysis, and improve data analysis speed and stability. Standardized data not only saves storage space, but also improves the efficiency of data calculation and retrieval.

[0173] Figure 2 An example of a standard data structure object 201 is shown, including a mechanical ventilation data object 202, a physiological parameter data object 203, a blood gas analysis data object 204, a laboratory diagnosis data object 205, an imaging data object 206, a clinical diagnosis information object 207 and a patient information object 208. The specific content of each data structure object can be preset according to the actual situation. In addition, Figure 2 The metadata information object 209 is also included, which is used to describe the structure, content and attributes of the standard data structure object 201 itself and the data structure objects it contains, and define the name, relationship, field, constraint and syntactic structure of each data structure object. Since blood gas analysis data, laboratory diagnosis data and imaging data are all periodic data, which occur intermittently or regularly during mechanical ventilation treatment, the standard data structure object 201 allows the relevant data to be set to null value when it does not appear. In addition, in order to improve the adaptability and extensibility of the standard data structure object 201, Figure 2 The standard data structure object 201 includes extended objects 1-n as shown, which are used to add other related mechanical ventilation treatment data. For example, for shock patients receiving mechanical ventilation treatment, CRRT blood filtration and urine volume dynamic monitoring need to be performed simultaneously. At this time, the mechanical ventilation treatment data can also include continuous blood filtration data and urine volume dynamic monitoring data. Accordingly, the standard data structure object 201 can expand the continuous blood filtration data object and the urine volume dynamic monitoring data object, so as to facilitate the addition of other related mechanical ventilation treatment data. The standard data structure object provided in this embodiment has better solved the defects of the existing medical equipment data communication protocol and data format being confusing and difficult to be uniformly processed, and has wide applicability. Specifically, the standardized data structure object can be implemented by a high-level language. Taking the object-oriented programming language JAVA as an example, the standardized data structure object is a specific class (class), the attributes (field) of the class are various data structure objects, and various methods (method) functions of the class are written for external calls, which are used to set specific data structure objects and the parameters contained therein, for example, airway resistance and respiratory work represented by floating point types, volume waveform data represented by linear tables, and Boolean attribute identifiers. The standardized data structure object class is distributed to the parsing programs. When parsing in parallel, each parsing program creates a standardized data structure class object (Object) for each data packet, calls the class method to fill in the data, and merges the standardized data structure class objects at the same time into one standardized data through the aggregation program.

[0174] Step S2: performing data cleaning on the standardized data to obtain a medical data set;

[0175] Data cleaning is the process of re-examining and verifying data, with the aim of deleting duplicate information, correcting existing errors, and providing data consistency. Existing data cleaning technology is relatively mature. In actual implementation, the corresponding data cleaning method can be selected according to the specific application scenario and the specific situation of mechanical ventilation treatment data in the application scenario. For example, the data cleaning process in this embodiment may include the following operations:

[0176] (1) Data constraint range judgment: range constraints of medical theory data values ​​and data constraints of empirical inference;

[0177] (2) Data unit / data precision conversion;

[0178] (3) Elimination of data outliers;

[0179] (4) For waveform data, the data length is calibrated and the data integrity is verified;

[0180] The medical data set obtained after data cleaning can be used for subsequent data analysis.

[0181] Step S3: data labeling is performed on the waveform and numerical data in the mechanical ventilation treatment data; the data labeling includes: labeling the amplitude, frequency and shape of the waveform data, labeling the measurement value of the numerical data, and labeling abnormal events; wherein the measurement value of the labeled numerical data may include a parameter higher than an upper threshold of the corresponding numerical data, or lower than a lower threshold of the corresponding numerical data; the labeled data is automatically stored in the medical data set;

[0182] Since the mechanical ventilation treatment data may include a lot of specific data, it is impossible to list them all. In this embodiment, only the common data in each mechanical ventilation treatment data and the common data annotations are highlighted. Specifically,

[0183] (1) In the mechanical ventilation data, the waveform data includes: mechanical ventilation pressure waveform data, volume waveform data, flow waveform data and end-tidal carbon dioxide waveform data; the numerical data includes: airway resistance, work of breathing, compliance, ventilation parameters and respiratory rate; in the process of data labeling of mechanical ventilation data, the form of the labeled waveform data includes: the start of the inspiratory phase, the inspiratory phase, the end of the inspiratory phase, the start of the expiratory phase, the expiratory phase and the end of the expiratory phase, the airway peak pressure, the expiratory peak flow rate of each mechanical ventilation waveform, and the inspiratory baseline, the expiratory platform height, the start and end of the expiratory platform of the end-tidal carbon dioxide waveform; the labeled content can be used directly or after statistics, calculation and analysis to generate a report summary page and a report analysis page; illustratively, the airway peak pressure and the expiratory peak flow rate can be determined by the following process:

[0184] (a) First, common filtering algorithms such as median / homogeneous filtering are used to filter and aggregate the high-sampling rate mechanical ventilation pressure, volume and flow waveform data to reduce the complexity of waveform processing. The pressure-time curve of the mechanical ventilation pressure waveform data is denoted as P(t), the volume-time curve of the mechanical ventilation volume waveform data is denoted as V(t), and the flow-time curve of the mechanical ventilation flow waveform data is denoted as F(t); an observation window is set. Exemplarily, two breaths require at least 3.428s, so the observation window duration is greater than 3.428s. Exemplarily, the observation window can be set to 4s. Steps (b)-(d) are all calculated based on this observation window;

[0185] (b) In the i-th observation window, the volume-time curve V(t) is processed as follows:

[0186] Use the peak finding algorithm to find the peaks and troughs in the observation window and store the results, and obtain:

[0187] The time when the i-th inspiration starts End time of the i-th inspiration It is also the start time of the i-th exhalation;

[0188] End time of the i-th exhalation It is also the start time of the i+1th inhalation;

[0189] (c) According to the start and end time nodes of the above-mentioned inhalation phase and exhalation phase, the pressure-time curve P(t) is calculated as follows to obtain the peak airway pressure PIP and store the result; the i-th peak airway pressure t represents time;

[0190] (d) Based on the start and end time nodes of the above-mentioned inspiratory phase and expiratory phase, the flow-time curve F(t) is calculated as follows to obtain the peak expiratory flow rate PEF and store the result; the peak expiratory flow rate PEF of the i-th time

[0191] The measured values ​​of the marked numerical data include: parameters that are higher than the upper threshold of the corresponding numerical data, or lower than the lower threshold of the corresponding numerical data; the marked content can be used directly or through statistics, calculation, and analysis for trend chart pages, mechanical ventilation setting parameter subpages, and mechanical ventilation monitoring parameter subpages; illustratively, the compliance data and its corresponding upper and lower thresholds are continuously compared numerically. When the compliance data is lower than the lower threshold (for example, the lower threshold is 25 ml / cmH2O), the compliance data and occurrence time at that moment are marked and stored.

[0192] The marked abnormal events include: human-machine asynchrony events, airway pressure too high / too low, positive end-expiratory pressure too high / too low, exhaled tidal volume too high / too low, minute ventilation too high / too low, inspired oxygen concentration too high / too low, oxygen concentration too high / too low, respiratory rate too high / too low, end-tidal carbon dioxide partial pressure too high / too low, tidal volume too high / too low, ventilator stop working, suffocation, airway pressure increase. The marked content can be used directly or after statistics, calculation, and analysis, for report summary pages, alarm event data subpages, and graphic pages.

[0193] Exemplarily, the human-machine asynchrony event may be determined based on the neural network model, and the specific process is described as follows:

[0194] (a) Construct model input based on pressure-time curve P(t), volume-time curve V(t), and flow-time curve F(t);

[0195] Specifically, the pressure-time curve P(t), volume-time curve V(t), and flow-time curve F(t) are combined into a three-dimensional time series as the model input: T = <S1,…,S c ,…,S C >, where S c represents the input vector of the cth sampling point, and C represents the total number of sampling points processed by the neural network model at a time, that is, the model input length; They respectively represent the flow rate, pressure and volume values ​​of the cth sampling point.

