Multi-dimensional quantitative filing method and system

Abstract

The invention relates to the technical field of medical data processing, in particular to a multi-dimensional quantitative filing method and system. The method comprises the following steps: acquiring historical medical data of a patient, and screening out unarchived medical data from the historical medical data; extracting medical data related to diagnosis and treatment of venous thromboembolism from the unarchived medical data by using keywords in a keyword library; calculating corresponding quality control data according to the extracted quality control calculation parameters; storing the extracted basic data in a database according to a time dimension and a space dimension; and storing the extracted quality control calculation parameters and the corresponding quality control data in a database according to a time dimension and a space dimension. By adopting the method, the query efficiency of the medical data and the quality control data can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of medical data processing. More specifically, the present invention relates to a multi-dimensional quantization filing method and system. Background Art

[0002] Venous thromboembolism (VTE) refers to the abnormal coagulation of blood in veins, which completely or incompletely blocks blood vessels and belongs to the disease of venous return disorder. It includes deep vein thrombosis (DVT) and pulmonary thromboembolism (PE).

[0003] VTE is two important clinical manifestation forms of the same disease at different stages and different parts. Epidemiological studies have shown that VTE is a global disease, the third most common cardiovascular disease after ischemic heart disease and stroke, and has become a disease that seriously threatens human health. The occurrence of fatal pulmonary embolism in hospitals has constituted a potential risk to medical quality and safety and has become a severe problem faced by clinical medical staff and hospital managers. Patients in many clinical departments have a risk of VTE; its onset is latent, clinical symptoms are atypical, it is easy to be misdiagnosed and missed diagnosed, and once it occurs, the fatality and disability rates are high.

[0004] During the hospitalization of patients for the treatment of venous thromboembolism, doctors need to regularly query the medical data and quality control parameters related to venous thromboembolism of patients to determine the physical rehabilitation of patients. In the prior art, when querying the medical data corresponding to venous thromboembolism, it is necessary to screen all the medical data in the database and extract keywords to obtain the medical data corresponding to venous thromboembolism. When the amount of data in the database is too large, the query speed will decrease; in addition, when querying the quality control parameters corresponding to venous thromboembolism, it is necessary to first screen out the medical data related to the diagnosis and treatment of venous thromboembolism from all the medical data, and then calculate the corresponding quality control parameters in real time according to the quality control calculation parameters therein combined with the calculation formula, resulting in a low query efficiency of the quality control parameters. Summary of the Invention

[0005] To solve the technical problem of the low query efficiency of the quality control parameters and medical data corresponding to venous thromboembolism in the prior art, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a multi-dimensional quantization filing method, including:

[0007] Obtain the historical medical data of a patient and screen out the unfiled medical data from it;

[0008] Extract medical data related to the diagnosis and treatment of venous thromboembolism from the unarchived medical data using the keywords in the keyword library; the extracted medical data includes basic data and quality control calculation parameters;

[0009] Calculate the corresponding quality control data based on the extracted quality control calculation parameters;

[0010] Store the extracted basic data in the database according to the time dimension and space dimension; and store the extracted quality control calculation parameters and the corresponding quality control data in the database according to the time dimension and space dimension.

[0011] Preferably, the ID of each medical data in the historical medical data is an integer ID; the screening of unarchived medical data includes:

[0012] Obtain the maximum ID value of the most recently archived medical data;

[0013] Screen out the medical data with ID values greater than the maximum ID value from the medical data, and record it as unarchived medical data.

[0014] Preferably, the basic data includes: department / medical record data, patient information data, and examination and test data, where the patient information data includes: patient name, gender, birthday, admission time, discharge time, department, and bed number.

[0015] Preferably, the quality control calculation parameters include: doctor's order content, and the corresponding quality control data for the doctor's order content are mechanical prophylaxis quality indicators, drug prophylaxis treatment indicators, and combined prophylaxis quality indicators.

[0016] Preferably, the quality control calculation parameters further include: correlation reporting data, and the corresponding quality control data for the correlation reporting data are: outcome situation, and the outcome situation includes: hospital-related DVT incidence rate, hospital-related PE incidence rate, hospital-related DVT+PE incidence rate, VTE-related DVT mortality rate, VTE-related PE mortality rate, hospital-related DVT+PE mortality rate, and VTE inpatient incidence rate; the correlation reporting data is used to characterize whether the patient is VTE-confirmed and the confirmation means adopted.

