Surgical anesthesia information early warning method and system

Abstract

The invention provides a surgical anesthesia information early warning method and system, and the method comprises the steps: obtaining a BMI index of a patient based on coded structured data, and carrying out the mining and analysis of a potential state of the patient through combining a mining analysis model and vectorized unstructured data; based on the unstructured data, collecting an anesthesia physiological signal of the patient and an electroencephalogram signal in an anesthesia state, and classifying the anesthesia state according to the anesthesia physiological signal and the electroencephalogram signal to obtain a classified anesthesia state; and collecting an anesthesia depth data set based on the potential state and the classified anesthesia state, extracting anesthesia features according to the anesthesia depth data set, constructing an early warning model, and inputting the feature input sample into the early warning model to enable the early warning model to generate early warning information. Early warning is performed through the early warning model to generate the early warning information, so that the state of the patient can be timely processed and adjusted in advance.

Description

Technical Field

[0001] The present invention relates to the technical field of information processing, and particularly relates to a surgical anesthesia information early warning method and system. Background Art

[0002] Anesthesia is a reversible functional inhibition of the central nervous system and (or) peripheral nervous system produced by drugs or other methods. The main feature of this inhibition is the loss of sensation, especially pain, to achieve painless surgical treatment while ensuring the life safety of patients during the perioperative period. The process must be completed by an anesthesiologist. The anesthesia information early warning system is an important tool for the combination of anesthesiology and intelligent monitoring technology, aiming to improve perioperative safety through real-time data analysis and risk prediction.

[0003] In the prior art, during the surgical process, it is necessary to accurately monitor various physiological indicators of the patient during the surgical anesthesia process. Traditional single judgment is difficult to explore the relationship between the patient's physiological indicators and the corresponding intraoperative symptoms, and it is impossible to comprehensively judge the patient's state, and it is difficult to quickly discover the patient's potential dangerous symptoms, and thus it is impossible to assist medical staff to make corresponding operations. Summary of the Invention

[0004] Based on this, the purpose of the present invention is to provide a surgical anesthesia information early warning method and system to solve the deficiencies in the above prior art.

[0005] In a first aspect, the present invention provides a surgical anesthesia information early warning method, and the method includes:

[0006] Obtain the structured data of the patient and collect the unstructured data during the surgical process; [[ID=2,4]]

[0007] Encode the structured data, and obtain the BMI index of the patient based on the encoded structured data, and process the unstructured data to obtain the vectorized unstructured data of a preset length;

[0008] Establish a mining and analysis model based on the vectorized representation of the surgical data in the vectorized unstructured data and the linear functions of several factor features, and combine the mining and analysis model and the vectorized unstructured data to mine and analyze the potential state of the patient;

[0009] Collect the anesthesia physiological signal and the electroencephalogram signal under anesthesia of the patient based on the unstructured data, and classify the anesthesia state according to the anesthesia physiological signal and the electroencephalogram signal to obtain the classified anesthesia state;

[0010] Collect an anesthesia depth data set based on the potential state and the classified anesthesia state, extract anesthesia features according to the anesthesia depth data set, generate feature input samples according to the anesthesia features, and generate sample labels;

[0011] Construct an early warning model according to a preset machine learning model and the sample labels, and input the feature input samples into the early warning model, so that the early warning model generates early warning information.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: By establishing a mining and analysis model based on vectorized unstructured data, and then combining the vectorized unstructured data and the mining and analysis model to analyze the potential state of the patient, not only can multiple physiological indicators of the patient be obtained, but also the potential state of the patient during the surgical anesthesia process can be obtained. Furthermore, before the patient shows dangerous symptoms, timely responses can be made. And by combining the input samples and sample labels generated from the potential state and the anesthesia state, and generating early warning information through the early warning model, it is possible to timely and proactively process and adjust the patient's state and take corresponding measures.

[0013] Further, the steps of obtaining the structured data of the patient and collecting the unstructured data during the operation include:

[0014] Obtain the basic information of the patient, where the basic information includes gender, height, age, weight, and department, and the gender, height, age, weight, and department constitute the structured data;

[0015] Collect the unstructured data during the operation through a monitor, an anesthesia monitor, and OAIS, where the unstructured data includes physiological monitoring sequence data, anesthesia physiological signals, electroencephalogram signals, and operation sequence data.