[0196] (b) Perform two sliding window processes to enrich features and obtain the time series after sliding window processing;

[0197] Two sliding window processing: Considering the inherent dependency of time series data, this embodiment uses two sliding windows to process the sequence data in blocks, extract the statistical features of each window, eliminate interference factors, enhance the time series feature law, better describe the central tendency and discrete degree of the distribution of waveform sampling point data values, the skewness and kurtosis characteristics of the waveform morphology change, and facilitate the subsequent capture of the dependency characteristics and local characteristics of the input sequence. The schematic diagram of the two sliding window processing is shown in Figure 3 (a). Among them,

[0198] The first sliding window processing is performed:

[0199] The model input T is processed by sliding window with b / 2 as the sliding window step length, where b represents the width of the first sliding window. After the first sliding window processing, the model input length changes from C to C′, and the value range of c changes to [1, C′].

[0200] After the first sliding window processing, based on the l-th type of waveform data in all sampling points in the c-th sliding window, the Euclidean distance of the l-th type of waveform data in the Euclidean matrix vector is calculated. Right now,

[0201] l is 1, 2, 3 respectively;

[0202] Finally, we get the Euclidean matrix vector

[0203] Based on the l-th type of waveform data in all sampling points in the c-th sliding window, the distance difference of the l-th type of waveform data in the distance difference vector is calculated. Right now,

[0204]

[0205] The final distance difference vector is expressed as

[0206] Therefore, the time series T′ after the first sliding window processing is <G1,…,G c ,…,G C′ >, where G c = <NOR c ,DON c >.

[0207] The second sliding window processing is performed:

[0208] The time series T′ is processed again with sliding window step length f / 2, where f represents the second sliding window width. After the second sliding window processing, the model input length changes from C′ to C″, and the value range of c changes to [1, C″];

[0209] During the second sliding window processing, the derived features of all Euclidean matrix vectors and distance difference vectors in the second sliding window can be obtained. Taking the c-th sliding window as an example, the following can be obtained:

[0210] Euclidean matrix and the average vector of distance differences (NOR c_Mean ,DON c_Mean ),

[0211] Euclidean matrix and maximum vector of distance difference (NOR c_Max ,DON c_Max ),

[0212] Euclidean matrix and the minimum vector of distance difference (NOR c_Min ,DON c_Min ),

[0213] The upper quartile vector of the Euclidean matrix and the distance difference (NORc_25Q ,DON c_25Q ),

[0214] Euclidean matrix and median vector of distance difference (NOR c_50Q ,DON c_50Q ),

[0215] The lower quartile vector of the Euclidean matrix and the distance difference (NOR c_75Q ,DON c_75Q ),

[0216] Euclidean matrix and standard deviation vector of distance difference (NOR c_Std ,DON c_Std ),

[0217] Euclidean matrix and deviation vector of distance difference (NOR c_p2p ,DON c_p2p ),

[0218] By calculating the Euclidean distance of the lth type of waveform data in the Euclidean matrix vector within the cth sliding window By averaging, we can get NOR c_Mean The first result in therefore,

[0219]

[0220] By calculating the distance difference of the lth type of waveform data in the distance difference vector within the cth sliding window By averaging, we can get DON c _ Mean The first result in therefore,

[0221]

[0222] Other derived features after the second sliding window processing can be identified based on their feature names. The corresponding calculation process is performed and will not be repeated here.

[0223] Therefore, the time series T″ after the second sliding window processing is c ,…,H C ″>, where

[0224]

[0225] The schematic diagram of the time series after two sliding window processing is as follows Figure 3 (b) as shown.

[0226] ​(c) Inputting the time series after sliding window processing into the neural network model, the neural network model outputs normal events and ventilator abnormal events (specifically the type of human-machine asynchrony events and their occurrence time) after processing; the types of ventilator abnormal events include invalid triggering, repeated triggering, automatic triggering, reverse triggering, flow rate asynchrony, overshoot, delayed switching and early switching;

[0227] Based on the output category results of the neural network model, the total number of human-computer asynchrony and the human-computer asynchrony index can also be calculated. These two items can be calculated according to the existing methods and will not be repeated here.

[0228] (2) In the physiological parameter data, waveform data includes: ECG waveform data, respiratory waveform data, volume pulse waveform data, invasive arterial pressure waveform data; numerical data includes: heart rate, blood pressure, blood oxygen saturation and body temperature; in the process of data labeling of physiological parameter data,

[0229] The forms of the marked ECG waveform data include: P wave, QRS complex, T wave, ST segment, PR interval, RR interval, QT interval of each ECG cycle waveform; the forms of the marked respiratory waveform data include: the peak and trough of each respiratory cycle waveform; the forms of the marked volume pulse waveform data include: the peak and trough of each pulse cycle waveform; the forms of the marked invasive arterial pressure waveform data include: the peak and trough of each arterial contraction-diastole cycle waveform; the marked content can be used in the report summary page and analysis page directly or after statistics, calculation and analysis;

[0230] The measured values ​​of the marked numerical data include: parameters that are higher than the threshold value or lower than the threshold value of the corresponding numerical data; the marked content can be directly used or used in the 24-hour physiological parameter data subpage after statistics, calculation and analysis;

[0231] The marked abnormal events include: tachycardia, bradycardia, arrhythmia, cardiac arrest, atrial flutter, atrial fibrillation, ventricular flutter, ventricular fibrillation, atrioventricular block, ST segment elevation / depression, high / low heart rate, high / low blood pressure, high / low body temperature, low blood oxygen saturation. The marked content can be used directly or after statistics, calculation, and analysis, for report summary page, analysis page, and alarm event data subpage.

[0232] (3) In the blood gas analysis data, the numerical data include: oxygen partial pressure, carbon dioxide partial pressure, total carbon dioxide, potassium ion concentration, sodium ion concentration, bicarbonate ion concentration and pH value;

[0233] In the process of data annotation of blood gas analysis data,

[0234] The measured values ​​of the marked numerical data include: parameters that are higher than or lower than the threshold of the corresponding numerical data; the marked content can be used directly or after statistics, calculation, and analysis for the 24-hour blood gas analysis and laboratory diagnosis data subpages;

[0235] The marked abnormal events include: excessively high / low oxygen partial pressure, excessively high / low carbon dioxide partial pressure, excessively high / low total carbon dioxide, excessively high / low potassium ion concentration, excessively high / low sodium ion concentration, excessively high / low bicarbonate ion concentration, and excessively high / low pH value. The marked contents can be used directly or after statistics, calculation, and analysis for 24-hour blood gas analysis and laboratory diagnosis data subpages.

[0236] (4) In laboratory diagnostic data, numerical data include: white blood cell count, red blood cell count, platelet count, hemoglobin, hematocrit, blood glucose, prothrombin time and activated partial thromboplastin time;

[0237] In the process of data annotation for experimental diagnostic data,

[0238] The marked numerical data include: parameters that are higher than the corresponding numerical data threshold or lower than the threshold; the marked content can be used directly or after statistics, calculation, analysis, for 24-hour blood gas analysis and laboratory diagnosis data subpages;

[0239] The marked abnormal events include: white blood cell count is too high / too low, red blood cell count is too high / too low, hemoglobin is too high / too low, hematocrit is too high / too low, blood sugar is too high / too low, platelet count is too high / too low, prothrombin time is too long / too short, activated partial thromboplastin time is too long / too short. The marked content can be used directly or after statistics, calculation, analysis, for 24-hour blood gas analysis and laboratory diagnosis data subpages;

[0240] Other unfinished mechanical ventilation treatment data and data annotations thereof are also easily conceivable by those skilled in the art based on the present embodiment, and therefore, also fall within the protection scope of this application.

[0241] Step S4: When a report generation instruction is received, a mechanical ventilation treatment data analysis report is generated based on the medical data set carrying the data annotations.

[0242] Exemplarily, in step S4, a medical data set with data annotations is retrieved, multi-dimensional data aggregation analysis is performed, and analysis conclusions based on correlation are generated. At the same time, various types of data are combined with time labels and dimension labels, and filled into a preset data analysis report framework to generate a mechanical ventilation treatment data analysis report and store it in a database. In addition, in order to further ensure the accuracy of the mechanical ventilation treatment data analysis report, a manual review operation can be added, and the mechanical ventilation treatment data analysis report after manual review is stored in the database.

[0243] It should be noted that various types of data in the medical data set are calculated, analyzed, and counted layer by layer to obtain data that meets the required content and accuracy requirements in the preset framework of the analysis report, which is used to generate a mechanical ventilation treatment data analysis report.

