A Smart Analysis Method for Pulmonary Tuberculosis Patient Medical Records for Hemoptysis Early Warning

Abstract

The present invention relates to the field of medical data processing technology, and in particular to an intelligent analysis method for medical records of pulmonary tuberculosis patients for hemoptysis early warning, wherein the medical records of historical patients in a database are extracted to obtain a standard hemoptysis deterioration curve and the personal bad habit value of each historical patient; and the medical records of patients to be examined are extracted to obtain an initial local hemoptysis curve and the personal bad habit value of the patient to be examined; and then the comparison segment in the standard hemoptysis deterioration curve and the revised local hemoptysis curve are determined, and the hemoptysis risk index of the patient to be examined is determined according to the similarity between the revised local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve, and the overall distribution level difference between the comparison segment in the revised local hemoptysis curve and the standard hemoptysis deterioration curve, and it is determined whether to perform hemoptysis early warning. The present invention effectively improves the accuracy of early warning of hemoptysis using the patient's medical records.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and in particular to an intelligent analysis method of medical records of pulmonary tuberculosis patients for early warning of hemoptysis. Background Art

[0002] Tuberculosis is a chronic infectious disease, and hemoptysis is a common symptom of tuberculosis. If it is not treated in time, suffocation may occur within minutes, leading to the patient's death. Therefore, when a patient enters the hospital for treatment, it is usually necessary to determine whether the patient will have hemoptysis.

[0003] In the prior art, when judging whether a patient will have hemoptysis, one way is to obtain images through chest X-rays and other examination methods, and then analyze them based on the doctor's experience. This method usually takes a long time, resulting in a lack of timeliness in the warning of hemoptysis for the patient. Another way is for the doctor to make a hemoptysis judgment based on the patient's medical records. This method often requires the doctor to have more clinical experience and is highly subjective, resulting in a lower accuracy rate of the warning. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent analysis method of the medical records of tuberculosis patients for hemoptysis warning, which is used to solve the problem of low accuracy of hemoptysis warning based on the existing medical records of patients.

[0005] In order to solve the above technical problems, in a first aspect, the present invention provides an intelligent analysis method of medical records of pulmonary tuberculosis patients for early warning of hemoptysis, comprising the following steps:

[0006] Data were extracted from the medical records of historical patients in the database to obtain the standard hemoptysis exacerbation curve and the individual bad habit value of each historical patient;

[0007] Data was extracted from the medical records of the patients to be examined to obtain the initial local hemoptysis curve and the personal bad habit values ​​of the patients to be examined;

[0008] Determining a comparison segment in the standard hemoptysis deterioration curve according to a difference in change fluctuations between the standard hemoptysis deterioration curve and the initial local hemoptysis curve;

[0009] According to the deviation of the individual bad habit values ​​of the patient to be examined relative to all historical patients, the initial local hemoptysis curve is corrected to obtain a corrected local hemoptysis curve;

[0010] Determining the hemoptysis risk index of the patient to be examined according to the similarity between the contrast segment in the modified local hemoptysis curve and the standard hemoptysis deterioration curve, and the difference in overall distribution level between the contrast segment in the modified local hemoptysis curve and the standard hemoptysis deterioration curve;

[0011] Whether to perform hemoptysis warning is determined according to the hemoptysis risk indicator.

[0012] In combination with the first aspect above, in some possible implementations, extracting data from the medical records of historical patients in the database to obtain a standard hemoptysis deterioration curve and a personal bad habit value of each historical patient includes:

[0013] Extract the medical records of historical patients in the database to obtain the personal bad habit values, hemoptysis indicators and the illness time corresponding to the hemoptysis indicators of historical patients;

[0014] According to the hemoptysis indexes of all historical patients and their corresponding illness time, curve fitting is performed to obtain a standard hemoptysis deterioration curve in which the hemoptysis indexes change with illness time.

[0015] In combination with the first aspect above, in some possible implementations, determining the comparison segment in the standard hemoptysis exacerbation curve includes:

[0016] Determining the hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined, and determining the hemoptysis index change rate of the standard hemoptysis deterioration curve;

[0017] Determine the end time point of the comparison segment in the standard hemoptysis deterioration curve according to the difference between the hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined and the hemoptysis index change rate of the standard hemoptysis deterioration curve, as well as the total length of illness time and the end illness time of the initial local hemoptysis curve of the patient to be examined;

[0018] The starting onset time of the initial local hemoptysis curve of the patient to be examined is used as the starting time point of the comparison segment in the standard hemoptysis deterioration curve, and the comparison segment in the standard hemoptysis deterioration curve is determined according to the starting time point and the ending time point of the comparison segment in the standard hemoptysis deterioration curve.

[0019] In combination with the first aspect above, in some possible implementations, obtaining the hemoptysis index includes:

[0020] Extracting hemoptysis information from medical records, and determining various hemoptysis vectors according to the hemoptysis information, wherein the hemoptysis vectors are composed of at least two different types of hemoptysis characteristic values;

[0021] Perform one-dimensional transformation on the hemoptysis vector to obtain a hemoptysis index.

[0022] In combination with the first aspect above, in some possible implementations, the different types of hemoptysis characteristic values ​​include at least a hemoptysis volume characteristic value, a hemoptysis color depth characteristic value, and a hemoptysis frequency characteristic value.

[0023] In combination with the first aspect above, in some possible implementations, determining the personal bad habit value includes:

[0024] Extract personal habit information from medical records, input the personal habit information into a semantic recognition network, and output a corresponding personal bad habit value from the semantic recognition network, where the personal bad habit value ranges from 0 to 1.

[0025] In combination with the first aspect above, in some possible implementations, determining the end time point of the comparison segment in the standard hemoptysis exacerbation curve includes:

[0026] Determining a time coefficient according to a difference between a hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined and a hemoptysis index change rate of the standard hemoptysis deterioration curve;

[0027] Calculating the product of the total length of the initial local hemoptysis curve of the patient to be examined and the time coefficient to obtain an adjustment value of the sick time;

[0028] The sum of the end time of the initial local hemoptysis curve of the patient to be examined and the adjustment value of the disease time is calculated to obtain the end time point of the comparison segment in the standard hemoptysis deterioration curve.