[0017] Preferably, the quality control calculation parameters further include: surgical record data, and the quality control data for the surgical record data include: preoperative risk assessment rate and postoperative risk assessment rate.

[0018] Preferably, the way to store the quality control data is to store the numerator and denominator of the quality control data in a key-value manner respectively, where the content of the key includes the name of the quality control data, the time and space positions corresponding to the quality control data, and the value is composed of the numerator or denominator of the corresponding quality control data.

[0019] Preferably, the time dimension includes days, weeks, months, quarters, and years; the space dimension includes departments and hospital areas.

[0020] Preferably, the extraction of medical data related to the diagnosis and treatment of venous thromboembolism from the unarchived medical data by using the keywords in the keyword library includes:

[0021] Input the unarchived medical data into the trained conditional random field model to obtain the corresponding labeled sequence;

[0022] According to the labeled sequence, segment the text into a word sequence;

[0023] Match the word sequence with the keywords in the keyword library and extract the sentences where the matched word sequences are located.

[0024] In a second aspect, the present invention provides a multi-dimensional quantization archiving system, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the multi-dimensional quantization archiving method of the present invention is implemented.

[0025] The beneficial effects of the present invention are as follows: By extracting medical data related to the diagnosis and treatment of venous thromboembolism from the historical medical data of patients and storing and archiving it according to the time dimension and space dimension, it helps to improve the query speed of medical data; Secondly, since the corresponding quality control data has been calculated when archiving medical data, there is no need to calculate quality control parameters in real time using the queried medical data when querying quality control data, and the quality control parameters can be directly obtained from the archived data, thus improving the efficiency of obtaining quality control data; In addition, when archiving the historical medical data of patients, first screen out the unarchived medical data from all medical data according to the ID of the medical data, thus avoiding duplicate archiving and helping to improve the efficiency of medical data archiving. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown in an exemplary rather than restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0027] Figure 1 is a schematic flowchart of a multi-dimensional quantization archiving method according to an embodiment of the present invention;

[0028] Figure 2 is a schematic diagram showing the storage method of medical data and quality control parameters according to an embodiment of the present invention;

[0029] Figure 3 is a schematic diagram showing the derived archived data according to an embodiment of the present invention;

[0030] Figure 4 is a schematic diagram showing the structure of a multi-dimensional quantization archiving system according to an embodiment of the present invention. Detailed implementation manners

[0031] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0032] Next, the specific implementation manners of the present invention will be described in detail with reference to the accompanying drawings.

[0033] Embodiment of the multi-dimensional quantization archiving method:

[0034] As Figure 1 shown, the multi-dimensional quantization archiving method of the present invention includes:

[0035] S101. Obtain the historical medical data of the patient and screen out the unarchived medical data therefrom;

[0036] S102. Extract the unarchived medical data, specifically: use the keywords in the keyword library to extract the medical data related to the diagnosis and treatment of venous thromboembolism from the unarchived medical data; the extracted medical data includes basic data and quality control calculation parameters;

[0037] The keywords in the keyword library include keywords related to the diagnosis and treatment of venous thromboembolism. For example, the diagnosis-related keywords include: venous thromboembolism, symptoms, D-dimer test, imaging examination, ultrasonic examination, CT scan. The treatment-related keywords include: anticoagulant therapy, thrombolytic therapy, surgical treatment, catheter thrombolytic therapy, etc. The basic data refers to the data only used to understand the diagnosis information of the patient.

[0038] S103. Calculate the corresponding quality control data according to the extracted quality control calculation parameters;

[0039] S104. Archive the basic data, quality control calculation parameters and the corresponding quality control data, specifically: store the extracted basic data in the database according to the time dimension and space dimension; and store the extracted quality control calculation parameters and the corresponding quality control data in the database according to the time dimension and space dimension.

[0040] The storage method for the extracted medical data and the calculated quality control parameters is as Figure 2As shown, calculations are performed based on the time dimension and the space dimension.

[0041] By storing the extracted medical data according to the time dimension and the space dimension, it is convenient for users to quickly query and search for the corresponding medical data.