[0016] Further, the steps of encoding the structured data and obtaining the BMI index of the patient based on the encoded structured data, and processing the unstructured data to obtain vectorized unstructured data of a preset length include:

[0017] Extract the height information and weight information from the unstructured data, and calculate the BMI value of the patient through the BMI calculation formula;

[0018] Merge different texts in the unstructured data into a text set, and segment the text set to obtain a number of words;

[0019] Perform vector representation on a number of the words to obtain vectorized word numbers, and map the vectorized word data to the same length. The expression of the vector representation process is:

[0020]

[0021] Wherein, WT represents the weighted vector representation of words, N represents the number of words, tfidf i represents the TFIDF value of the i-th word, WV i represents the word vector of the i-th word.

[0022] Furthermore, the steps of establishing a mining and analysis model based on the vectorized representation of surgical data in the unstructured data and a linear function of several factor features, and mining and analyzing the potential state of the patient by combining the mining and analysis model and the vectorized unstructured data include:

[0023] Taking the vectorized representation of surgical data in the unstructured data as a feature item, and establishing a mining and analysis model through a linear function that combines the feature item with several factor features. The expression of the mining and analysis model is:

[0024]

[0025] Wherein, represents the predicted value of the i-th sample, D represents the feature dimension, x ij represents the j-th feature value of the i-th sample, w j the weight corresponding to the j-th feature, X i represents the feature row vector of the i-th sample, W represents the weight vector, W T represents the transposed weight vector;

[0026] Based on the mining and analysis model and the vectorized unstructured data, mining the potential high-risk situations, respiratory system conditions, circulatory system conditions, and nervous system conditions of the patient during the operation.

[0027] Furthermore, the steps of classifying the anesthesia state according to the anesthesia physiological signal and the electroencephalogram signal to obtain the classified anesthesia state include:

[0028] Drawing a depth-of-consciousness map based on the anesthesia physiological signal and the electroencephalogram signal;

[0029] Classifying the depth-of-consciousness map according to the level of consciousness to obtain depth-of-consciousness maps with different depth values, so as to obtain different anesthesia states of the patient.

[0030] Furthermore, the steps of extracting anesthesia features according to the anesthesia depth data set, generating feature input samples according to the anesthesia features, and generating sample labels include:

[0031] Extract the anesthesia parameters and situation characteristics from the anesthesia depth dataset, and generate feature input samples based on the anesthesia parameters and situation characteristics;

[0032] Obtain the generation time of the feature input sample, and generate a sample label for the feature input sample according to the generation time.

[0033] Further, the expression of the warning model is:

[0034]

[0035] In the formula, represents the prediction result, f represents a preset machine learning model, H represents the length of the observation window, B represents the interval time, P represents the length of the prediction window, x represents the feature input sample set, and t represents the sample label.

[0036] In a second aspect, the present invention also provides a surgical anesthesia information warning system, and the system includes:

[0037] An acquisition module, configured to acquire structured data of a patient and acquire unstructured data during a surgical procedure;

[0038] An encoding processing module, configured to encode the structured data, obtain the BMI index of the patient based on the encoded structured data, process the unstructured data to obtain vectorized unstructured data of a preset length;

[0039] A building and mining module, configured to establish a mining and analysis model based on the vectorized representation of surgical data and the linear function of several factor characteristics in the vectorized unstructured data, and mine and analyze the potential state of the patient in combination with the mining and analysis model and the vectorized unstructured data;

[0040] An acquisition and classification module, configured to acquire the anesthesia physiological signal of the patient and the electroencephalogram signal under anesthesia based on the unstructured data, and classify the anesthesia state according to the anesthesia physiological signal and the electroencephalogram signal to obtain the classified anesthesia state;

[0041] A collection and extraction module, configured to collect an anesthesia depth dataset based on the potential state and the classified anesthesia state, extract anesthesia features according to the anesthesia depth dataset, generate feature input samples according to the anesthesia features, and generate sample labels;

[0042] A construction and generation module, configured to construct a warning model according to a preset learning model, and input the feature input sample set and the sample label into the warning model, so that the warning model generates warning information.

[0043] In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above-mentioned surgical anesthesia information warning method is implemented.

[0044] In a fourth aspect, the present invention further provides a storage medium, on which a computer program is stored. When the program is executed by a processor, the above-mentioned surgical anesthesia information warning method is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flowchart of the surgical anesthesia information warning method in the first embodiment of the present invention;

[0046] Figure 2 It is a structural block diagram of the surgical anesthesia information warning system in the second embodiment of the present invention;

[0047] Figure 3 It is a schematic hardware structure diagram of the electronic device in the third embodiment of the present invention.