[0029] In combination with the first aspect, in some possible implementations, the initial local hemoptysis curve is corrected in combination with the deviation of the personal bad habit value of the patient to be examined relative to all historical patients to obtain a corrected local hemoptysis curve, including:

[0030] According to the overall distribution level and dispersion of the individual bad habit values ​​of all historical patients, the mean value of bad habits and the dispersion index of bad habit values ​​are determined;

[0031] Determine the hemoptysis index correction parameter of the patient to be examined according to the difference between the individual bad habit value of the patient to be examined and the mean value of the bad habit value, and the size of the bad habit value discrete index;

[0032] According to the hemoptysis index correction parameter, the hemoptysis index corresponding to each disease time in the initial local hemoptysis curve is updated to obtain a corrected local hemoptysis curve.

[0033] In combination with the first aspect above, in some possible implementations, determining the hemoptysis risk indicator of the patient to be examined includes:

[0034] Determine the DTW distance between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve, and perform negative correlation mapping on the DTW distance to obtain a similarity index value;

[0035] Determine the average values ​​of the hemoptysis indexes corresponding to all the disease time in the comparison segments of the modified local hemoptysis curve and the standard hemoptysis deterioration curve, respectively, and obtain the first hemoptysis index average value and the second hemoptysis index average value;

[0036] Performing mapping processing on the difference between the first hemoptysis index mean and the second hemoptysis index mean to obtain a difference index value;

[0037] The hemoptysis risk index of the patient to be examined is determined according to the similarity index value and the difference index value, and both the similarity index value and the difference index value are positively correlated with the hemoptysis risk index.

[0038] In combination with the first aspect above, in some possible implementations, judging whether to perform hemoptysis warning according to the hemoptysis risk indicator includes:

[0039] Determining whether the hemoptysis risk index is greater than the hemoptysis risk threshold;

[0040] If the hemoptysis risk index is greater than the hemoptysis risk threshold, a hemoptysis warning is performed.

[0041] In order to solve the above technical problems, in a second aspect, the present invention also provides an intelligent medical record analysis device for tuberculosis patients for early warning of hemoptysis, the device comprising:

[0042] The historical data acquisition module is used to: extract the medical records of historical patients in the database to obtain the standard hemoptysis deterioration curve and the personal bad habit value of each historical patient;

[0043] The real-time data acquisition module is used to extract the medical records of the patient to be examined, and obtain the initial local hemoptysis curve and the personal bad habit value of the patient to be examined;

[0044] A comparison segment acquisition module is used to: determine the comparison segment in the standard hemoptysis deterioration curve according to the change fluctuation difference between the standard hemoptysis deterioration curve and the initial local hemoptysis curve;

[0045] A local hemoptysis curve acquisition module is used to: correct the initial local hemoptysis curve in combination with the deviation of the individual bad habit values ​​of the patient to be examined relative to all historical patients, to obtain a corrected local hemoptysis curve;

[0046] A risk index acquisition module, used to determine the hemoptysis risk index of the patient to be examined according to the similarity between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve, and the overall distribution level difference between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve;

[0047] The judgment and early warning module is used to: judge whether to issue a hemoptysis early warning according to the hemoptysis risk indicator.

[0048] To solve the above technical problems, in a third aspect, the present invention also provides a tuberculosis patient medical record intelligent analysis system for hemoptysis warning, comprising a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the method in the above first aspect or any possible implementation of the first aspect.

[0049] To solve the above technical problems, in a fourth aspect, the present invention further provides a computer program product, which includes: a computer program code, when the computer program code runs on a computer, enables the computer to execute the method in the above first aspect or any possible implementation of the first aspect.

[0050] To solve the above technical problems, in a fifth aspect, the present invention further provides a computer-readable storage medium, which stores a computer program code. When the computer program code runs on a computer, the computer executes the method in the above first aspect or any possible implementation of the first aspect.

[0051] The present invention has the following beneficial effects: by extracting data from the medical records of historical patients, a standard hemoptysis deterioration curve and a personal bad habit value of each historical patient are obtained, and the standard hemoptysis deterioration curve reflects the standard deterioration of the patient's condition without intervention. At the same time, data is extracted from the medical records of the patient to be examined to obtain an initial local hemoptysis curve and a personal bad habit value of the patient to be examined. The initial local hemoptysis curve reflects the current deterioration of the condition of the patient to be examined. Since the disease progression speed of different patients is different, some patients even reach a more serious degree in a short time. Therefore, according to the change fluctuation difference between the standard hemoptysis deterioration curve and the initial local hemoptysis curve, the deterioration of the condition of the patient to be examined is evaluated, so as to determine the comparison section of the initial local hemoptysis curve in the standard hemoptysis deterioration curve. Considering that personal habit history information often affects the development of pulmonary tuberculosis, the personal habit history information of different patients is different, and it is also necessary to combine the personal bad habit value of the patient to be examined, that is, to compare the personal bad habit value of the patient to be examined with the personal bad habit value of the historical patient in the database, analyze the influence of its bad personal habits on the initial local hemoptysis curve, and then correct the initial local hemoptysis curve, so as to obtain the final accurate corrected local hemoptysis curve. When the degree of similarity between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve is higher, it means that the patient to be examined is more in line with the standard disease deterioration law under the condition of no intervention. At the same time, when the overall distribution level of the modified local hemoptysis curve is higher than the overall distribution level of the comparison segment in the standard hemoptysis deterioration curve, it means that the degree of deterioration of the patient to be tested is higher, and the corresponding hemoptysis risk is greater. Therefore, the hemoptysis risk index of the patient to be examined can be determined, so as to accurately judge whether to perform hemoptysis warning. The present invention extracts data from the medical records of historical patients in the database and the medical records of the patient to be examined, thereby determining the local hemoptysis curve affected by personal bad habits and the standard hemoptysis deterioration curve reflecting the standard disease deterioration of the patient under the condition of no intervention, combining the degree of similarity between the comparison segment in the local hemoptysis curve and the standard hemoptysis deterioration curve, and the overall distribution level difference between the comparison segment in the local hemoptysis curve and the standard hemoptysis deterioration curve, to determine the hemoptysis risk index of the patient to be examined, and then accurately determine whether to perform hemoptysis warning, effectively improving the accuracy of hemoptysis warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 A flowchart of the steps of a method for intelligent analysis of medical records of tuberculosis patients for early warning of hemoptysis according to an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of the structure of an intelligent medical record analysis device for tuberculosis patients for early warning of hemoptysis according to an embodiment of the present invention;

[0055] Figure 3 The present invention is a schematic diagram of the structure of an intelligent analysis system for medical records of tuberculosis patients for early warning of hemoptysis according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to clearly illustrate the technical features of the present invention, the present invention is described in detail below through specific implementation methods in conjunction with the accompanying drawings.