[0042] In the prior art, when querying the medical data corresponding to venous thromboembolism, it is necessary to screen and extract keywords from all the medical data in the database to obtain the medical data corresponding to venous thromboembolism. When the amount of data in the database is too large, the query speed will decrease; when querying the quality control parameters corresponding to venous thromboembolism, it is necessary to first screen out the medical data related to the diagnosis and treatment of venous thromboembolism from all the medical data, and then calculate the corresponding quality control parameters in real time according to the quality control calculation parameters therein and the calculation formula, resulting in a low query efficiency of the quality control parameters. The multi-dimensional quantification and archiving method of the present invention extracts the medical data related to the diagnosis and treatment of venous thromboembolism from the historical medical data of patients and stores and archives it according to the time dimension and the space dimension, thereby helping to improve the query speed of medical data; secondly, since the corresponding quality control data has been calculated when archiving the medical data, there is no need to calculate the quality control parameters in real time using the queried medical data when querying the quality control data, and the quality control parameters can be directly obtained from the archived data, thereby improving the efficiency of obtaining the quality control data; in addition, when archiving the historical medical data of patients, first screen out the unarchived medical data from all the medical data according to the ID of the medical data, thereby avoiding duplicate archiving and helping to improve the efficiency of archiving medical data.

[0043] In this embodiment, the time dimension includes days, weeks, months, quarters, and years; the space dimension includes departments and hospital areas.

[0044] By setting the time dimension to include days, weeks, months, quarters, and years, it is convenient to query the medical data of a specific day. By setting the space dimension to departments and hospital areas, it is convenient to query the medical data corresponding to a certain hospital area or a certain department. In this embodiment, the departments include: operating department, oncology department 1, intensive care medicine department, general internal medicine department, orthopedics and bone disease department, general surgery department, oncology department 3, neurology department, etc.

[0045] In one embodiment, the ID of each piece of medical data in the historical medical data is an integer ID; the screening out of the unarchived medical data includes:

[0046] S401. Obtain the maximum ID value of the most recently archived medical data;

[0047] S402. Screen out the medical data with ID values greater than the maximum ID value from the medical data, and record it as unarchived medical data. Since the ID of medical data is an integer ID, integer IDs are usually set with the Auto Increment attribute. In this way, whenever a new record is inserted, a new integer that is 1 greater than the current maximum ID will be automatically generated as the ID of the new record. Therefore, for the newly generated medical data, the value of its corresponding ID will be greater than the IDs of the previously generated medical data. When archiving the historical medical data of a patient, first screen out the unarchived medical data from all the medical data based on the ID of the medical data, so as to avoid duplicate archiving and help improve the efficiency of medical data archiving.

[0048] In one embodiment, the basic data includes: department / medical record data, patient information data, and examination and test data, where the patient information data includes: patient name, gender, birthday, admission time, discharge time, department, and bed number.

[0049] In this embodiment, the keywords corresponding to the department / medical record data include: department name, department ID. The keywords corresponding to the patient information data include: name, gender, birthday, admission time, discharge time, inpatient number, patient ID, department ID. The keywords corresponding to the examination and test data include: examination type, abnormal value range, examination result, patient ID.

[0050] In one embodiment, the quality control calculation parameters include: doctor's order content, and the quality control data corresponding to the doctor's order content are mechanical prevention quality indicators, drug prevention and treatment indicators, and combined prevention quality indicators.

[0051] The keywords corresponding to the doctor's order content include: doctor's order content, doctor's order time, patient ID.

[0052] In this embodiment, the mechanical prevention quality indicators include: total mechanical prevention implementation rate, mechanical prevention implementation rate within 24 hours after admission, mechanical prevention implementation rate within 24 hours before surgery, mechanical prevention implementation rate during surgery, mechanical prevention implementation rate within 24 hours after surgery, mechanical prevention implementation rate within 24 hours after transfer to another department. The drug prevention and treatment indicators include: total drug prevention implementation rate, drug prevention implementation rate within 24 hours after admission, drug prevention implementation rate within 24 hours to 72 hours before surgery, drug prevention implementation rate within 24 hours after surgery, drug prevention implementation rate within 24 hours to 72 hours after surgery, mechanical prevention implementation rate within 24 hours after transfer to another department, proportion of anticoagulant drugs in discharge instructions. The combined prevention quality indicators include: total combined prevention implementation rate, combined prevention implementation rate within 24 hours after admission, combined prevention implementation rate within 24 hours to 72 hours before surgery, combined prevention implementation rate within 24 hours after surgery, combined prevention implementation rate within 24 hours to 72 hours after surgery, combined prevention implementation rate within 24 hours after transfer to another department.

