CN114496196A - Automatic auditing system for clinical biochemical inspection in medical laboratory - Google Patents
Automatic auditing system for clinical biochemical inspection in medical laboratory Download PDFInfo
- Publication number
- CN114496196A CN114496196A CN202210148648.8A CN202210148648A CN114496196A CN 114496196 A CN114496196 A CN 114496196A CN 202210148648 A CN202210148648 A CN 202210148648A CN 114496196 A CN114496196 A CN 114496196A
- Authority
- CN
- China
- Prior art keywords
- module
- audit
- data
- sample
- random forest
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Business, Economics & Management (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses an automatic auditing system for clinical biochemical inspection in a medical laboratory, and relates to the technical field of medical information processing. The system comprises a data acquisition module, a feature extraction module, a feature identification module and an audit result output module which are connected in sequence, wherein the data acquisition module is used for acquiring the data to be audited; the characteristic extraction module is used for extracting characteristics according to the data to be audited and dividing the data into a sample characteristic set and a sample label; the feature recognition module is used for inputting the sample feature set into an audit model, the audit model is composed of three classifiers of a random forest, an XGboost and a GBDT + LR, and the three classifiers respectively recognize the sample feature set to obtain a sub-recognition result; and the auditing result output module is used for carrying out combined decision on the sub-recognition results through the auditing model to obtain a final auditing result. The invention establishes the clinical examination automatic auditing system based on machine learning, reduces the workload of the clinical laboratory staff and improves the examination accuracy.
Description
Technical Field
The invention relates to the technical field of medical information processing, in particular to an automatic auditing system for clinical biochemical inspection in a medical laboratory.
Background
With the improvement of the automation degree of the inspection equipment, the current clinical inspection items are basically detected by large-scale biochemical and immune automatic analyzers, and the high-throughput inspection equipment provides huge challenges for the examination and verification of inspection results while meeting clinical requirements. In the prior art, workers in a clinical testing laboratory generally need to review and confirm a large amount of data in a short time to meet the requirement of sample testing.
The clinical examination report sheet is increasing with the increase of the diagnosis and treatment amount of the hospital, the pressure of manual examination and examination is very high, workers need to judge and examine according to the clinical diagnosis, the reference range of the detection items and the past experience, massive data is time-consuming and labor-consuming, and the judgment level of the workers is uneven, so that the detection results of patients are different. In recent years, medical inspection analysis technology and information technology are rapidly developed, but intelligent verification of inspection results is still in a starting stage, and a clinical medical clinical laboratory information system can only perform abnormal reminding on the inspection results and cannot perform verification decision by combining personalized information. Therefore, it is an urgent problem for those skilled in the art to accurately and rapidly perform automatic review of clinical biochemical test results.
Disclosure of Invention
In view of this, the invention provides an automatic auditing system for clinical biochemical examination in a medical laboratory, which can accurately and quickly perform automatic auditing on clinical biochemical examination results.
In order to achieve the purpose, the invention adopts the following technical scheme: an automatic auditing system for clinical biochemical examination in a medical laboratory comprises a data acquisition module, a feature extraction module, a feature identification module and an auditing result output module which are connected in sequence, wherein,
the data acquisition module is used for acquiring the data to be checked;
the characteristic extraction module is used for extracting characteristics according to the data to be audited and dividing the data into a sample characteristic set and a sample label;
the feature recognition module is used for inputting the sample feature set into an audit model, the audit model is composed of three classifiers of a random forest, XGboost and GBDT + LR, and the random forest, the XGboost and the GBDT + LR respectively recognize the sample feature set to obtain three sub-recognition results;
and the auditing result output module is used for carrying out combined decision on the three sub-identification results through the auditing model to obtain a final auditing result.
Optionally, the system further comprises a data preprocessing module, wherein the data preprocessing module comprises a denoising submodule and a normalization processing submodule, the denoising submodule is connected with the data acquisition module and the normalization processing submodule, and the normalization processing submodule is connected with the feature extraction module.
Optionally, the sample label is 0 or 1; wherein, the "audit passed" is marked as "0", and the "audit failed" is marked as "1".
Optionally, the combined decision of the audit result output module is as follows: if the sample labels of the sub-identification results are all '0', the audit result is '0'; and if at least one sample label of the sub-identification result is '1', the auditing result is '1'.
Optionally, the feature recognition module inputs the sample feature set into each decision tree in a random forest model to obtain a classification result output by each decision tree in the random forest model, and the random forest model is obtained by training using historical audit data as a training sample;
and determining a sub-recognition result according to the classification result output by each decision tree in the random forest model and the weight corresponding to each decision tree in the random forest model.
Optionally, the feature recognition module adds weights of decision trees with the same classification result in the random forest model to determine a classification weight of each classification result; and taking the classification result with the maximum classification weight as a sub-identification result.
According to the technical scheme, compared with the prior art, the invention discloses and provides an automatic auditing system for clinical biochemical examination in a medical laboratory, which has the following beneficial technical effects:
(1) the working efficiency of medical inspection and audit is improved, the return time of inspection results is shortened, and the error rate is reduced;
(2) the automatic clinical examination and verification method based on machine learning is established by fusing historical examination data and a data set of individualized information of an examiner, the workload of clinical examination and verification staff of a clinical laboratory is reduced, clinical examination and verification of a medical laboratory are carried out by the machine learning method, and the accuracy of the examination is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a block diagram of the system of the present invention;
FIG. 2 is a flow chart of the method of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment 1 of the invention discloses an automatic auditing system for clinical examination based on machine learning, which comprises a data acquisition module, a feature extraction module, a feature identification module and an auditing result output module, as shown in figure 1, wherein,
the data acquisition module is used for acquiring the data to be checked;
the characteristic extraction module is used for extracting characteristics according to the data to be audited and dividing the data into a sample characteristic set and a sample label;
the feature recognition module is used for inputting the sample feature set into an audit model, the audit model is composed of three classifiers of a random forest, XGboost and GBDT + LR, and the random forest, the XGboost and the GBDT + LR respectively recognize the sample feature set to obtain three sub-recognition results;
and the audit result output module is used for carrying out combined decision on the three sub-identification results through the audit model to obtain a final audit result.