[0048] MAIN ELEMENT SYMBOL DESCRIPTION:

[0049] 10. Acquisition and collection module; 20. Encoding and processing module; 30. Establishment and mining module; 40. Acquisition and classification module; 50. Collection and extraction module; 60. Construction and generation module;

[0050] 70. Bus; 71. Processor; 72. Memory; 73. Communication interface.

[0051] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. SPECIFIC EMBODIMENTS

[0052] For the convenience of understanding the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0053] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there may also be a middle element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be a middle element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this invention belongs. The terms used in the description of the present invention herein are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0055] Embodiment 1

[0056] Please refer to Figure 1 , which shows the surgical anesthesia information warning method in the first embodiment of the present invention. The method includes steps S1 to S6:

[0057] S1. Obtain the structured data of the patient and collect the unstructured data during the operation;

[0058] Specifically, step S1 includes steps S11 to S12:

[0059] S11. Obtain the basic information of the patient, where the basic information includes gender, height, age, weight, and department. The gender, height, age, weight, and department constitute the structured data;

[0060] S12. Collect the unstructured data during the operation through a monitor, an anesthesia monitor, and OAIS. The unstructured data includes physiological monitoring sequence data, anesthesia physiological signals, electroencephalogram signals, and surgical sequence data;

[0061] It can be understood that the structured data includes the basic information of the patient itself, such as gender, height, age, and weight, and also includes the department situation of the symptoms suffered by the patient. The unstructured data includes physiological monitoring sequence data and surgical sequence data, and also includes anesthesia physiological signals and the electroencephalogram signals of the patient. Some relevant data of the patient are collected through a monitor, an anesthesia detector, and OAIS. It should be noted that OAIS is an information management framework based on international standards (ISO 14721) and is designed to long-term preserve and reliably access digital information.

[0062] S2. Encode the structured data, and obtain the BMI index of the patient based on the encoded structured data. Process the unstructured data to obtain vectorized unstructured data of a preset length;

[0063] Specifically, step S2 includes steps S21 to S23:

[0064] S21. Extract the height information and weight information from the unstructured data, and calculate the BMI index of the patient through the BMI calculation formula;

[0065] It should be noted that the height information and weight information in the structured data are extracted and encoded into digital form. The calculation of the BMI index is to divide the weight of the patient by the square of the height of the patient, so as to obtain the BMI index of the patient, which can measure whether the patient is too thin or too fat. When mining or warning the patient's status in the follow-up, being too thin or too fat is also an important indicator.

[0066] S22. Merge different texts in the unstructured data into a text set, and segment the text set to obtain a number of word counts.

[0067] S23. Perform vector representation on a number of the word counts to obtain vectorized word counts, and map the vectorized word data to the same length.

[0068] It can be understood that the unstructured data includes physiological monitoring sequence data, anesthesia physiological signals, electroencephalogram signals, and surgical sequence data, which contain many medical special names, as well as some Chinese symbols and numbers. In this embodiment, first, the text containing medical special names, Chinese symbols, and numbers is normalized in format, and segmented to obtain a number of word counts. The expression of the vector representation process is:

[0069]

[0070] In the formula, WT represents the weighted vector representation of the word, N represents the word count, tfidfi i represents the TFIDF value of the i-th word, and WV i represents the word vector of the i-th word.

[0071] It is worth noting that in order to be used in the subsequent mining analysis model, the vectorized word data needs to be mapped to the same length.

[0072] S3. Establish a mining analysis model based on the vectorized representation of the surgical data in the unstructured data and the linear function of several factor features, and combine the mining analysis model and the vectorized unstructured data to mine and analyze the potential status of the patient.

[0073] Specifically, step S3 includes steps S31 to S32:

[0074] S31. Use the vectorized representation of the surgical data in the unstructured data as a feature item, and establish a mining analysis model through the feature item combined with the linear function of several factor features.

[0075] It is understandable that the surgical data in the unstructured data needs to be vectorized as feature items, and a mining and analysis model is established based on the feature items and a linear function. In this embodiment, the expression of the mining and analysis model is:

[0076]

[0077] In the formula, represents the predicted value of the i-th sample, D represents the feature dimension, x ij represents the j-th feature value of the i-th sample, w j is the weight corresponding to the j-th feature, X i represents the feature row vector of the i-th sample, W represents the weight vector, and W T represents the transposed weight vector.