[0053] In one embodiment, the quality control calculation parameters further include: correlation reporting data, and the quality control data corresponding to the correlation reporting data is: outcome situation, and the outcome situation includes: hospital-related DVT incidence rate, hospital-related PE incidence rate, hospital-related DVT+PE incidence rate, VTE-related DVT mortality rate, VTE-related PE mortality rate, hospital-related DVT+PE mortality rate, and VTE inpatient incidence rate; the correlation reporting data is used to characterize whether the patient has a VTE diagnosis and the diagnosis means adopted.

[0054] The keywords corresponding to the correlation reporting data include: patient ID, reporting time, and reporting type.

[0055] In one embodiment, the quality control calculation parameters further include: surgical record data, and the quality control data of the surgical record data includes: preoperative risk assessment rate and postoperative risk assessment rate.

[0056] The keywords corresponding to the surgical record data include: surgical start time, surgical end time, and patient ID.

[0057] In one embodiment, the quality control calculation parameters further include: transfer record data, and the quality control data of the transfer record data includes: risk assessment rate after transfer.

[0058] The keywords corresponding to the transfer record data include: transfer time, department before transfer, department after transfer, and patient ID.

[0059] In one embodiment, the method of storing the quality control data is to store the numerator and denominator of the quality control data in a key-value manner respectively, where the content of the key includes the name of the quality control data, the time and spatial location corresponding to the quality control data, and the value is composed of the numerator or denominator of the corresponding quality control data.

[0060] In this embodiment, if the name of the quality control data is the total risk assessment coefficient, the value of the quality control data is 540 / 553, the time corresponding to the quality control data is July 18, 2022, and the corresponding spatial location is the whole hospital, then the content stored in the corresponding key-value manner is:

[0061] Key = Total risk assessment numerator number 2022-07-18 Whole hospital, Value = 540;

[0062] Key = Total risk assessment denominator number 2022-07-18 Whole hospital, Value = 553.

[0063] If the name of the quality control data is the total risk assessment coefficient, the value of the quality control data is 325 / 334, the time corresponding to the quality control data is July 18, 2022, and the corresponding spatial location is the Department of Neurointervention, then the content stored in the corresponding key-value format is:

[0064] Key = Total risk assessment numerator 2022-07-18 Department of Neurointervention, Value = 325;

[0065] Key = Total risk assessment denominator 2022-07-18 Department of Neurointervention, Value = 334.

[0066] If the name of the quality control data is the total risk assessment coefficient, the value of the quality control data is 215 / 219, the time corresponding to the quality control data is July 18, 2022, and the corresponding spatial location is the Department of Neurosurgery, then the content stored in the corresponding key-value format is:

[0067] Key = Total risk assessment numerator 2022-07-18 Department of Neurosurgery, Value = 215;

[0068] Key = Total risk assessment denominator 2022-07-18 Department of Neurosurgery, Value = 219.

[0069] In one embodiment, the extracting of medical data related to the diagnosis and treatment of venous thromboembolism from the unarchived medical data by using the keywords in the keyword library includes:

[0070] S201. Input the unarchived medical data into the trained conditional random field model to obtain the corresponding labeled sequence;

[0071] S202. According to the labeled sequence, segment the text into a word sequence;

[0072] S203. Match the word sequence with the keywords in the keyword library and extract the sentence where the matched word sequence is located.

[0073] In this embodiment, by using the conditional random field model to segment the unarchived medical data, the accuracy of the segmentation result can be greatly improved, and thus the accuracy of the extracted medical data can be improved.

[0074] In this embodiment, the training process of the conditional random field model includes:

[0075] S301. Collect the text data of the medical data to be segmented and perform preprocessing, where the preprocessing includes removing irrelevant characters and standardization;

[0076] To ensure the training effect of the model, as much text data of medical data to be segmented as possible should be collected.

[0077] S302. Annotate the preprocessed text data to form a training set;

[0078] S303. Use the text data in the training set to train a conditional random field model.

[0079] The training process usually includes selecting appropriate feature functions, setting model parameters (such as the weights of feature functions), and maximizing the conditional probability P(Y|X) through iterative optimization algorithms (such as gradient descent, quasi-Newton method, etc.).

[0080] As Figure 3 shown, in order to further solidify the archived data, in one embodiment, the archived data can be exported and stored in the form of a table.