[0244] The report framework of the mechanical ventilation treatment data analysis report includes a report summary page and a report analysis page; wherein, Figure 4 As shown, the report summary page 400 is divided into a basic information area 401, a mechanical ventilation data area 402, and an analysis conclusion area 403;

[0245] The basic information area 401 includes at least patient information and key time node information; it also includes the following optional items: clinical diagnosis information, device ID for outputting mechanical ventilation treatment data, and mechanical ventilation mode and setting information; illustratively, patient information may include hospital information, patient basic information, ward bed information, hospital department, etc.; key time node information may include report time, data start time, data end time, device startup time, and device weaning time, etc.

[0246] The mechanical ventilation data area 402 at least includes respiratory analysis results, human-machine asynchrony event analysis results, airway resistance change analysis results, compliance change analysis results and lung injury warning analysis results; various analysis results are obtained based on mechanical ventilation data and data annotation; illustratively, the respiratory analysis results include: total respiratory cycle, spontaneous respiratory frequency ratio of spontaneous respiratory cycle and shallow rapid breathing index;

[0247] Specifically, in this embodiment, the total respiratory cycle, spontaneous respiratory cycle and tidal volume annotated with the mechanical ventilation waveform data are statistically calculated to obtain the total number of respiratory cycles, the number of spontaneous respiratory cycles, and the proportion of spontaneous respiratory frequency. At the same time, the shallow rapid breathing index can also be obtained to evaluate the possibility of patient weaning. The data annotated with mechanical ventilation human-machine asynchrony events are statistically calculated to obtain the total number of human-machine asynchrony events, and the frequency and proportion of each asynchrony event are statistically calculated. The compliance data annotated with the mechanical ventilation numerical data are calculated to obtain the value and occurrence time of the compliance reduction, and the number of compliance reductions, cumulative duration and longest reduction duration within the preset time range are statistically calculated. Figure 4 Examples of analysis results in the mechanical ventilation data area are given. These examples are indicator data that clinicians are more concerned about and have high clinical value.

[0248] Figure 4The analysis conclusion area 403 in the figure is used to display the mechanical ventilation treatment data analysis description and conclusion; wherein, the analysis conclusion area may include mechanical ventilation monitoring time measurement conclusion, respiratory analysis conclusion (including "shallow rapid breathing index"), human-machine asynchrony event analysis conclusion, airway resistance change analysis conclusion, compliance change analysis conclusion, and lung injury warning analysis conclusion. Specifically, various types of data in the mechanical ventilation data area are statistically analyzed to obtain two types of data: normal interval and abnormal interval; events are stratified according to the number of occurrences, occurrence time, duration, and type and risk level of abnormal interval events, and are associated with corresponding analysis conclusion terms to describe the patient's mechanical ventilation treatment status, generate analysis conclusions, and fill in the preset data analysis report framework. Exemplarily, the compliance change analysis conclusion is based on the marked compliance reduction event data, and the number of events that meet the preset compliance reduction abnormal interval (<25ml / cmH2O) is counted, and the duration of all compliance change abnormal events is compared to obtain the longest lasting compliance reduction event. Finally, according to the number of compliance reduction events, the longest lasting compliance reduction event and the occurrence time, an analysis conclusion is generated and filled into the analysis conclusion area 403 .

[0249] The report analysis page is used to display detailed analysis reports of one or more mechanical ventilation treatment data in the form of charts.

[0250] The report analysis page includes a trend graph page 500, such as Figure 5 As shown; the trend chart page includes: a pressure trend analysis area 501, a volume trend analysis area 502, a ventilation trend analysis area 503, a respiratory rate trend analysis area 504, an inspired oxygen concentration trend analysis area 505, and a trend analysis description and conclusion area 506 for mechanical ventilation data. Among them, the trend analysis description and conclusion area 506 is used to synthesize the trend trends of each trend analysis area to obtain a trend analysis description and corresponding conclusions. In addition, based on the mechanical ventilation data and data annotation, the contents of each trend analysis area are obtained, divided into 24 hourly data units by hour, and the average value of the relevant parameters reported in each unit is calculated. The average value of the hourly parameters is plotted and connected to draw a trend chart.

[0251] The report analysis page also includes a data page, as shown in FIG6 ; the data page includes at least: a mechanical ventilation setting parameter data subpage 610, a mechanical ventilation monitoring parameter data subpage 620; the data page also includes the following optional data subpages: a 24-hour physiological parameter data subpage 630, a 24-hour blood gas analysis and laboratory diagnosis data subpage 640, and an alarm event data subpage 650; wherein,

[0252] The mechanical ventilation setting parameter data subpage 610, as shown in FIG6(a), includes a mechanical ventilation setting parameter record table 611, a mechanical ventilation ventilation mode and setting parameter change trend chart 612;

[0253] The mechanical ventilation monitoring parameter data subpage 620, as shown in FIG6(b), includes a mechanical ventilation monitoring parameter record table 621; based on the mechanical ventilation data and data annotation, the contents of the mechanical ventilation setting parameter data subpage and the mechanical ventilation monitoring parameter data subpage 620 are obtained;

[0254] 24-hour physiological parameter data subpage 630, as shown in FIG6(c); includes 24-hour vital sign monitoring record table 631, heart rate / respiratory rate / blood pressure trend chart 632; based on the physiological parameter data and data annotation, the content of the 24-hour physiological parameter data subpage 630 is obtained;

[0255] 24-hour blood gas analysis and laboratory diagnosis data subpage 640, as shown in FIG6(d), includes a 24-hour blood gas analysis monitoring record table 641, an oxygen partial pressure / carbon dioxide partial pressure / bicarbonate trend chart 642, and a laboratory diagnosis data table 643; based on the blood gas analysis data, the laboratory diagnosis data and the data annotation, the content of the 24-hour blood gas analysis and laboratory diagnosis data subpage 640 is obtained;

[0256] The alarm event data subpage 650, as shown in FIG6(e), includes an alarm event record table 651, a 24-hour alarm event distribution diagram 652, and an alarm event comprehensive analysis table 653. The content of the alarm event data subpage 650 is obtained based on the abnormal events marked in the mechanical ventilation treatment data.

[0257] The extended image data subpage 660, as shown in FIG6(e), is used to display the image data. For example, the image data may include an ultrasound image inspection area 661, a CT image inspection area 662, and a DR image inspection area 663.

[0258] The report analysis page also includes a graphics page 700, such as Figure 7 As shown; the graphic page is divided into a mechanical ventilation waveform data area 701, a physiological parameter waveform data area 702, and a mechanical ventilation loop diagram area 703; wherein,

[0259] The mechanical ventilation waveform data area 701 includes continuous normal mechanical ventilation data waveforms and abnormal mechanical ventilation event data waveforms, and is accompanied by waveform measurement values, marks for each waveform cycle, and marked abnormal event types;

[0260] The physiological parameter waveform data area 702 includes a continuous physiological parameter data waveform diagram synchronized with the mechanical ventilation data waveform diagram, and is accompanied by a waveform measurement value, a mark of each waveform cycle, and a marked abnormal event type. Based on the physiological parameter data and its data annotation, the content of the physiological parameter waveform data area is obtained.

[0261] The mechanical ventilation loop diagram area 703 includes a pressure-volume loop diagram and a flow-volume loop diagram drawn by combining the pressure, volume and flow data during the mechanical ventilation event cycle, and highlights abnormal loop diagrams in an overlay manner; based on the mechanical ventilation data and its data annotations, the contents of the mechanical ventilation waveform data area and the mechanical ventilation loop diagram area are obtained; illustratively, the mechanical ventilation data is used to draw a normal mechanical ventilation data waveform diagram, an abnormal mechanical ventilation event data waveform diagram, and a mechanical ventilation loop diagram, and the loop diagrams formed by each mechanical ventilation are overlaid to highlight the loop diagrams with changed morphology, and mark the occurrence time and duration.

[0262] The above-mentioned graphic page can be arranged in the order of the mechanical ventilation waveform data area and the physiological parameter waveform data area; the arrangement order of the mechanical ventilation waveform data and the physiological parameter waveform data can also be reorganized according to the clinician's data usage habits and analytical thinking mode.