[0057] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0058] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0059] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0060] It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0061] Although operations or steps are described in a specific order in the drawings in the embodiments of the present invention, it should not be understood that it is required to perform these operations or steps in the specific order shown or in a serial order, or that all the operations or steps shown must be performed to obtain the desired results. In the embodiments of the present invention, these operations or steps may be performed in series; these operations or steps may also be performed in parallel; or some of these operations or steps may be performed.

[0062] At the same time, it is understood that the data involved in the technical solution of the present invention (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of the relevant laws, regulations and relevant provisions. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meanings as those generally understood by technicians in the technical field of the present invention, and all parameters or indicators in the formulas involved in the present invention are normalized values ​​that eliminate the influence of dimensions.

[0063] In order to solve the problem of low accuracy of the existing early warning of hemoptysis based on the patient's medical records, an embodiment of the present invention provides an intelligent analysis method of the medical records of pulmonary tuberculosis patients for early warning of hemoptysis. The method first extracts data from the medical records of historical patients in the database to construct a standard hemoptysis deterioration curve and the personal bad habit value of each historical patient, and at the same time extracts data from the medical records of the patient to be examined to obtain an initial local hemoptysis curve and the personal bad habit value of the patient to be examined; then, according to the change fluctuation difference between the standard hemoptysis deterioration curve and the initial local hemoptysis curve, the comparison segment in the standard hemoptysis deterioration curve is determined; then, the personal bad habit value of the patient to be examined is compared with the personal bad habit value of the historical patients in the database, and the influence of the bad personal habits on the initial local hemoptysis curve is analyzed, and then the initial local hemoptysis curve is corrected; finally, according to the similarity between the corrected local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve, and the overall distribution level difference between the corrected local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve, the hemoptysis risk index of the patient to be examined is determined, thereby realizing early warning of hemoptysis. The embodiment of the present invention analyzes the medical records of historical patients in the database and the medical records of patients to be examined to accurately determine the hemoptysis risk index, thereby realizing hemoptysis early warning. This can effectively avoid the problem of low accuracy of hemoptysis risk early warning due to lack of experience of doctors, improve the accuracy of hemoptysis early warning, and assist doctors in diagnosis and treatment.

[0064] The following will introduce in detail a method for intelligent analysis of medical records of pulmonary tuberculosis patients for early warning of hemoptysis provided by an embodiment of the present invention in conjunction with the accompanying drawings.

[0065] Figure 1The basic flow chart of a method for intelligent analysis of medical records of pulmonary tuberculosis patients for early warning of hemoptysis provided by an embodiment of the present invention is shown as follows: Figure 1 As shown, the method specifically comprises the following steps:

[0066] Step S100: extracting medical records of historical patients in the database to obtain a standard hemoptysis deterioration curve and a personal bad habit value of each historical patient.

[0067] The medical records of pulmonary tuberculosis patients (hereinafter referred to as historical patients) at the time of admission are counted, and the medical records of each historical patient are digitally stored. The medical records mainly include medical records before admission and medical records during the treatment process. The medical records need to record the identity information of each historical patient, as well as the symptom information and hemoptysis information before admission (during the period before the visit), personal habit history, and whether hemoptysis occurs after admission and other related information. Among them, the identity information includes the name, gender, age and other information of the historical patient; the symptom information refers to the tuberculosis-related symptoms that appeared before the patient was admitted to the hospital and the time point when the symptoms appeared; the hemoptysis information refers to the description of the hemoptysis condition before the patient was admitted to the hospital, which includes the time point of each hemoptysis, and the corresponding hemoptysis amount, color, frequency and other data for each hemoptysis; the personal habit history includes whether there is a history of smoking for many years, work experience in dusty environment, and bad living habits and other personal habit information that affects the condition of pulmonary tuberculosis. These medical records are stored in the database, and the database is used as the standard database for pulmonary tuberculosis patients. The larger the number of medical records in the database, the better.

[0068] Furthermore, when the medical records of historical patients stored in the database are sufficient, the age of the historical patients is determined according to the identity information in the medical records, and the medical records of historical patients belonging to the same age group are stored together, so that the medical records of all historical patients corresponding to each age group can be used to facilitate the subsequent call of the medical records of historical patients, so as to more accurately warn of hemoptysis. In the embodiment of the present invention, 0-6 years old, 7-12 years old, 13-17 years old, 18-45 years old, 46-69 years old, and 70 years old and later are respectively regarded as the same age group.

[0069] Since the above-mentioned database storing the medical records of historical patients upon admission is constructed, when a new patient (hereinafter referred to as the patient to be examined) is admitted to the hospital, the medical records of the patient to be examined upon admission can be obtained in accordance with the above-mentioned method of obtaining the medical records of each historical patient. By comparing the medical records of the patient to be examined with the medical records of historical patients in the database, it is possible to quickly and accurately determine whether the patient to be examined will have the risk of hemoptysis, and issue a hemoptysis warning.

[0070] To achieve the above purpose, firstly, the medical records of historical patients in the database are extracted to obtain the standard hemoptysis deterioration curve and the personal bad habit value of each historical patient. The implementation steps include:

[0071] Extract the medical records of historical patients in the database to obtain the personal bad habit values, hemoptysis indicators and the illness time corresponding to the hemoptysis indicators of historical patients;

[0072] According to the hemoptysis indexes of all historical patients and their corresponding illness time, curve fitting is performed to obtain a standard hemoptysis deterioration curve in which the hemoptysis indexes change with illness time.

[0073] For the above steps, the age of the patient to be examined is determined based on the identity information in the medical records of the patient to be examined, so that the age group to which the age of the patient to be examined belongs can be determined, and the medical records of all historical patients corresponding to the age group in the database can be determined. It should be understood that the medical records of all historical patients below refer to the medical records of all historical patients corresponding to the age group to which the age of the patient to be examined belongs in the database, and these historical patients all had hemoptysis before admission.