[0081] Embodiment of a multi-dimensional quantization archiving system:

[0082] The present invention also provides a multi-dimensional quantization archiving system. As Figure 4 shown, the multi-dimensional quantization archiving system includes a processor and a memory, and the memory stores computer program instructions. When the computer program instructions are executed by the processor, a multi-dimensional quantization archiving method described in the above embodiments is implemented.

[0083] The multi-dimensional quantization archiving system further includes other components well-known to those skilled in the art, such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be described in detail here.

[0084] In the present invention, the aforementioned memory may be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as, resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application program, module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that can be stored or otherwise held by such a computer-readable medium.

[0085] In the description of this specification, "a plurality of" and "several" mean at least two, such as two, three, or more, etc., unless otherwise specifically defined.

[0086] Although this specification has shown and described multiple embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, alterations, and alternative ways without departing from the spirit and scope of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in the practice of the present invention.

Claims

1. A multi-dimensional quantitative archiving method, characterized in that: include: Obtain the patient's historical medical data and filter out the unarchived medical data; Extracting medical data related to the diagnosis and treatment of venous thromboembolism from the unarchived medical data using the keywords in the keyword library; the extracted medical data includes basic data and quality control calculation parameters; Calculate the corresponding quality control data according to the extracted quality control calculation parameters; The extracted basic data are stored in the database according to the time dimension and the space dimension; and the extracted quality control calculation parameters and the corresponding quality control data are stored in the database according to the time dimension and the space dimension.

2. The multi-dimensional quantitative archiving method according to claim 1, characterized in that: The ID of each piece of medical data in the historical medical data is an integer ID; The screening of unarchived medical data includes: Get the maximum ID value of the most recently archived medical data; Medical data with an ID value greater than the maximum ID value is screened out from the medical data and recorded as unarchived medical data.

3. The multi-dimensional quantitative archiving method according to claim 1, characterized in that: The basic data includes: department / medical record data, patient information data and examination and testing data, wherein the patient information data includes: patient name, gender, birthday, admission time, discharge time, department and bed number.

4. The multi-dimensional quantitative archiving method according to claim 1, characterized in that: The quality control calculation parameters include: the content of the doctor's order, and the quality control data corresponding to the content of the doctor's order are the quality index of mechanical prevention, the index of drug prevention and treatment, and the quality index of combined prevention.

5. The multi-dimensional quantitative archiving method according to claim 4, characterized in that: The quality control calculation parameters also include: correlation reporting data, the quality control data corresponding to the correlation reporting data are: outcome conditions, the outcome conditions include: hospital-related DVT incidence, hospital-related PE incidence, hospital-related DVT+PE incidence, VTE-related DVT mortality, VTE-related PE mortality, hospital-related DVT+PE mortality and VTE inpatient incidence; the correlation reporting data is used to characterize whether the patient is diagnosed with VTE and the diagnosis measures taken.

6. The multi-dimensional quantitative archiving method according to claim 4, characterized in that: The quality control calculation parameters also include: surgery record data, and the quality control data of the surgery record data includes: preoperative risk assessment rate and postoperative risk assessment rate.

7. The multi-dimensional quantitative archiving method according to any one of claims 1 to 6, characterized in that: The way to store quality control data is to store the numerator and denominator of the quality control data respectively in key-value mode, where the key content includes the name of the quality control data, the time and space position corresponding to the quality control data, and the value is composed of the numerator or denominator of the corresponding quality control data.

8. The multi-dimensional quantitative archiving method according to claim 7, characterized in that: The time dimension includes day, week, month, quarter and year; the space dimension includes department and hospital area.

9. The multi-dimensional quantitative archiving method according to claim 7, characterized in that: The extracting of medical data related to the diagnosis and treatment of venous thromboembolism from the unarchived medical data using the keywords in the keyword library includes: Input the unarchived medical data into the trained conditional random field model to obtain the corresponding annotation sequence; According to the tag sequence, the text is segmented into word sequences; The word sequence is matched with the keywords in the keyword library, and the sentence containing the matched word sequence is extracted.

10. A multi-dimensional quantitative archiving system, comprising a processor and a memory, wherein the memory stores computer program instructions, characterized in that: When the computer program instructions are executed by the processor, the multi-dimensional quantitative archiving method according to any one of claims 1 to 9 is implemented.

Images