[0263] It should be emphasized that the mechanical ventilation treatment data analysis report is a multi-category, multi-time zone summary data set, which provides summary analysis and conclusions of key indicator data of mechanical ventilation treatment, as well as various key indicator data and abnormal event data, evolution trends, including physiological parameter data, blood gas analysis and laboratory diagnosis data during treatment; it also provides waveform graphic data reflecting the functional status of the ventilator, the state of human / machine interaction, the state of respiratory mechanics, the state of physiological parameters, and the state of various events; it can also provide lung, heart, and abdominal diaphragm imaging data. According to the mechanical ventilation treatment data analysis report, doctors and nurses can quickly grasp the comprehensive mechanical ventilation treatment data, disease status and evolution trends, evaluate the effectiveness of treatment plans, and provide timely intervention measures to reduce complications such as lung injury, improve data processing efficiency and mechanical ventilation treatment quality, and promote patient recovery. It has wide applicability.

[0264] During actual application, in order to expand the application scenarios of the solutions in this embodiment, the report generation instruction can be a customized report generation instruction; the customized report generation instruction includes one or more customized parameters for generating a report summary page, and a report analysis page matching the customized parameters; the customized parameters include a trend chart page, a data page or a graphics page. For example, when the customized parameters only include a trend chart page, the report analysis page in the generated mechanical ventilation treatment data analysis report also only includes a trend chart page, and no longer includes a data page and a graphics page. The generation method of each content in the mechanical ventilation treatment data analysis report can refer to the above description and will not be repeated here.

[0265] In addition, the method further comprises:

[0266] Step S5: During the data labeling process, when one or more abnormal events such as ventilator cessation, suffocation, increased airway pressure, cardiac arrest, tachycardia, bradycardia, ventricular fibrillation, ventricular flutter, and hypotension (shock) occur, an emergency instruction is issued, and the emergency instruction carries information about the abnormal events that occurred.

[0267] Step S6: The method further comprises:

[0268] During the data labeling process, when one or more abnormal events such as human-machine asynchrony, high airway pressure, low airway pressure, high positive end-expiratory pressure, low positive end-expiratory pressure, high minute ventilation, low minute ventilation, high tidal volume, low tidal volume, high respiratory rate, low respiratory rate, high oxygen concentration, and low oxygen concentration occur, a ventilator parameter adjustment instruction is issued, and the ventilator parameter adjustment instruction carries information about the abnormal events that have occurred.

[0269] In addition, step S7: the method further comprises:

[0270] While receiving mechanical ventilation treatment data in real time, streaming media data of treatment scenes are also received in real time (for example, streaming media data can be collected through bedside audio and video equipment); waveform data, numerical data, treatment scenes and mechanical ventilation treatment data analysis reports in the mechanical ventilation treatment data of the medical data set of some or all patients in one or more wards are broadcast live in real time. For example, the clinical data terminal device can broadcast the mechanical ventilation waveform data and numerical data and streaming media data of the medical data set. When the data comes from multiple patients in multiple wards, the patient data and treatment scenes of multiple wards are displayed in a visual manner; in addition, the streaming media data can be associated with the medical data set and all stored in the database.

[0271] The mechanical ventilation treatment data management method provided in this embodiment enables doctors and nurses to centrally manage the mechanical ventilation treatment data services of multiple wards in a visual manner over a long distance, process a large amount of heterogeneous data generated by multiple devices into standard data structure data, and generate a mechanical ventilation treatment data analysis report. On a clinical data terminal device, it is possible to observe and analyze the real-time data, event data, data analysis reports, and treatment scenarios of mechanical ventilation treatment of patients in multiple wards, including experimental diagnosis data and imaging data from the clinical information system, grasp the current and past mechanical ventilation treatment status and disease evolution trend, detect complications at an early stage, and provide timely intervention measures, thereby improving the quality of mechanical ventilation treatment and work efficiency, reducing the labor intensity of doctors and nurses, and solving the difficulties caused by huge multi-source heterogeneous data in clinical applications. The method has wide applicability.

[0272] Example 2

[0273] The embodiment of the present invention discloses a mechanical ventilation treatment data management system, such as Figure 8 As shown, the system includes an application software system 801, a clinical data terminal device 802 and a database server 803; wherein,

[0274] The application software system 801 includes a data collection and standardization module, a data cleaning module, a data annotation module and a report generation module; specifically,

[0275] A data acquisition and standardization module, used to place patient information and mechanical ventilation treatment data received in real time into standard data structure objects to form standardized data; the mechanical ventilation treatment data includes mechanical ventilation data;

[0276] A data cleaning module, used for cleaning the standardized data to obtain a medical data set;

[0277] A data annotation module, used to annotate the mechanical ventilation treatment data in the medical data set; the data annotation includes: marking the amplitude, frequency and shape of the waveform data, marking the measurement value of the numerical data, and marking abnormal events;

[0278] A report generation module, for generating a mechanical ventilation treatment data analysis report based on a medical data set carrying data annotations when a report generation instruction is received;

[0279] The clinical data terminal device 802 is used to issue the report generation instruction and receive and display the mechanical ventilation treatment data analysis report;

[0280] The database server 803 is used to store the standardized data, medical data set and mechanical ventilation treatment data analysis report in partitions.

[0281] In Example 1, other data included in the mechanical ventilation treatment data, the data annotation process and the mechanical ventilation treatment data analysis report generation process have been described in detail, which will not be repeated here.

[0282] In addition, in this embodiment, during the data annotation process of the data annotation module, when one or more abnormal events such as ventilator cessation, asphyxia, airway pressure increase, cardiac arrest, tachycardia, bradycardia, ventricular flutter, ventricular fibrillation and hypotension occur, an emergency instruction is issued to the clinical data terminal device, and the emergency instruction carries information about the abnormal event that occurred; at this time, the clinical data terminal device 802 is also used to receive and respond to the emergency instruction. Exemplarily, the clinical data terminal device 802 can respond to the emergency instruction by issuing an abnormal alarm to attract the attention of medical staff, and can also display information about the abnormal event through the clinical data terminal device.

[0283] During the data annotation process of the data annotation module, when one or more abnormal events occur, including human-machine asynchrony events, high airway pressure, low airway pressure, high positive end-expiratory pressure, low positive end-expiratory pressure, high minute ventilation, low minute ventilation, high tidal volume, low tidal volume, high respiratory rate, low respiratory rate, high oxygen concentration, and low oxygen concentration, a ventilator parameter adjustment instruction is sent to the clinical data terminal device 802, and the ventilator parameter adjustment instruction carries information about the abnormal event that has occurred; at this time, the clinical data terminal device 802 is also used to receive and respond to the ventilator parameter adjustment instruction. The response to the ventilator parameter adjustment instruction may be to issue an abnormal alarm to attract the attention of medical staff. The information of the abnormal event may also be displayed through the clinical data terminal device 802, and suggestions for ventilator parameter adjustment methods for the current abnormal event may also be given.

[0284] The application software system 801 also includes a streaming media management module, which is used to receive streaming media data of treatment scenes in real time through bedside audio and video equipment; the streaming media management module can also manage the operation of bedside audio and video equipment, including streaming media business management, equipment scheduling, consultation management, ward rounds and medical management, as well as live broadcast, storage, review, and retrieval of streaming media data. At this time, the database server 803 is also used to partition and store the streaming media data, and associate the streaming media data with the medical data set and mechanical ventilation treatment data analysis report; the clinical data terminal device 802 is also used to broadcast in real time the waveform data, numerical data, treatment scenes, and mechanical ventilation treatment data analysis report in the mechanical ventilation treatment data of the medical data set of some or all patients in one or more wards.

[0285] like Figure 8 As shown, the mechanical ventilation treatment data management system may also include a multi-mode gateway 804 and a main control server 805; wherein the multi-mode gateway is used to connect ventilator equipment 807, multi-parameter monitor 808, bedside audio and video equipment 809, clinical information system 810 and clinical data terminal equipment 802 of different manufacturers in various wards of the hospital through the network; the multi-mode gateway 804 is embedded with multiple communication protocols, including at least TCP / IP, UDP, RS232 / 485, Restful-API, WiFi, MQTT, HL7, for realizing data communication between the ventilator equipment 807, multi-parameter monitor 808, bedside audio and video equipment 809, clinical information system 810 and the data acquisition and standardization module, and for realizing data communication between the application software system 801 and the clinical data terminal equipment 802. The main control server 805 is embedded with a multi-tasking operating system and an application software system 801 running on the operating system, for controlling the operation of the application software system 801.