[0074] Pulmonary tuberculosis develops slowly, and the early symptoms are not obvious. Usually, patients are admitted to the hospital for treatment only after entering the onset period, accompanied by symptoms such as long-term low fever and fatigue, and may have symptoms such as coughing and hemoptysis. For the medical records of all historical patients corresponding to the age group of the patient to be examined, the hemoptysis information of each historical patient before admission is extracted, and the hemoptysis amount, color, frequency and other data corresponding to each hemoptysis can be determined based on the hemoptysis information. If the hemoptysis amount is large, the color is red and dark, and the frequency is high, the historical patient has a high risk of long-term hemoptysis. The hemoptysis amount, color, frequency and other data corresponding to each hemoptysis are converted into numerical values ​​to obtain the hemoptysis characteristic value corresponding to each data, namely, the hemoptysis amount characteristic value, the hemoptysis color depth characteristic value and the hemoptysis frequency characteristic value. The value range of each hemoptysis characteristic value is 0-1. The larger the value of the hemoptysis characteristic value, the larger the hemoptysis amount, the redder the color or the higher the frequency. In an embodiment of the present invention, a LSTM (Recurrent Neural Network Long Short Term Memory Networks) semantic recognition network is used to convert the data such as the amount of hemoptysis, color, frequency, etc. corresponding to each hemoptysis, that is, the data such as the amount of hemoptysis, color, frequency, etc. corresponding to each hemoptysis are respectively input into the corresponding pre-trained LSTM semantic recognition network, and the LSTM semantic recognition network outputs the corresponding hemoptysis feature values ​​such as the amount of hemoptysis feature value, the hemoptysis color depth feature value, and the hemoptysis frequency feature value. Since the training process of the LSTM semantic recognition network belongs to the prior art, it will not be repeated here. In this way, the hemoptysis feature values ​​corresponding to the amount of hemoptysis, color, frequency, etc. corresponding to each hemoptysis of all historical patients in the database can be obtained.

[0075] A hemoptysis vector is constructed using the hemoptysis amount characteristic value, hemoptysis color depth characteristic value, hemoptysis frequency characteristic value and other hemoptysis characteristic values ​​corresponding to each hemoptysis, so that the hemoptysis vectors corresponding to each historical patient can be obtained. Each hemoptysis vector is converted into one dimension, so that a hemoptysis index can be obtained. In an embodiment of the present invention, a pre-trained Embedding network is used to perform a one-dimensional conversion on each hemoptysis vector, that is, each hemoptysis vector is input into the Embedding network, and the Embedding network outputs the corresponding one-dimensional hemoptysis index. Since the training process of the Embedding network belongs to the prior art, it will not be repeated here. In this way, a hemoptysis index corresponding to each hemoptysis of all historical patients in the database can be obtained.

[0076] Since the medical records of historical patients record the tuberculosis-related symptoms that occurred before hospitalization and the time point of the symptoms, the time point corresponding to the first occurrence of tuberculosis-related symptoms is used as the reference time point, so that the time interval between the time point corresponding to each hemoptysis and the reference time point can be determined. At the same time, since a one-dimensional hemoptysis index is determined each time hemoptysis occurs, the time interval corresponding to each one-dimensional hemoptysis index of historical patients in the database can be obtained, and this time interval is called the illness time corresponding to the hemoptysis index.

[0077] At the same time, since the medical records of historical patients record the personal habit history information of each historical patient, and these personal habit history information often affect the development of tuberculosis, for example, when there is a history of smoking and working in a dusty environment for many years, the disease progression of such patients is faster and the possibility of hemoptysis is greater. Therefore, the personal habit history information in the medical records of historical patients in the database is extracted, and the personal habit history information is numerically converted to obtain the corresponding personal bad habit value. The value range of the personal bad habit value is 0-1. The larger the value of the personal bad habit value, the more serious the patient's bad habit is, and the faster the disease progression may be. In the embodiment of the present invention, the same as the above-mentioned method of obtaining the characteristic value of each type of hemoptysis, the LSTM semantic recognition network is also used to convert the personal habit history information numerically, that is, the personal habit history information of each historical patient is input into the corresponding pre-trained LSTM semantic recognition network, and the LSTM semantic recognition network outputs the corresponding personal bad habit value. In this way, the personal bad habit values ​​of all historical patients in the database can be obtained.

[0078] For the medical records of all historical patients corresponding to the age group to which the age of the patient to be examined belongs in the database, the hemoptysis index of all historical patients obtained and the illness time corresponding to the hemoptysis index are curve-fitted, so as to obtain a standard hemoptysis deterioration curve in which the hemoptysis index changes with the illness time. In an embodiment of the present invention, the hemoptysis index is used as the ordinate and the illness time is used as the abscissa to determine the data points corresponding to the hemoptysis index of all historical patients obtained and the illness time corresponding to the hemoptysis index, and the least squares method is used to fit all data points to obtain a standard hemoptysis deterioration curve. It should be understood that the standard hemoptysis deterioration curve is obtained using the medical records of existing hemoptysis patients who have been treated at the time of admission. One patient may correspond to one or more data points, and one data point is determined by the hemoptysis index corresponding to the hemoptysis vector during hemoptysis and the corresponding illness time. The standard hemoptysis deterioration curve refers to the standard disease deterioration curve of the patient without intervention.

[0079] Step S200: extracting data from the medical records of the patient to be examined, and obtaining an initial local hemoptysis curve and a personal bad habit value of the patient to be examined.

[0080] Due to the different stages of patients' medical treatment, some patients only seek medical treatment after experiencing multiple severe hemoptysis. For such patients, the medical records obtained during their medical treatment are extracted to obtain the corresponding local hemoptysis curve. If the local hemoptysis curve is closer to the corresponding segment of the standard hemoptysis deterioration curve and the distribution level is higher, it means that the patient's condition is worsening and the risk of hemoptysis is higher.

[0081] In order to obtain the local hemoptysis curve of the patient to be examined, the medical records of the patient when he is admitted to the hospital are extracted to obtain the personal bad habit value, hemoptysis index and the illness time corresponding to the hemoptysis index of the patient to be examined. Since the method of obtaining the medical records of the patient to be examined when he is admitted to the hospital, extracting the data from the medical records, obtaining the personal bad habit value, hemoptysis index and the illness time corresponding to the hemoptysis index of the patient to be examined is exactly the same as the method of obtaining the medical records of historical patients in the database when they are admitted to the hospital, extracting the data from the medical records, obtaining the personal bad habit value, hemoptysis index and the illness time corresponding to the hemoptysis index of the historical patients, and will not be repeated here.