[0078] S32. Based on the mining and analysis model and the vectorized unstructured data, mine the potential high-risk situations, respiratory system conditions, circulatory system conditions, and nervous system conditions of the patient during the operation;

[0079] It is understandable that the mining and analysis model combined with unstructured data can predict and mine the potential state of the patient during the operation. The high-risk situations include airway problems and patient blood pressure problems. There are residual muscle relaxation problems in the respiratory system conditions, postoperative hypotension problems in the circulatory system conditions, and cognitive impairment in the nervous system conditions.

[0080] S4. Collect the anesthetic physiological signals and electroencephalogram signals under anesthesia of the patient based on the unstructured data, and classify the anesthesia state according to the anesthetic physiological signals and the electroencephalogram signals to obtain the classified anesthesia state;

[0081] It should be explained that the physiological monitoring sequence data, anesthetic physiological signals, and electroencephalogram signals included in the unstructured data can obtain the anesthetic physiological signals and electroencephalogram signals of the patient under anesthesia.

[0082] Specifically, step S4 includes steps S41 to S42:

[0083] S41. Draw a depth-of-consciousness map based on the anesthetic physiological signals and the electroencephalogram signals;

[0084] S42. Classify the depth-of-consciousness map according to the level of consciousness to obtain depth-of-consciousness maps with different depth values, so as to obtain different anesthesia states of the patient;

[0085] It can be understood that the EEG signals on the patient's forehead are directly collected by an anesthetic detector, and the collected data is transferred to a notebook computer for storage using an RS232 transmission line, and a depth-of-consciousness map is drawn. Then, according to the patient's current level of consciousness during anesthesia, the depth-of-consciousness map is classified, and depth-of-consciousness maps with different depth values at different times can be obtained, and thus the different anesthesia states of the patient at different times can be obtained.

[0086] It should be noted that the δ-wave frequency band distribution is 0.5 - 4 Hz, which mainly appears in the EEG signals of patients under deep anesthesia and hardly exists in the EEG signals of awake patients.

[0087] S5, collect an anesthetic depth data set based on the potential state and the classified anesthesia state, extract anesthetic features according to the anesthetic depth data set, generate a feature input sample according to the anesthetic features, and generate a sample label;

[0088] Specifically, the step S5 includes steps S51 to S52:

[0089] S51, extract the anesthetic parameters and condition features in the anesthetic depth data set, and generate a feature input sample based on the anesthetic parameters and condition features;

[0090] S52, obtain the generation time of the feature input sample, and generate a sample label for the feature input sample according to the generation time;

[0091] It can be understood that anesthetic parameters, vital sign indicators, and detection indicators are extracted from the anesthetic depth data set according to medical indicators and the patient's condition. And the anesthetic parameters, vital sign indicators, and detection indicators are statistically constructed to obtain an input sample, and based on the anesthetic parameters, vital sign indicators, and detection indicators of the patient in different time periods, a sample label is generated according to the time period, and the name of the feature input sample is marked.

[0092] S6, construct an early warning model according to a preset machine learning model and the sample label, and input the feature input sample into the early warning model so that the early warning model generates early warning information;

[0093] It should be noted that in this embodiment, the expression of the early warning model is:

[0094]

[0095] In the formula, represents the prediction result, f represents the preset machine learning model, H represents the length of the observation window, B represents the interval time, P represents the length of the prediction window, x represents the feature input sample set, and t represents the sample label.

[0096] It should be noted that in this embodiment, the preset machine model adopts the LightGBM model. The LightGBM model is trained with sample labels, and an early warning model is constructed by combining the feature input samples. Then, by inputting the feature input samples into the early warning model, early warning information is generated. The early warning information can provide the medical staff with time for preparation, avoiding accidents during the anesthesia process of the patient.

[0097] In summary, for the surgical anesthesia information early warning method and system in the above embodiments of the present invention, through the mining and analysis model established based on the vectorized unstructured data, and then combining the vectorized unstructured data and the mining and analysis model to analyze the potential state of the patient, not only can multiple physiological indicators of the patient be obtained, but also the potential state of the patient during the surgical anesthesia process can be obtained. Furthermore, before the patient shows dangerous symptoms, timely responses can be made. By combining the input samples and sample labels generated from the potential state and the anesthesia state, and generating early warning information through the early warning model, it is possible to timely and proactively adjust the state of the patient and take corresponding measures.