[0286] The database server 803 is composed of a database cluster composed of multiple servers, which is used for the system to store various types of data and files. In addition, real-time analysis and processing, multi-dimensional search and query, multi-user network operation, and database disaster recovery management can also be realized. It should be noted that the database server 803 can use multiple servers to form a database cluster, one of which is set as the main node, and the rest are set as sub-nodes. As copies of the main node, the service pressure of the main node server is diverted. When an abnormality occurs in the main node during operation, the sub-node can be adaptively switched to the main node, which improves the security and high availability of the database, meets the storage of massive mechanical ventilation treatment data, data analysis report files, and streaming media data of the system, and the database cluster can be more conveniently dynamically expanded or migrated to the cloud.

[0287] The clinical data terminal device 802 includes at least a desktop computer and a desktop device with computing functions, which is used for data communication with the system and displaying the patient's mechanical ventilation treatment data and status; common clinical data terminal devices such as doctors and nurses' handheld terminal devices 811 include at least a laptop computer and a smart phone, which display the patient's mechanical ventilation treatment data and status.

[0288] The bedside audio and video equipment 809 includes at least video monitoring equipment and an audio and video mobile vehicle, which is used to display the patient treatment scene on the screen of the clinical data terminal device.

[0289] In this embodiment, the modules in the application software system 801 cooperate with each other to jointly complete functions such as data collection, data cleaning, data annotation and report generation, and can realize multi-dimensional analysis and comprehensive judgment of the patient's mechanical ventilation treatment process, thereby predicting the disease evolution trend, facilitating doctors to formulate intervention measures, effectively improving the quality and work efficiency of mechanical ventilation treatment, and reducing the labor intensity of doctors and nurses. The screen of the clinical data terminal device 802 can be a single screen or a combined screen. Doctors and nurses can hold the data terminal device 811 to seamlessly monitor and manage patients undergoing mechanical ventilation treatment in any occasion.

[0290] Preferably, in this embodiment Fig. 9 An implementation example of displaying patients in a visual manner is given, wherein a clinical data terminal device 901 receives mechanical ventilation treatment data and streaming media data of patients 902 in multiple wards in real time, dynamically displays waveform data, numerical data and treatment scenarios of patients in multiple wards in a visual manner, and centrally manages mechanical ventilation treatment data services of patients in multiple wards over a long distance.

[0291] Preferably, the ward management panel 1000 provided by the clinical data terminal device screen is as follows: Fig.10As shown, the mechanical ventilation data services of patients in multiple wards are centrally managed, the mechanical ventilation treatment data and evolution trends of patients are visually displayed, various data and event data, and treatment scenarios are observed and analyzed, and clinical diagnosis information, as well as doctor's orders and nursing information, and business instructions can be output. The business instructions at least include emergency instructions, ventilator parameter adjustment instructions, and instructions for generating analysis reports. The ward management panel is provided with content windows including the function operation area 1001, the ward management area 1002, the patient data management area 1003, and the clinical application management area 1004, as well as function buttons. All or a specific patient in multiple wards can be selected to observe the patient's real-time data, event data, historical data, clinical doctor's orders and nursing information, and treatment scenarios, analyze mechanical ventilation treatment, and issue business instructions or doctor's orders. Specifically, the function operation area 1001 is provided with function buttons for switching between the various panels of the clinical data terminal device and for selecting a designated patient; the ward management area 1002 is used to display and retrieve patient information, ward information and report information; the patient data management area 1003 is used to display and retrieve the patient's mechanical ventilation treatment data; the clinical application management area 1004 is used to display and retrieve the patient's streaming media data, mechanical ventilation treatment data, and to issue business instructions or medical orders.

[0292] Preferably, Fig.11 The data analysis report panel 1100 of the clinical data terminal device provided in Example 2 of the present invention is an implementation example; the screen of the clinical data terminal device provides a data analysis report panel, which visually displays the mechanical ventilation analysis reports of all or a specified patient in multiple wards, and reviews the mechanical ventilation data analysis report according to the manual review request automatically prompted at the agreed time of the business instruction or doctor's order, distributes it over the network, prints it out, and archives it in the database. The data analysis report panel includes a content window for setting the function operation area 1101, the ward management area 1102, the analysis report area 1103, and the clinical doctor's order area 1104, as well as function buttons, centrally manages the mechanical ventilation data analysis report business of multiple wards, can analyze the reasons for data changes in combination with clinical treatment doctor's order information, and can also compare and analyze the mechanical ventilation treatment data analysis reports of a patient, judge the trend of the disease evolution, and trigger the printer to output the mechanical ventilation treatment data analysis report. Specifically, the function operation area 1101 is provided with function buttons for switching between the data analysis panels of the clinical data terminal device, and for managing reports; the ward management area 1102 is used to display and retrieve patient information, ward information and report information, and is also used to select designated patients or reports; the analysis report area 1103 is used to display and retrieve patient data analysis reports, and present them in a visual manner; the clinical medical order area 1104 is used to display and retrieve patient mechanical ventilation treatment data, edit the report content, and trigger a print command.

[0293] To further expand the scalability and flexibility of the mechanical ventilation therapy data management system in this embodiment, illustratively, the mechanical ventilation therapy data management system can be deployed under a cloud computing architecture, that is, all functional modules in the mechanical ventilation therapy data management system are deployed under a cloud computing architecture; or, some functional modules in the mechanical ventilation therapy data management system are deployed on a cloud platform, such as a report generation module and a streaming media management module, etc., to realize various business functions of the system, including cloud-edge collaborative real-time broadcast of waveform data, numerical data, and treatment scenarios; other functional modules are deployed on edge computing nodes, such as data acquisition and standardization modules, data cleaning modules, data labeling modules, etc., to perform data cleaning and data standardization processing, as well as data labeling and event labeling at the data source, to provide support for cloud-edge collaboration and cloud platform data storage management, fast multi-dimensional data retrieval, and data update. Fig.12 As shown, it includes a multi-mode gateway 1201, a main control server 1202, a database server 1203, an application software system 1204, a bedside audio and video device 1205, and a clinical data terminal device 1206; the multi-mode gateway 1201 is connected to the bedside audio and video devices 1205, ventilators 1207 and multi-parameter monitors 1208, clinical information systems 1209 and clinical data terminal devices 1206 of different manufacturers in various wards of the hospital through the network, and establishes a connection with the cloud platform 1210.

[0294] The mechanical ventilation treatment data management system in this embodiment can also connect to a third-party medical data service, entrusting the complex and cumbersome real-time monitoring and data analysis and processing of mechanical ventilation treatment data to a third party for management, so that doctors and nurses can focus on the clinical medical issues of mechanical ventilation treatment, further reduce the work pressure of doctors and nurses, and improve the quality and work efficiency of mechanical ventilation treatment.

[0295] The above-mentioned method and system embodiments improve the existing clinical application method of mechanical ventilation treatment data. Doctors and nurses can centrally manage patients in multiple wards remotely, quickly grasp comprehensive mechanical ventilation data and disease evolution trends, solve the difficulties caused by huge multi-source heterogeneous data in clinical applications, reduce the work pressure of doctors and nurses, and improve the quality of mechanical ventilation treatment. It has wide applicability.

[0296] This embodiment reduces hardware costs and business pressure, performs data preprocessing at the data source, effectively reduces the delay of network data communication and cloud platform processing, and can flexibly determine the scale of user data storage. For example, data files are limited to be stored for three months, reducing the pressure and storage cost of mechanical ventilation treatment data storage. When an external network fails, the edge computing node can independently complete the processing of the mechanical ventilation treatment data analysis report.