[0082] According to the acquired hemoptysis index of the patient to be examined and the illness time corresponding to the hemoptysis index, curve fitting is performed to obtain an initial local hemoptysis curve of the hemoptysis index changing with the illness time. In the embodiment of the present invention, the method of obtaining the standard hemoptysis deterioration curve is the same as the above-mentioned method, that is, the hemoptysis index is used as the ordinate and the illness time is used as the abscissa to determine the data points corresponding to all the hemoptysis indexes of the patient to be examined and the illness time corresponding to the hemoptysis index, and the least squares method is used to perform curve fitting on all the data points to obtain the initial local hemoptysis curve of the patient to be examined. The first data point of the initial local hemoptysis curve is the data point corresponding to the first hemoptysis index of the patient to be examined and the illness time corresponding to the first hemoptysis index, and the last data point is the data point corresponding to the last hemoptysis index of the patient to be examined and the illness time corresponding to the last hemoptysis index.

[0083] Step S300: determining a comparison segment in the standard hemoptysis deterioration curve according to the difference in change fluctuation between the standard hemoptysis deterioration curve and the initial local hemoptysis curve.

[0084] Since the disease progression speeds of different patients are different, some patients even reach a more serious degree in a relatively short period of time. Therefore, it is necessary to determine the stage position of the local hemoptysis curve in the standard hemoptysis deterioration curve according to the disease deterioration rate of the patient to be examined, that is, it is necessary to determine the starting time point and the ending time point of the comparison segment of the standard hemoptysis deterioration curve for comparison with the local hemoptysis curve, so as to obtain the comparison segment of the local hemoptysis curve in the standard hemoptysis deterioration curve.

[0085] In the embodiment of the present invention, the comparison segment of the local hemoptysis curve in the standard hemoptysis deterioration curve is determined, and the implementation steps include:

[0086] Determining the hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined, and determining the hemoptysis index change rate of the standard hemoptysis deterioration curve;

[0087] Determine the end time point of the comparison segment in the standard hemoptysis deterioration curve according to the difference between the hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined and the hemoptysis index change rate of the standard hemoptysis deterioration curve, as well as the total length of illness time and the end illness time of the initial local hemoptysis curve of the patient to be examined;

[0088] The starting onset time of the initial local hemoptysis curve of the patient to be examined is used as the starting time point of the comparison segment in the standard hemoptysis deterioration curve, and the comparison segment in the standard hemoptysis deterioration curve is determined according to the starting time point and the ending time point of the comparison segment in the standard hemoptysis deterioration curve.

[0089] For the above steps, when determining the end time point of the comparison segment of the local hemoptysis curve in the standard hemoptysis deterioration curve, if the initial local hemoptysis curve deterioration rate of the patient to be tested is greater than the standard hemoptysis deterioration curve deterioration rate, it means that the disease progression of the patient to be tested is faster, and the time of the standard hemoptysis deterioration curve to be compared with it will also be correspondingly longer. Otherwise, it means that the disease progression of the patient to be tested is slower, that is, the condition of the patient to be tested is better, and it takes a long time for the deterioration to be similar to the short-term deterioration of other ordinary patients. At this time, the time of the standard hemoptysis deterioration curve to be compared with it will also be correspondingly shortened.

[0090] To this end, the hemoptysis index change rate of the patient to be examined is determined based on the initial local hemoptysis curve of the patient to be examined, that is, the difference between the maximum and minimum values ​​of all hemoptysis indicators in the initial local hemoptysis curve is determined, and the difference is divided by the total length of the illness time of the local hemoptysis curve, the total length of the illness time refers to the difference between the illness time corresponding to the last hemoptysis of the patient to be examined and the illness time corresponding to the first hemoptysis, and the obtained quotient is the hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined. At the same time, the hemoptysis index change rate of the standard hemoptysis deterioration curve is determined, that is, the difference between the maximum and minimum values ​​of all hemoptysis indicators in the standard hemoptysis deterioration curve is determined, and the difference is divided by the total length of the illness time of the standard hemoptysis deterioration curve, the total length of the illness time refers to the illness time corresponding to the last hemoptysis indicator on the standard hemoptysis deterioration curve, and the obtained quotient is the hemoptysis index change rate of the standard hemoptysis deterioration curve. The change rate of the hemoptysis index of the patient to be examined is compared with the change rate of the hemoptysis index of the standard hemoptysis deterioration curve. When the change rate of the hemoptysis index of the patient to be examined is greater than that of the standard hemoptysis deterioration curve, the end time point of the comparison segment of the local hemoptysis curve in the standard hemoptysis deterioration curve is later, thereby determining the end time point of the comparison segment of the initial local hemoptysis curve of the patient to be examined in the standard hemoptysis deterioration curve.

[0091] In the embodiment of the present invention, the end time point of the comparison segment of the initial local hemoptysis curve of the patient to be examined in the standard hemoptysis deterioration curve is determined, and the corresponding calculation formula is:

[0092] ;

[0093] in, Indicates patients to be examined The initial local hemoptysis curve is at the end time point of the comparison segment in the standard hemoptysis exacerbation curve; Indicates patients to be examined The duration of illness corresponding to the last hemoptysis index of the initial local hemoptysis curve; Indicates patients to be examined The total length of illness from the initial local hemoptysis curve; Indicates patients to be examined The rate of change of hemoptysis index of the initial local hemoptysis curve; It represents the change rate of hemoptysis index of the standard hemoptysis deterioration curve; represents the sickness time adjustment value, represents the time coefficient, Represents the inverse tangent function.

[0094] In the above calculation formula, It indicates the adjustment value of the illness time based on the illness time corresponding to the last hemoptysis index of the patient to be examined. When the difference between the hemoptysis index change rate of the patient to be examined and the hemoptysis index change rate of the standard hemoptysis deterioration curve is greater, it means that the patient to be examined is The faster the disease progresses, the longer the standard hemoptysis deterioration curve that needs to be compared is. The closer it is to 1, the closer the adjusted value of illness time is to the patient to be examined. The total length of illness time of the initial local hemoptysis curve On the contrary, when the difference between the change rate of hemoptysis index of the patient to be examined and the change rate of hemoptysis index of the standard hemoptysis deterioration curve is smaller, it means that the patient to be examined is The slower the disease progresses, the closer the time adjustment is to the patient to be examined. The total length of illness time of the initial local hemoptysis curve By adding the time adjustment amount to the illness time corresponding to the last hemoptysis index of the patient to be examined, the patient's The initial local hemoptysis curve is at the end time point of the comparison segment in the standard hemoptysis exacerbation curve.