[0098] Embodiment 2

[0099] The present invention also provides a surgical anesthesia information early warning system. Please refer to Figure 2 , which shows the surgical anesthesia information early warning system in the second embodiment of the present invention. The system includes:

[0100] An acquisition and collection module 10, configured to acquire the structured data of the patient and collect the unstructured data during the surgical process;

[0101] An encoding and processing module 20, configured to encode the structured data, and obtain the BMI index of the patient based on the encoded structured data, and process the unstructured data to obtain vectorized unstructured data of a preset length;

[0102] A model establishment and mining module 30, configured to establish a mining and analysis model based on the vectorized representation of the surgical data in the vectorized unstructured data and the linear functions of several factor features, and combine the mining and analysis model and the vectorized unstructured data to mine and analyze the potential state of the patient;

[0103] An acquisition and classification module 40, configured to collect the anesthesia physiological signals and electroencephalogram signals under anesthesia of the patient based on the unstructured data, and classify the anesthesia state according to the anesthesia physiological signals and the electroencephalogram signals to obtain the classified anesthesia state;

[0104] A collection and extraction module 50, configured to collect an anesthesia depth data set based on the potential state and the classified anesthesia state, extract anesthesia features according to the anesthesia depth data set, generate a feature input sample according to the anesthesia features, and generate a sample label;

[0105] A construction and generation module 60, configured to construct an early warning model according to a preset learning model, and input the feature input sample set and the sample label into the early warning model, so that the early warning model generates early warning information;

[0106] The expression of the early warning model is:

[0107]

[0108] In the formula, represents the prediction result, f represents a preset machine learning model, H represents the length of the observation window, B represents the interval time, P represents the length of the prediction window, x represents the feature input sample set, and t represents the sample label.

[0109] In some alternative embodiments, the acquisition and collection module 10 includes:

[0110] A first acquisition unit, configured to acquire the basic information of the patient, where the basic information includes gender, height, age, weight, and department, and the gender, the height, the age, the weight, and the department form the structured data;

[0111] An acquisition unit, configured to acquire unstructured data during the operation through a monitor, an anesthesia monitor, and an OAIS, where the unstructured data includes physiological monitoring sequence data, anesthesia physiological signals, electroencephalogram signals, and operation sequence data.

[0112] In some alternative embodiments, the acquisition and encoding processing module 20 includes:

[0113] An extraction and calculation unit, configured to extract the height information and the weight information from the unstructured data, and calculate the BMI index of the patient through a BMI calculation formula;

[0114] A merging unit, configured to merge different texts in the unstructured data into a text set, and perform word segmentation on the text set to obtain a number of words;

[0115] A mapping unit, configured to perform vector representation on a number of the words to obtain vectorized word numbers, and map the vectorized word data to the same length. The expression of the vector representation process is:

[0116]

[0117] Wherein, WT represents the weighted vector representation of the word, N represents the number of words, tfidfi i represents the TFIDF value of the i-th word, WV i represents the word vector of the i-th word.

[0118] In some alternative embodiments, the establishing and mining module 30 includes:

[0119] A combining unit, configured to vectorize and represent the surgical data in the unstructured data as feature items, and establish a mining and analysis model through a linear function combining several factor features. The expression of the mining and analysis model is:

[0120]

[0121] Wherein, represents the predicted value of the i-th sample, D represents the feature dimension, x ij represents the j-th feature value of the i-th sample, w j the weight corresponding to the j-th feature, X i represents the feature row vector of the i-th sample, W represents the weight vector, W T represents the transposed weight vector;

[0122] A mining unit, configured to mine the potential high-risk conditions, respiratory system conditions, circulatory system conditions, and nervous system conditions of the patient during the operation based on the mining and analysis model and the vectorized unstructured data.

[0123] In some alternative embodiments, the acquisition and classification module 40 includes:

[0124] A plotting unit, configured to plot a depth-of-consciousness map based on the anesthesia physiological signal and the electroencephalogram signal;

[0125] A classification unit, configured to classify the depth-of-consciousness map according to the level of consciousness to obtain depth-of-consciousness maps with different depth values, so as to obtain different anesthesia states of the patient.

[0126] In some alternative embodiments, the collection and extraction module 50 includes:

[0127] An extraction unit, configured to extract the anesthesia parameters and situation features in the anesthesia depth dataset, and generate a feature input sample based on the anesthesia parameters and situation features;

[0128] A second acquisition unit, configured to obtain the generation time of the feature input sample, and generate a sample label for the feature input sample according to the generation time.