[0297] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0298] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for managing mechanical ventilation treatment data, characterized in that: The method comprises: Patient information and real-time received mechanical ventilation treatment data are respectively placed into standard data structure objects to form standardized data; the mechanical ventilation treatment data includes mechanical ventilation data; the mechanical ventilation treatment data also includes one or more of the following: physiological parameter data, blood gas analysis data, laboratory diagnosis data, imaging data, clinical diagnosis information and device ID for outputting mechanical ventilation treatment data; wherein the physiological parameter data is output by a multi-parameter monitor; the patient information, blood gas analysis data, laboratory diagnosis data, imaging data and clinical diagnosis information are all output by a clinical information system; the mechanical ventilation data is output by a ventilator device; Performing data cleaning on the standardized data to obtain a medical data set; Data annotation is performed on the mechanical ventilation treatment data in the medical data set; the data annotation includes: marking the amplitude, frequency and shape of the waveform data, marking the measurement value of the numerical data, and marking abnormal events; When a report generation instruction is received, a mechanical ventilation treatment data analysis report is generated based on the medical data set carrying the data annotations; In the mechanical ventilation data, the waveform data includes: mechanical ventilation pressure waveform data, volume waveform data, flow waveform data and end-tidal carbon dioxide waveform data; the numerical data includes: airway resistance, work of breathing, compliance, ventilation parameters and respiratory rate; During the data labeling process of the mechanical ventilation data, The marked waveform data include: the start of inspiration, inspiration, end of inspiration, start of expiratory, expiratory, end of expiratory, peak airway pressure, peak expiratory flow rate of each mechanical ventilation waveform, and the inspiration baseline, expiratory plateau height, start and end of expiratory plateau of the end-tidal carbon dioxide waveform; The marked abnormal events include: patient-ventilator asynchrony events, airway pressure too high / too low, positive end-expiratory pressure too high / too low, minute ventilation too high / too low, oxygen concentration too high / too low, respiratory rate too high / too low, end-tidal carbon dioxide partial pressure too high / too low, tidal volume too high / too low, ventilator stoppage, suffocation and airway pressure increase; Among them, based on the neural network model, the human-machine asynchronous event is determined and executed: Based on the pressure-time curve of the mechanical ventilation pressure waveform data, the volume-time curve of the mechanical ventilation volume waveform data, and the flow-time curve of the mechanical ventilation flow waveform data, the model input is constructed; T = < S1,…,S c ,…,S C> , where S c represents the input vector of the cth sampling point, and C represents the total number of sampling points processed by the neural network model at a time, that is, the model input length; Respectively represent the flow rate, pressure, and volume values ​​of the cth sampling point; Perform two sliding window processes to enrich features and obtain the time series after sliding window processing; where: The first sliding window processing is performed: the model input T is subjected to sliding window processing with a sliding window step length of b / 2, where b represents the first sliding window width; after the first sliding window processing, the model input length changes from C to C′, and the value range of c changes to [1, C′]; After the first sliding window processing, based on the l-th type of waveform data in all sampling points in the c-th sliding window, the Euclidean distance of the l-th type of waveform data in the Euclidean matrix vector is calculated. Right now, l takes 1, 2, and 3 respectively; finally, the Euclidean matrix vector is obtained Based on the l-th type of waveform data in all sampling points in the c-th sliding window, the distance difference of the l-th type of waveform data in the distance difference vector is calculated. Right now, The final distance difference vector is expressed as Therefore, the time series T′ after the first sliding window processing is < G1,…,G c ,…,G C′> ,in, G c = < NOR c ,DON c > ; The second sliding window processing is performed: the time series T′ is subjected to sliding window processing again with a sliding window step length of f / 2, where f represents the second sliding window width; after the second sliding window processing, the model input length changes from C′ to C″, and the value range of c changes to [1, C″]; During the second sliding window processing, the derived features of all Euclidean matrix vectors and distance difference vectors in the second sliding window are obtained to obtain the time series after the second sliding window processing; The time series after sliding window processing is input into the neural network model, and the neural network model outputs normal events and abnormal ventilator events after processing, including the type of human-machine asynchrony events and their occurrence time.

2. The method for managing mechanical ventilation therapy data according to claim 1, characterized in that: In the physiological parameter data, waveform data includes: electrocardiogram waveform data, respiratory waveform data, volume pulse waveform data and invasive arterial pressure waveform data; numerical data includes: heart rate, blood pressure, blood oxygen saturation and body temperature; In the process of labeling the physiological parameter data, The forms of the marked ECG waveform data include: P wave, QRS complex, T wave, ST segment, PR interval, RR interval and QT interval of each ECG cycle waveform; The morphology of the marked respiratory waveform data includes: the peaks and troughs of each respiratory cycle waveform; The morphology of the marked volume pulse waveform data includes: the peaks and troughs of each pulse cycle waveform; The morphology of the marked invasive arterial pressure waveform data includes: the peaks and troughs of each arterial systolic-diastolic cycle waveform; Abnormal events marked include: tachycardia, bradycardia, cardiac arrest, atrial flutter, atrial fibrillation, ventricular flutter, ventricular fibrillation, atrioventricular block, ST segment elevation / depression, high / low blood pressure, high / low body temperature and low blood oxygen saturation.

3. The method for managing mechanical ventilation therapy data according to claim 2, characterized in that: In the blood gas analysis data, the numerical data include: oxygen partial pressure, carbon dioxide partial pressure, total carbon dioxide, potassium ion concentration, sodium ion concentration, bicarbonate ion concentration and pH value; In the process of labeling the blood gas analysis data, The marked abnormal events include: oxygen partial pressure too high / too low, carbon dioxide partial pressure too high / too low, total carbon dioxide too high / too low, potassium ion concentration too high / too low, sodium ion concentration too high / too low, and pH too high / too low.

4. The method for managing mechanical ventilation treatment data according to claim 3, characterized in that: In the laboratory diagnosis data, the numerical data include: white blood cell count, red blood cell count, platelet count, hemoglobin, hematocrit, blood glucose, prothrombin time and activated partial thromboplastin time; In the process of data annotation for experimental diagnostic data, The marked abnormal events include: white blood cell count is too high / too low, red blood cell count is too high / too low, hemoglobin is too high / too low, hematocrit is too high / too low, blood sugar is too high / too low, platelet count is too high / too low, prothrombin time is too long / too short, activated partial thromboplastin time is too long / too short.

5. The method for managing mechanical ventilation treatment data according to claim 4, characterized in that: The report framework of the mechanical ventilation treatment data analysis report includes a report summary page and a report analysis page; wherein the report summary page is divided into a basic information area, a mechanical ventilation data area and an analysis conclusion area; The basic information area includes at least patient information and key time node information; and also includes the following optional items: clinical diagnosis information, device ID for outputting mechanical ventilation treatment data, and mechanical ventilation mode and setting information; The mechanical ventilation data area at least includes respiratory analysis results, human-machine asynchrony event analysis results, airway resistance change analysis results, compliance change analysis results and lung injury warning analysis results; each analysis result of the mechanical ventilation data area is obtained based on mechanical ventilation data and its data annotation; the respiratory analysis results include shallow rapid breathing index; The analysis conclusion area is used to display the analysis description and conclusion of the mechanical ventilation treatment data; The report analysis page is used to display detailed analysis reports of one or more mechanical ventilation treatment data in the form of charts.

6. The method for managing mechanical ventilation treatment data according to claim 5, characterized in that: The report analysis page includes a trend graph page; the trend graph page includes: Pressure, volume, and ventilation trend analysis area for mechanical ventilation data; Respiratory rate trend analysis area; Inspired oxygen concentration trend analysis area; and, The trend analysis description and conclusion area is used to summarize the trend trends of each trend analysis area and give a trend analysis description and corresponding conclusions; The contents of each trend analysis area in the trend graph page are obtained based on mechanical ventilation data and data annotations thereof.

7. The method for managing mechanical ventilation therapy data according to claim 6, characterized in that: The report analysis page also includes a data page; The data page at least includes: Mechanical ventilation setting parameter data subpage, including: mechanical ventilation setting parameter record table, mechanical ventilation ventilation mode and setting parameter change trend chart; Mechanical ventilation monitoring parameter data subpage, including mechanical ventilation monitoring parameter record table; The contents of the mechanical ventilation setting parameter data subpage and the mechanical ventilation monitoring parameter data subpage are obtained based on the mechanical ventilation data and data annotation thereof; The data page also includes the following optional data subpages: 24-hour physiological parameter data subpage, including 24-hour vital signs monitoring record sheet, heart rate / respiratory rate / blood pressure trend chart; the content of the 24-hour physiological parameter data subpage is obtained based on the physiological parameter data and its data annotation; 24-hour blood gas analysis and laboratory diagnosis data subpage, including 24-hour blood gas analysis monitoring record sheet, oxygen partial pressure / carbon dioxide partial pressure / bicarbonate trend chart, and laboratory diagnosis data sheet; The content of the 24-hour blood gas analysis and laboratory diagnosis data subpage is obtained based on the blood gas analysis data, laboratory diagnosis data and data annotation thereof; Alarm event data subpage, including: alarm event record table, 24-hour alarm event distribution diagram, alarm event comprehensive analysis table; The alarm event data subpage is obtained based on abnormal events marked by mechanical ventilation treatment data; The extended image data subpage is used to display the image data.