[0095] After obtaining the end time point of the comparison segment of the initial local hemoptysis curve of the patient to be examined in the standard hemoptysis deterioration curve in the above manner, the starting illness time of the initial local hemoptysis curve of the patient to be examined is used as the starting time point of the comparison segment in the standard hemoptysis deterioration curve, and the curve segment corresponding to the starting time point and the ending time point is obtained on the standard hemoptysis deterioration curve. The curve segment is the comparison segment of the local hemoptysis curve of the patient to be examined in the standard hemoptysis deterioration curve.

[0096] Step S400: Based on the deviation of the personal bad habit values ​​of the patient to be examined relative to all historical patients, the initial local hemoptysis curve is corrected to obtain a corrected local hemoptysis curve.

[0097] Taking into account that personal habit history information often affects the development of pulmonary tuberculosis, and different patients have different personal habit history information, it is also necessary to combine the patient's personal bad habit value, that is, combine the deviation of the personal bad habit value of the patient to be examined relative to all historical patients, and correct the initial local hemoptysis curve obtained above to obtain the final accurate local hemoptysis curve.

[0098] In the embodiment of the present invention, according to the deviation of the personal bad habit values ​​of the patient to be examined from all the historical patients, the initial local hemoptysis curve is corrected to obtain a corrected local hemoptysis curve, and the implementation steps include:

[0099] According to the overall distribution level and dispersion of the individual bad habit values ​​of all historical patients, the mean value of bad habits and the dispersion index of bad habit values ​​are determined;

[0100] Determine the hemoptysis index correction parameter of the patient to be examined according to the difference between the individual bad habit value of the patient to be examined and the mean value of the bad habit value, and the size of the bad habit value discrete index;

[0101] According to the hemoptysis index correction parameter, the hemoptysis index corresponding to each disease time in the initial local hemoptysis curve is updated to obtain a corrected local hemoptysis curve.

[0102] For the above steps, based on the medical records of all historical patients corresponding to the age group of the patient to be examined in the database, the personal bad habit values ​​of all historical patients are obtained, and the overall difference of these personal bad habit values ​​is determined. If the overall difference is greater, it means that the difference in bad habits of historical patients in the database is greater, there are patients with good habits, and there are patients with poor habits, that is, the overall bad habits are unevenly distributed, and the lower the personal bad habit value of the patient to be examined is than the overall distribution of the personal bad habit values ​​of all historical patients in the database, the smaller the correction amplitude of the hemoptysis index of the patient to be examined will be.

[0103] In the embodiment of the present invention, the calculation formula for determining the overall difference is:

[0104] ;

[0105] in, It indicates the overall difference of individual bad habit values ​​of all historical patients obtained based on the medical records of all historical patients corresponding to the age group of the patient to be examined in the database, that is, the bad habit value dispersion index; Indicates j The personal bad habit value of each patient; It represents the mean bad habit value corresponding to the personal bad habit value of all historical patients; Represents the total number of historical patients.

[0106] Then, according to the difference between the personal bad habit value of the patient to be examined and the personal bad habit values ​​of all historical patients, and combined with the overall difference of the personal bad habit values ​​of all historical patients, the correction parameter of the hemoptysis index of the patient to be examined is determined, and the corresponding calculation formula is:

[0107] ;

[0108] in, Indicates patients to be examined Correction parameters for hemoptysis index, Indicates patients to be examined Personal bad habits value; It represents the mean bad habit value corresponding to the personal bad habit value of all historical patients; It represents the overall difference of individual bad habit values ​​of all historical patients, that is, the dispersion index of bad habit values; Indicated by natural constant The exponential function of base .

[0109] According to the hemoptysis index correction parameter of the patient to be examined obtained above, the hemoptysis index corresponding to each illness time in the initial local hemoptysis curve of the patient to be examined is updated, that is, the sum of the hemoptysis index correction parameter of the patient to be examined obtained above and the value 1 is multiplied by the hemoptysis index corresponding to each illness time in the initial local hemoptysis curve, and the hemoptysis index corresponding to each illness time in the initial local hemoptysis curve is updated with the obtained multiplication value, so as to obtain a corrected local hemoptysis curve.

[0110] Through the above method, the local hemoptysis curve of the patient to be examined affected by personal bad habits can be obtained, that is, the corrected local hemoptysis curve.

[0111] Step S500: Determine the hemoptysis risk index of the patient to be examined according to the similarity between the contrast segments in the modified local hemoptysis curve and the standard hemoptysis deterioration curve, and the difference in overall distribution levels between the contrast segments in the modified local hemoptysis curve and the standard hemoptysis deterioration curve.

[0112] By comparing the similarity between the local hemoptysis curve of the patient to be examined affected by personal bad habits and the comparison segment in the standard hemoptysis deterioration curve, when the similarity is higher, it means that the patient to be examined is more in line with the standard disease deterioration law under the condition of no intervention; at the same time, by comparing the overall distribution level difference between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve, when the overall distribution level of the modified local hemoptysis curve is higher than the overall distribution level of the comparison segment in the standard hemoptysis deterioration curve, it means that the degree of deterioration of the patient to be tested is higher, and the corresponding hemoptysis risk is greater. Therefore, the hemoptysis risk index of the patient to be examined can be determined according to the similarity between the comparison segment in the modified local hemoptysis curve and the standard hemoptysis deterioration curve, and the overall distribution level difference between the comparison segment in the modified local hemoptysis curve and the standard hemoptysis deterioration curve.

[0113] In an embodiment of the present invention, the hemoptysis risk index of the patient to be examined is determined, and the implementation steps include:

[0114] Determine the DTW distance between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve, and perform negative correlation mapping on the DTW distance to obtain a similarity index value;

[0115] Determine the average values ​​of the hemoptysis indexes corresponding to all the disease time in the comparison segments of the modified local hemoptysis curve and the standard hemoptysis deterioration curve, respectively, and obtain the first hemoptysis index average value and the second hemoptysis index average value;

[0116] Performing mapping processing on the difference between the first hemoptysis index mean and the second hemoptysis index mean to obtain a difference index value;

[0117] The hemoptysis risk index of the patient to be examined is determined according to the similarity index value and the difference index value, and both the similarity index value and the difference index value are positively correlated with the hemoptysis risk index.