[0129] The functions or operation steps implemented when the above-mentioned modules and units are executed are substantially the same as those in the above method embodiment, and will not be described in detail here.

[0130] The surgical anesthesia information warning system provided by the embodiment of the present invention has the same implementation principle and technical effects as those in the foregoing method embodiment. For a brief description, for the parts not mentioned in the system embodiment, reference may be made to the corresponding content in the foregoing method embodiment.

[0131] Embodiment III

[0132] The present invention further provides an electronic device. Please refer to Figure 3 , which shows the electronic device in the third embodiment of the present invention.

[0133] The electronic device may include a processor 71 and a memory 72 storing computer program instructions.

[0134] Specifically, the above-mentioned processor 71 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the present application.

[0135] Among them, the memory 72 may include a mass storage for data or instructions. By way of example and not limitation, the memory 72 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disc, a magneto-optical disc, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In suitable cases, the memory 72 may include removable or non-removable (or fixed) media. In suitable cases, the memory 72 may be internal or external to the data processing device. In a particular embodiment, the memory 72 is a non-volatile memory. In a particular embodiment, the memory 72 includes a read-only memory (ROM) and a random access memory (RAM). In suitable cases, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or a flash memory, or a combination of two or more of these. In suitable cases, the RAM may be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM may be a fast page mode dynamic random access memory (FPMDRAM), an extended date out dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0136] The memory 72 can be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 71.

[0137] The processor 71 reads and executes the computer program instructions stored in the memory 72 to implement the surgical anesthesia information warning method in the first embodiment above.

[0138] In some of the embodiments, the electronic device may further include a communication interface 73 and a bus 70. Among them, as Figure 3 shown, the processor 71, the memory 72, and the communication interface 73 are connected through the bus 70 and complete communication with each other.

[0139] The communication interface 73 is used to implement communication between the various modules, devices, units, and / or devices in the present application. The communication interface 73 can also implement data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0140] The bus 70 includes hardware, software, or both, and couples components of the device to each other. The bus 70 includes, but is not limited to, at least one of the following: Data Bus, Address Bus, Control Bus, Expansion Bus, Local Bus. By way of example and not limitation, the bus 70 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable bus or a combination of two or more of these. In suitable cases, the bus 70 may include one or more buses. Although this application describes and illustrates specific buses, this application contemplates any suitable bus or interconnect.

[0141] The electronic device can obtain a surgical anesthesia information warning system and execute the surgical anesthesia information warning method of Embodiment 1.

[0142] In addition, in combination with the surgical anesthesia information warning method in Embodiment 1 above, this application can be implemented by providing a storage medium. Computer program instructions are stored on the storage medium; when the computer program instructions are executed by a processor, the surgical anesthesia information warning method of Embodiment 1 above is implemented.

[0143] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0144] The above-described embodiments merely represent several implementation manners of the present invention. The descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the patent for the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention should be subject to the appended claims.

Claims

1. A surgical anesthesia information warning method, characterized in that, The method includes: Obtaining the structured data of the patient and collecting the unstructured data during the operation; Encoding the structured data, obtaining the BMI index of the patient based on the encoded structured data, and processing the unstructured data to obtain the vectorized unstructured data of a preset length; Establishing a mining and analysis model based on the vectorized representation of the surgical data in the vectorized unstructured data and the linear functions of several factor features, and mining and analyzing the potential state of the patient in combination with the mining and analysis model and the vectorized unstructured data; Collecting the anesthesia physiological signals and electroencephalogram signals under anesthesia of the patient based on the unstructured data, and classifying the anesthesia state according to the anesthesia physiological signals and the electroencephalogram signals to obtain the classified anesthesia state; Collecting an anesthesia depth data set based on the potential state and the classified anesthesia state, extracting anesthesia features according to the anesthesia depth data set, generating feature input samples according to the anesthesia features, and generating sample labels; Constructing an early warning model according to a preset machine learning model and the sample labels, and inputting the feature input samples into the early warning model so that the early warning model generates early warning information.