8. The method for managing mechanical ventilation therapy data according to claim 7, characterized in that: The report analysis page also includes a graphics page; the graphics page is divided into a mechanical ventilation waveform data area, a mechanical ventilation loop diagram area, and a physiological parameter waveform data area; wherein, The mechanical ventilation waveform data area includes continuous normal mechanical ventilation data waveform graphs and abnormal mechanical ventilation event data waveform graphs; The mechanical ventilation loop diagram area includes a pressure-volume loop diagram and a flow-volume loop diagram drawn by combining mechanical ventilation pressure, volume, and flow data; The contents of the mechanical ventilation waveform data area and the mechanical ventilation ring diagram area are obtained based on the mechanical ventilation data and data annotation thereof; The physiological parameter waveform data area includes a continuous physiological parameter data waveform graph synchronized with the mechanical ventilation data waveform graph; The content of the physiological parameter waveform data area is obtained based on the physiological parameter data and its data annotation.

9. The method for managing mechanical ventilation therapy data according to any one of claims 5 to 8, characterized in that: The report generation instruction is a customized report generation instruction; The customized report generation instruction includes one or more customized parameters for generating a report summary page and a report analysis page matching the customized parameters; The customized parameters include a trend graph page, a data page or a graphic page.

10. The method for managing mechanical ventilation therapy data according to any one of claims 5 to 8, characterized in that: The method further comprises: During the data labeling process, when one or more abnormal events such as ventilator cessation, suffocation, increased airway pressure, cardiac arrest, tachycardia, bradycardia, ventricular flutter, ventricular fibrillation and hypotension occur, an emergency instruction is issued, and the emergency instruction carries information about the abnormal event that occurred.

11. The method for managing mechanical ventilation therapy data according to any one of claims 5 to 8, characterized in that: The method further comprises: During the data labeling process, when one or more abnormal events occur, including human-machine asynchrony events, airway pressure too high, airway pressure too low, positive end-expiratory pressure too high, positive end-expiratory pressure too low, minute ventilation too high, minute ventilation too low, tidal volume too high, tidal volume too low, respiratory rate too high, respiratory rate too low, oxygen concentration too high, and oxygen concentration too low, a ventilator parameter adjustment instruction is issued, and the ventilator parameter adjustment instruction carries information about the abnormal events that have occurred.

12. The method for managing mechanical ventilation therapy data according to any one of claims 1 to 4, characterized in that: In the standard data structure object, a corresponding data structure object is configured for each mechanical ventilation treatment data and patient information; Each decoupled mechanical ventilation treatment data and patient information are placed into corresponding data structure objects to form standardized data.

13. The method for managing mechanical ventilation therapy data according to any one of claims 1 to 4, characterized in that: While receiving mechanical ventilation treatment data in real time, it also receives streaming media data of the treatment scene in real time; Live broadcast of waveform data, numerical data, treatment scenarios and mechanical ventilation treatment data analysis reports in the mechanical ventilation treatment data of some or all patients in one or more wards.

14. A mechanical ventilation treatment data management system, characterized in that: The system includes an application software system, a clinical data terminal device and a database server; wherein, The application software system comprises: The data acquisition and standardization module is used to place the patient information and the real-time received mechanical ventilation treatment data into standard data structure objects to form standardized data; the mechanical ventilation treatment data includes mechanical ventilation data; the mechanical ventilation treatment data also includes one or more of the following: physiological parameter data, blood gas analysis data, laboratory diagnosis data, imaging data, clinical diagnosis information and device ID for outputting mechanical ventilation treatment data; wherein, Physiological parameter data are output by multi-parameter monitors; patient information, blood gas analysis data, laboratory diagnosis data, imaging data and clinical diagnosis and treatment information are all output by clinical information systems; mechanical ventilation data are output by ventilator equipment; A data cleaning module, used for cleaning the standardized data to obtain a medical data set; A data annotation module is used to annotate the mechanical ventilation treatment data in the medical data set; the data annotation includes: marking the amplitude, frequency and shape of the waveform data, Mark the measured values ​​of numerical data and mark abnormal events; A report generation module, for generating a mechanical ventilation treatment data analysis report based on a medical data set carrying data annotations when a report generation instruction is received; A clinical data terminal device, used to issue the report generation instruction, and also used to receive and display the mechanical ventilation treatment data analysis report; A database server, used for storing the standardized data, medical data set, and mechanical ventilation treatment data analysis report in partitions; In the mechanical ventilation data, the waveform data includes: mechanical ventilation pressure waveform data, volume waveform data, flow waveform data and end-tidal carbon dioxide waveform data; the numerical data includes: airway resistance, work of breathing, compliance, ventilation parameters and respiratory rate; During the data labeling process of the mechanical ventilation data, The marked waveform data include: the start of inspiration, inspiration, end of inspiration, start of expiratory, expiratory, end of expiratory, peak airway pressure, peak expiratory flow rate of each mechanical ventilation waveform, and the inspiration baseline, expiratory plateau height, start and end of expiratory plateau of the end-tidal carbon dioxide waveform; The marked abnormal events include: patient-ventilator asynchrony events, airway pressure too high / too low, positive end-expiratory pressure too high / too low, minute ventilation too high / too low, oxygen concentration too high / too low, respiratory rate too high / too low, end-tidal carbon dioxide partial pressure too high / too low, tidal volume too high / too low, ventilator stoppage, suffocation and airway pressure increase; Among them, based on the neural network model, the human-machine asynchronous event is determined and executed: Based on the pressure-time curve of the mechanical ventilation pressure waveform data, the volume-time curve of the mechanical ventilation volume waveform data, and the flow-time curve of the mechanical ventilation flow waveform data, the model input is constructed; T = < S1,…,S c ,…,S C> , where S c represents the input vector of the cth sampling point, and C represents the total number of sampling points processed by the neural network model at a time, that is, the model input length; Respectively represent the flow rate, pressure, and volume values ​​of the cth sampling point; Perform two sliding window processes to enrich features and obtain the time series after sliding window processing; where: The first sliding window processing is performed: the model input T is subjected to sliding window processing with a sliding window step length of b / 2, where b represents the first sliding window width; after the first sliding window processing, the model input length changes from C to C′, and the value range of c changes to [1, C′]; After the first sliding window processing, based on the l-th type of waveform data in all sampling points in the c-th sliding window, the Euclidean distance of the l-th type of waveform data in the Euclidean matrix vector is calculated. Right now, l takes 1, 2, and 3 respectively; finally, the Euclidean matrix vector is obtained Based on the l-th type of waveform data in all sampling points in the c-th sliding window, the distance difference of the l-th type of waveform data in the distance difference vector is calculated. Right now, The final distance difference vector is expressed as Therefore, the time series T′ after the first sliding window processing is < G1,…,G c ,…,G C′> ,in, G c = < NOR c ,DON c > ; The second sliding window processing is performed: the time series T′ is subjected to sliding window processing again with a sliding window step length of f / 2, where f represents the second sliding window width; after the second sliding window processing, the model input length changes from C′ to C″, and the value range of c changes to [1, C″]; During the second sliding window processing, the derived features of all Euclidean matrix vectors and distance difference vectors in the second sliding window are obtained to obtain the time series after the second sliding window processing; The time series after sliding window processing is input into the neural network model, and the neural network model outputs normal events and abnormal ventilator events after processing, including the type of human-machine asynchrony events and their occurrence time.

15. The mechanical ventilation treatment data management system according to claim 14, characterized in that: In the physiological parameter data, waveform data includes: electrocardiogram waveform data, respiratory waveform data, volume pulse waveform data and invasive arterial pressure waveform data; numerical data includes: heart rate, blood pressure, blood oxygen saturation and body temperature; In the process of labeling the physiological parameter data, The forms of the marked ECG waveform data include: P wave, QRS complex, T wave, ST segment, PR interval, RR interval and QT interval of each ECG cycle waveform; The morphology of the marked respiratory waveform data includes: the peaks and troughs of each respiratory cycle waveform; The morphology of the marked volume pulse waveform data includes: the peaks and troughs of each pulse cycle waveform; The morphology of the marked invasive arterial pressure waveform data includes: the peaks and troughs of each arterial systolic-diastolic cycle waveform; Abnormal events marked include: tachycardia, bradycardia, cardiac arrest, atrial flutter, atrial fibrillation, ventricular flutter, ventricular fibrillation, atrioventricular block, ST segment elevation / depression, high / low blood pressure, high / low body temperature and low blood oxygen saturation.