[0118] For the above steps, when determining the similarity between the local hemoptysis curve of the patient to be examined affected by personal bad habits and the comparison segment in the standard hemoptysis deterioration curve, since the curve lengths of the local hemoptysis curve and the comparison segment are inconsistent, the dynamic time warping (DTW) method is used to calculate the DTW distance between the local hemoptysis curve and the comparison segment, and the DTW distance is negatively correlated to obtain a similarity index value. At the same time, the average value of the hemoptysis index corresponding to all the sick time in the corrected local hemoptysis curve is determined to obtain the first hemoptysis index mean, and the average value of the hemoptysis index corresponding to all the sick time in the comparison segment in the standard hemoptysis deterioration curve is determined to obtain the second hemoptysis index mean. The higher the degree to which the first hemoptysis index mean is greater than the second hemoptysis index mean, the higher the possibility of hemoptysis. Therefore, the hemoptysis risk index of the patient to be examined can be determined by combining the similarity index value and the difference between the first hemoptysis index mean and the second hemoptysis index mean.

[0119] In the embodiment of the present invention, the hemoptysis risk index of the patient to be examined is determined, and the corresponding calculation formula is:

[0120] ;

[0121] in, Indicates patients to be examined Risk indicators for hemoptysis; Indicates patients to be examined Corrected local hemoptysis curve Comparison with the standard hemoptysis exacerbation curve DTW distance; Indicates patients to be examined The mean value of the first hemoptysis index corresponding to the corrected local hemoptysis curve; It represents the mean value of the second hemoptysis index corresponding to the comparison segment in the standard hemoptysis deterioration curve; Indicated by natural constant The exponential function with base ; Indicates the denominator correction value, which is used to prevent the denominator from being taken as 0. In this embodiment of the present invention, ; Represents the normalization function.

[0122] In the above calculation formula, the similarity index value The larger the value of and the greater the difference between the mean of the first hemoptysis index and the mean of the second hemoptysis index, the corresponding difference index value The larger the value of , the larger the corresponding hemoptysis risk index value, indicating that the patient to be examined The risk of hemoptysis is higher.

[0123] Step S600: Determine whether to issue a hemoptysis warning based on the hemoptysis risk indicator.

[0124] A hemoptysis risk threshold is set in advance, and it is determined whether the hemoptysis risk index determined above is greater than the hemoptysis risk threshold; if the hemoptysis risk index is greater than the hemoptysis risk threshold, it indicates that the patient to be examined is very likely to have a risk of hemoptysis, and a hemoptysis warning is performed, otherwise no hemoptysis warning is performed, so as to assist the doctor in the next step of various examinations, diagnosis and treatment. In the embodiment of the present invention, the value of the hemoptysis risk threshold is set to 0.5.

[0125] Based on the same inventive concept, the embodiment of the present invention also provides an intelligent medical record analysis device for tuberculosis patients for early warning of hemoptysis, such as Figure 2 As shown, the device comprises:

[0126] The historical data acquisition module is used to: extract the medical records of historical patients in the database to obtain the standard hemoptysis deterioration curve and the personal bad habit value of each historical patient;

[0127] The real-time data acquisition module is used to extract the medical records of the patient to be examined, and obtain the initial local hemoptysis curve and the personal bad habit value of the patient to be examined;

[0128] A comparison segment acquisition module is used to: determine the comparison segment in the standard hemoptysis deterioration curve according to the change fluctuation difference between the standard hemoptysis deterioration curve and the initial local hemoptysis curve;

[0129] A local hemoptysis curve acquisition module is used to: correct the initial local hemoptysis curve in combination with the deviation of the individual bad habit values ​​of the patient to be examined relative to all historical patients, to obtain a corrected local hemoptysis curve;

[0130] A risk index acquisition module, used to determine the hemoptysis risk index of the patient to be examined according to the similarity between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve, and the overall distribution level difference between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis deterioration curve;

[0131] The judgment and early warning module is used to: judge whether to issue a hemoptysis early warning according to the hemoptysis risk indicator.

[0132] It should be noted that the device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the above embodiment provides an intelligent analysis method for the medical records of tuberculosis patients for hemoptysis warning, which belongs to the same concept. The specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0133] Based on the same inventive concept, the embodiment of the present invention also provides an intelligent analysis system for medical records of pulmonary tuberculosis patients for early warning of hemoptysis, such as Figure 3 As shown, the control system includes: a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the system can execute any one of the intelligent analysis methods for medical records of tuberculosis patients for hemoptysis warning introduced above.

[0134] The embodiment of the present invention can divide the system into functional modules according to the above method example. For example, each functional module can be corresponded to, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0135] Based on the same inventive concept, an embodiment of the present invention also provides a computer program product, which includes: computer program code, when the computer program code runs on a computer, the computer executes any one of the intelligent analysis methods of medical records of tuberculosis patients for hemoptysis warning introduced above.

[0136] Based on the same inventive concept, an embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program code. When the computer program code runs on a computer, the computer executes any one of the intelligent analysis methods for medical records of tuberculosis patients for hemoptysis warning described above.