2. The surgical anesthesia information warning method according to claim 1, wherein The steps of obtaining the structured data of the patient and collecting the unstructured data during the operation include: Obtaining the basic situation of the patient, where the basic situation includes gender, height, age, weight, and department, and the gender, height, age, weight, and department constitute the structured data; Collecting the unstructured data during the operation through a monitor, an anesthesia monitor, and an OAIS, where the unstructured data includes physiological monitoring sequence data, anesthesia physiological signals, electroencephalogram signals, and surgical sequence data.

3. The surgical anesthesia information warning method according to claim 1, characterized in that, The steps of encoding the structured data, obtaining the BMI index of the patient based on the encoded structured data, and processing the unstructured data to obtain the vectorized unstructured data of a preset length include: Extracting the height information and weight information in the unstructured data, and calculating the BMI index of the patient through the BMI calculation formula; Combining different texts in the unstructured data into a text set, and segmenting the text set to obtain a number of words; Performing vector representation on the number of words to obtain vectorized word numbers, and mapping the vectorized word data to the same length. The expression of the vector representation process is: where WT represents the weighted vector representation of words, N represents the number of words, tfidfi i represents the TFIDF value of the i-th word, WV i represents the word vector of the i-th word.

4. The surgical anesthesia information warning method according to claim 1, wherein, The steps of establishing a mining and analysis model based on the vectorized representation of the surgical data in the vectorized unstructured data and the linear functions of several factor features, and mining and analyzing the potential state of the patient in combination with the mining and analysis model and the vectorized unstructured data include: Taking the vectorized representation of the surgical data in the unstructured data as a feature item, and establishing a mining and analysis model through the feature item combined with the linear functions of several factor features. The expression of the mining and analysis model is: Wherein, represents the predicted value of the i-th sample, D represents the feature dimension, and 0 ij represents the j-th feature value of the i-th sample, and w j is the weight corresponding to the j-th feature, X i represents the feature row vector of the i-th sample, W represents the weight vector, and W T represents the transposed weight vector; Based on the mining analysis model and the vectorized unstructured data, potential high-risk situations, respiratory system conditions, circulatory system conditions, and nervous system conditions of the patient during the operation are mined.

5. The surgical anesthesia information warning method according to claim 1, wherein, The step of classifying the anesthesia state according to the anesthesia physiological signal and the electroencephalogram signal to obtain the classified anesthesia state includes: Drawing a depth of consciousness map based on the anesthesia physiological signal and the electroencephalogram signal; Classifying the depth of consciousness map according to the level of consciousness to obtain depth of consciousness maps with different depth values, so as to obtain different anesthesia states of the patient.

6. The surgical anesthesia information warning method according to claim 1, characterized in that The step of extracting anesthesia features according to the anesthesia depth data set, generating feature input samples according to the anesthesia features, and generating sample labels includes: Extracting anesthesia parameters and situation features in the anesthesia depth data set, and generating feature input samples based on the anesthesia parameters and situation features; Obtaining the generation time of the feature input sample, and generating a sample label for the feature input sample according to the generation time.

7. The surgical anesthesia information warning method according to claim 1, wherein The expression of the early warning model is: In the formula, represents the prediction result, f represents a preset machine learning model, H represents the length of the observation window, B represents the interval time, P represents the length of the prediction window, x represents the feature input sample set, and t represents the sample label.

8. A surgical anesthesia information warning system, characterized in that, The system includes: An acquisition module, configured to acquire structured data of the patient and acquire unstructured data during the operation; An encoding processing module, configured to encode the structured data, obtain the BMI index of the patient based on the encoded structured data, and process the unstructured data to obtain vectorized unstructured data of a preset length; A mining module, configured to establish a mining analysis model based on the vectorized representation of the surgical data in the vectorized unstructured data and the linear functions of several factor features, and mine and analyze the potential state of the patient in combination with the mining analysis model and the vectorized unstructured data; An acquisition and classification module, configured to acquire the anesthesia physiological signal of the patient and the electroencephalogram signal under the anesthesia state based on the unstructured data, and classify the anesthesia state according to the anesthesia physiological signal and the electroencephalogram signal to obtain the classified anesthesia state; A collection and extraction module, configured to collect an anesthesia depth data set based on the potential state and the classified anesthesia state, extract anesthesia features according to the anesthesia depth data set, generate feature input samples according to the anesthesia features, and generate sample labels; A construction and generation module, configured to construct an early warning model according to a preset learning model, and input the feature input sample set and the sample label into the early warning model, so that the early warning model generates early warning information.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the surgical anesthesia information early warning method according to any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the surgical anesthesia information early warning method according to any one of claims 1 to 7.

Images