16. The mechanical ventilation treatment data management system according to claim 15, characterized in that: In the blood gas analysis data, the numerical data include: oxygen partial pressure, carbon dioxide partial pressure, total carbon dioxide, potassium ion concentration, sodium ion concentration, bicarbonate ion concentration and pH value; In the process of labeling the blood gas analysis data, The marked abnormal events include: oxygen partial pressure too high / too low, carbon dioxide partial pressure too high / too low, total carbon dioxide too high / too low, potassium ion concentration too high / too low, sodium ion concentration too high / too low, and pH too high / too low.

17. The mechanical ventilation treatment data management system according to claim 16, characterized in that: In the laboratory diagnosis data, the numerical data include: white blood cell count, red blood cell count, platelet count, hemoglobin, hematocrit, blood glucose, prothrombin time and activated partial thromboplastin time; In the process of data annotation for experimental diagnostic data, The marked abnormal events include: white blood cell count is too high / too low, red blood cell count is too high / too low, hemoglobin is too high / too low, hematocrit is too high / too low, blood sugar is too high / too low, platelet count is too high / too low, prothrombin time is too long / too short, activated partial thromboplastin time is too long / too short.

18. The mechanical ventilation treatment data management system according to claim 17, characterized in that: The report framework of the mechanical ventilation treatment data analysis report includes a report summary page and a report analysis page; wherein the report summary page is divided into a basic information area, a mechanical ventilation data area and an analysis conclusion area; The basic information area includes at least patient information and key time node information; and also includes the following optional items: clinical diagnosis information, device ID for outputting mechanical ventilation treatment data, and mechanical ventilation mode and setting information; The mechanical ventilation data area at least includes respiratory analysis results, human-machine asynchrony event analysis results, airway resistance change analysis results, compliance change analysis results and lung injury warning analysis results; each analysis result of the mechanical ventilation data area is obtained based on mechanical ventilation data and its data annotation; the respiratory analysis results include shallow rapid breathing index; The analysis conclusion area is used to display the analysis description and conclusion of the mechanical ventilation treatment data; The report analysis page is used to display detailed analysis reports of one or more mechanical ventilation treatment data in the form of charts.

19. The mechanical ventilation therapy data management system according to claim 18, characterized in that: The report analysis page includes a trend graph page; the trend graph page includes: Pressure, volume, and ventilation trend analysis area for mechanical ventilation data; Respiratory rate trend analysis area; Inspired oxygen concentration trend analysis area; and, The trend analysis description and conclusion area is used to summarize the trend trends of each trend analysis area and give a trend analysis description and corresponding conclusions; The contents of each trend analysis area in the trend graph page are obtained based on mechanical ventilation data and data annotations thereof.

20. The mechanical ventilation therapy data management system according to claim 19, characterized in that: The report analysis page also includes a data page; The data page at least includes: Mechanical ventilation setting parameter data subpage, including: mechanical ventilation setting parameter record table, mechanical ventilation ventilation mode and setting parameter change trend chart; Mechanical ventilation monitoring parameter data subpage, including mechanical ventilation monitoring parameter record table; The contents of the mechanical ventilation setting parameter data subpage and the mechanical ventilation monitoring parameter data subpage are obtained based on the mechanical ventilation data and data annotation thereof; The data page also includes the following optional data subpages: 24-hour physiological parameter data subpage, including 24-hour vital signs monitoring record sheet, heart rate / respiratory rate / blood pressure trend chart; the content of the 24-hour physiological parameter data subpage is obtained based on the physiological parameter data and its data annotation; 24-hour blood gas analysis and laboratory diagnosis data subpage, including 24-hour blood gas analysis monitoring record sheet, oxygen partial pressure / carbon dioxide partial pressure / bicarbonate trend chart, and laboratory diagnosis data sheet; The content of the 24-hour blood gas analysis and laboratory diagnosis data subpage is obtained based on the blood gas analysis data, laboratory diagnosis data and data annotation thereof; Alarm event data subpage, including: alarm event record table, 24-hour alarm event distribution diagram, alarm event comprehensive analysis table; The alarm event data subpage is obtained based on abnormal events marked by mechanical ventilation treatment data; The extended image data subpage is used to display the image data.

21. The mechanical ventilation treatment data management system according to claim 20, characterized in that: The report analysis page also includes a graphics page; the graphics page is divided into a mechanical ventilation waveform data area, a mechanical ventilation loop diagram area, and a physiological parameter waveform data area; wherein, The mechanical ventilation waveform data area includes continuous normal mechanical ventilation data waveform graphs and abnormal mechanical ventilation event data waveform graphs; The mechanical ventilation loop diagram area includes a pressure-volume loop diagram and a flow-volume loop diagram drawn by combining mechanical ventilation pressure, volume, and flow data; The contents of the mechanical ventilation waveform data area and the mechanical ventilation ring diagram area are obtained based on the mechanical ventilation data and data annotation thereof; The physiological parameter waveform data area includes a continuous physiological parameter data waveform graph synchronized with the mechanical ventilation data waveform graph; The content of the physiological parameter waveform data area is obtained based on the physiological parameter data and its data annotation.

22. The mechanical ventilation therapy data management system according to any one of claims 18 to 21, characterized in that: The report generation instruction is a customized report generation instruction; The customized report generation instruction includes one or more customized parameters for generating a report summary page and a report analysis page matching the customized parameters; The customized parameters include a trend graph page, a data page or a graphic page.

23. The mechanical ventilation therapy data management system according to any one of claims 18 to 21, characterized in that: During the data annotation process of the data annotation module, when one or more abnormal events of ventilator stopping working, asphyxia, airway pressure increase, cardiac arrest, tachycardia, bradycardia, ventricular flutter, ventricular fibrillation and hypotension occur, an emergency instruction is sent to the clinical data terminal device, and the emergency instruction carries information of the abnormal event that occurred; The clinical data terminal device is also used to receive and respond to the emergency instruction.

24. The mechanical ventilation therapy data management system according to any one of claims 18 to 21, characterized in that: During the data annotation process of the data annotation module, when one or more abnormal events of human-machine asynchrony event, too high airway pressure, too low airway pressure, too high positive end-expiratory pressure, too low positive end-expiratory pressure, too high minute ventilation, too low minute ventilation, too high tidal volume, too low tidal volume, too high respiratory rate, too low respiratory rate, too high oxygen concentration, too low oxygen concentration occur, a ventilator parameter adjustment instruction is sent to the clinical data terminal device, and the ventilator parameter adjustment instruction carries information of the abnormal event that occurred; The clinical data terminal device is also used to receive and respond to ventilator parameter adjustment instructions.

25. The mechanical ventilation therapy data management system according to any one of claims 14-17, characterized in that: In the standard data structure object, a corresponding data structure object is configured for each mechanical ventilation treatment data and patient information; Each decoupled mechanical ventilation treatment data and patient information are placed into corresponding data structure objects to form standardized data.

26. The mechanical ventilation therapy data management system according to any one of claims 14-17, characterized in that: The application software system also includes a streaming media management module, which is used to receive streaming media data of the treatment scene in real time; the streaming media data is output by the bedside audio and video equipment; The database server is further used to partition and store the streaming media data, and associate the streaming media data with the medical data set and the mechanical ventilation treatment data analysis report; The clinical data terminal device is also used to broadcast in real time the waveform data, numerical data, treatment scenarios and mechanical ventilation treatment data analysis report in the mechanical ventilation treatment data of the medical data set of some or all patients in one or more wards.

27. The mechanical ventilation therapy data management system according to any one of claims 16-21, characterized in that: The system also includes a multi-mode gateway and a master control server; wherein, The multi-mode gateway is used to connect ventilator equipment, multi-parameter monitors, bedside audio and video equipment, clinical information systems and clinical data terminal equipment from different manufacturers in various wards of the hospital through the network; the multi-mode gateway is embedded with multiple communication protocols to realize data communication between ventilator equipment, multi-parameter monitors, bedside audio and video equipment, clinical information systems and data acquisition and standardization modules, and to realize data communication between application software systems and clinical data terminal equipment where patients are located; The main control server is embedded with a multi-task operating system and an application software system running on the multi-task operating system, and is used to control the operation of the application software system.

Images