[0137] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for intelligent analysis of medical records of pulmonary tuberculosis patients for early warning of hemoptysis, characterized in that: The following steps are involved: Data were extracted from the medical records of historical patients in the database to obtain the standard hemoptysis exacerbation curve and the individual bad habit value of each historical patient; Data extraction was performed on the medical records of the patients to be examined to obtain the initial local hemoptysis curve and the personal bad habit values ​​of the patients to be examined; Determining a comparison segment in the standard hemoptysis deterioration curve according to a difference in change fluctuations between the standard hemoptysis deterioration curve and the initial local hemoptysis curve; According to the deviation of the individual bad habit values ​​of the patient to be examined relative to all historical patients, the initial local hemoptysis curve is corrected to obtain a corrected local hemoptysis curve; Determining the hemoptysis risk index of the patient to be examined according to the similarity between the contrast segment in the modified local hemoptysis curve and the standard hemoptysis deterioration curve, and the difference in overall distribution level between the contrast segment in the modified local hemoptysis curve and the standard hemoptysis deterioration curve; Determining whether to perform hemoptysis warning according to the hemoptysis risk indicator; The data extraction of the medical records of historical patients in the database is performed to obtain the standard hemoptysis deterioration curve and the personal bad habit value of each historical patient, including: Extract the medical records of historical patients in the database to obtain the personal bad habit values, hemoptysis indicators and the illness time corresponding to the hemoptysis indicators of historical patients; According to the hemoptysis indexes of all historical patients and their corresponding illness time, curve fitting is performed to obtain a standard hemoptysis deterioration curve of the hemoptysis indexes changing with the illness time; Determining the comparison segment in the standard hemoptysis exacerbation curve comprises: Determining the hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined, and determining the hemoptysis index change rate of the standard hemoptysis deterioration curve; Determine the end time point of the comparison segment in the standard hemoptysis deterioration curve according to the difference between the hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined and the hemoptysis index change rate of the standard hemoptysis deterioration curve, as well as the total length of illness time and the end illness time of the initial local hemoptysis curve of the patient to be examined; Taking the starting time of the initial local hemoptysis curve of the patient to be examined as the starting time point of the comparison segment in the standard hemoptysis deterioration curve, and determining the comparison segment in the standard hemoptysis deterioration curve according to the starting time point and the ending time point of the comparison segment in the standard hemoptysis deterioration curve; Determining the end time point of the comparison segment in the standard hemoptysis curve includes: Determining a time coefficient according to a difference between a hemoptysis index change rate of the initial local hemoptysis curve of the patient to be examined and a hemoptysis index change rate of the standard hemoptysis deterioration curve; Calculating the product of the total length of the initial local hemoptysis curve of the patient to be examined and the time coefficient to obtain an adjustment value of the sick time; The sum of the end time of the initial local hemoptysis curve of the patient to be examined and the adjustment value of the disease time is calculated to obtain the end time point of the comparison segment in the standard hemoptysis curve. The corresponding calculation formula is: ； in, Indicates patients to be examined The initial local hemoptysis curve is at the end time point of the comparison segment in the standard hemoptysis exacerbation curve; Indicates patients to be examined The duration of illness corresponding to the last hemoptysis index of the initial local hemoptysis curve; Indicates patients to be examined The total length of illness from the initial local hemoptysis curve; Indicates patients to be examined The rate of change of hemoptysis index of the initial local hemoptysis curve; It represents the change rate of hemoptysis index of the standard hemoptysis deterioration curve; represents the sickness time adjustment value, represents the time coefficient, represents the inverse tangent function; The initial local hemoptysis curve is corrected in combination with the deviation of the individual bad habit values ​​of the patient to be examined relative to all historical patients to obtain a corrected local hemoptysis curve, including: According to the overall distribution level and dispersion of the individual bad habit values ​​of all historical patients, the mean value of bad habits and the dispersion index of bad habit values ​​are determined; According to the difference between the individual bad habit value of the patient to be examined and the mean of the bad habit value, and the size of the discrete index of the bad habit value, the hemoptysis index correction parameter of the patient to be examined is determined, and the corresponding calculation formula is: ； in, Indicates patients to be examined Correction parameters for hemoptysis index, Indicates patients to be examined Personal bad habits value; It represents the mean bad habit value corresponding to the personal bad habit value of all historical patients; The bad habit value discrete index representing the individual bad habit values ​​of all historical patients; Expressed as a natural constant The exponential function with base ; ； in, The bad habit value discrete index represents the individual bad habit values ​​of all historical patients obtained based on the medical records of all historical patients corresponding to the age group to which the patient to be examined belongs in the database; Indicates j The personal bad habit value of each patient; It represents the mean bad habit value corresponding to the personal bad habit value of all historical patients; represents the total number of historical patients; According to the hemoptysis index correction parameter, the hemoptysis index corresponding to each disease time in the initial local hemoptysis curve is updated to obtain a corrected local hemoptysis curve; Determine the DTW distance between the modified local hemoptysis curve and the comparison segment in the standard hemoptysis curve, and perform negative correlation mapping on the DTW distance to obtain a similarity index value; Determine the average values ​​of the hemoptysis indexes corresponding to all the disease time in the comparison segments of the modified local hemoptysis curve and the standard hemoptysis curve, respectively, and obtain the first hemoptysis index average value and the second hemoptysis index average value; Performing mapping processing on the difference between the first hemoptysis index mean and the second hemoptysis index mean to obtain a difference index value; The hemoptysis risk index of the patient to be examined is determined according to the similarity index value and the difference index value, wherein both the similarity index value and the difference index value are positively correlated with the hemoptysis risk index, and the corresponding calculation formula is: ; in, Indicates patients to be examined Risk indicators for hemoptysis; Indicates patients to be examined Corrected local hemoptysis curve Comparison with the standard hemoptysis exacerbation curve DTW distance; Indicates patients to be examined The mean value of the first hemoptysis index corresponding to the corrected local hemoptysis curve; It represents the mean value of the second hemoptysis index corresponding to the comparison segment in the standard hemoptysis deterioration curve; Expressed as a natural constant The exponential function with base ; represents the denominator correction value, Represents the normalization function.

2. According to claim 1, a method for intelligent analysis of medical records of pulmonary tuberculosis patients for early warning of hemoptysis, characterized in that: Obtaining the hemoptysis index includes: Extracting hemoptysis information from medical records, and determining various hemoptysis vectors according to the hemoptysis information, wherein the hemoptysis vectors are composed of at least two different types of hemoptysis characteristic values; Perform one-dimensional transformation on the hemoptysis vector to obtain a hemoptysis index.

3. The intelligent analysis method of medical records of pulmonary tuberculosis patients for early warning of hemoptysis according to claim 2 is characterized in that: The different types of hemoptysis characteristic values ​​at least include a hemoptysis amount characteristic value, a hemoptysis color depth characteristic value and a hemoptysis frequency characteristic value.

4. The intelligent analysis method of medical records of pulmonary tuberculosis patients for early warning of hemoptysis according to claim 1, characterized in that: Determining the personal bad habit value includes: Extract personal habit information from medical records, input the personal habit information into a semantic recognition network, and output a corresponding personal bad habit value from the semantic recognition network, where the personal bad habit value ranges from 0 to 1.

5. The intelligent analysis method of medical records of pulmonary tuberculosis patients for early warning of hemoptysis according to claim 1 is characterized in that: The determining whether to perform hemoptysis warning according to the hemoptysis risk indicator includes: Determining whether the hemoptysis risk index is greater than the hemoptysis risk threshold; If the hemoptysis risk index is greater than the hemoptysis risk threshold, a hemoptysis warning is performed.

Images