The data preprocessing module comprises a denoising submodule and a normalization processing submodule, the denoising submodule is connected with the data acquisition module and the normalization processing submodule, and the normalization processing submodule is connected with the feature extraction module.
Further, the sample label is 0 or 1; wherein, the "audit passed" is marked as "0", and the "audit failed" is marked as "1".
The combination decision of the audit result output module is as follows: if the sample labels of the sub-identification results are all '0', the auditing result is '0'; and if at least one sample label of the sub-identification result is 1, the auditing result is 1.
A random forest is a classifier that contains multiple decision trees and whose output classes are dependent on the mode of the class output by the individual trees. The random forest model can not only overcome the problem of decision tree overfitting, but also has good tolerance on noise and abnormal values; the XGboost is an optimized distributed gradient enhancement library and aims to realize high efficiency, flexibility and portability. The method realizes a machine learning algorithm under a Gradient Boosting framework. XGboost provides parallel tree lifting (also called GBDT, GBM), and can quickly and accurately solve a plurality of data science problems; GBDT + LR is a classifier model with the idea of stacking, and the fusion of GBDT and LR means using the path of the decision tree generated by GBDT as the input feature of LR.
Training three classifiers by using historical data to find optimal parameters, and training a random forest model as follows:
the method comprises the following steps: acquiring a random forest model, wherein the random forest model comprises a plurality of decision trees;
step two: acquiring a plurality of training sample data, wherein the training sample data comprises category data and characteristic sample data; the category data is used to mark a pass or fail audit, with "pass audit" marked as "0" and "fail audit" marked as "1"; the characteristic sample data comprises a patient name, detection items, detection data and a detection result;
step three: inputting a plurality of training sample data into a random forest model in sequence to obtain a plurality of prediction data;
step four: calculating the precision of the random forest model according to the plurality of prediction data and the plurality of category data;
step five: and optimizing the random forest model according to the precision to obtain the tuned and optimized random forest model.
Furthermore, in the feature identification module, the step of identifying the sample feature set by the random forest comprises:
s31, inputting the sample feature set into each decision tree in the random forest model to obtain a classification result output by each decision tree in the random forest model, wherein the random forest model is obtained by training with historical audit data as training samples;
s32, determining a sub-recognition result according to the classification result output by each decision tree in the random forest model;
specifically, the sub-recognition result is determined according to the classification result output by each decision tree in the random forest model and the weight corresponding to each decision tree in the random forest model.
S33, determining a sub-recognition result according to the classification result output by each decision tree in the random forest model and the weight corresponding to each decision tree in the random forest model, wherein the sub-recognition result comprises the following steps: adding the weights of the decision trees with the same classification result in the random forest model to determine the classification weight of each classification result; and taking the classification result with the maximum classification weight as an auditing result.
The embodiment 2 of the invention discloses an automatic auditing method for clinical examination based on machine learning, which comprises the following specific steps as shown in figure 2:
s1, acquiring the data to be checked;
s2, extracting features according to the data to be audited, and dividing the data into a sample feature set and a sample label;
s3, inputting the sample feature set into an audit model, wherein the audit model is composed of three classifiers of a random forest, an XGboost and a GBDT + LR, and the random forest, the XGboost and the GBDT + LR respectively identify the sample feature set to obtain three sub-identification results;
and S4, the auditing model makes a combined decision on the three sub-recognition results to obtain a final auditing result.
Further, the method also comprises the step of preprocessing the data to be audited, and the specific steps are as follows:
s11, denoising the data to be audited by a wavelet threshold method to obtain first data;
and S12, carrying out normalization processing on the first data to obtain second data.
Finally, a computer storage medium is provided, having a computer program stored thereon, which, when executed by a processor, performs the steps of a machine learning-based clinical testing automatic audit method.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. An automatic auditing system for clinical biochemical examination in a medical laboratory is characterized by comprising a data acquisition module, a feature extraction module, a feature identification module and an auditing result output module which are sequentially connected, wherein,
the data acquisition module is used for acquiring the data to be checked;
the characteristic extraction module is used for extracting characteristics according to the data to be audited and dividing the data into a sample characteristic set and a sample label;
the feature recognition module is used for inputting the sample feature set into an audit model, the audit model is composed of three classifiers of a random forest, XGboost and GBDT + LR, and the random forest, the XGboost and the GBDT + LR respectively recognize the sample feature set to obtain three sub-recognition results;
and the audit result output module is used for carrying out combined decision on the three sub-identification results through the audit model to obtain a final audit result.
2. The medical laboratory clinical biochemical examination automatic auditing system according to claim 1, further comprising a data preprocessing module including a denoising sub-module connected with the data acquisition module and a normalization processing sub-module connected with the feature extraction module.
3. The medical laboratory clinical biochemical test automatic auditing system according to claim 1, where in the sample label is 0 or 1; wherein, the "audit passed" is marked as "0", and the "audit failed" is marked as "1".
4. The medical laboratory clinical biochemical inspection automatic auditing system according to claim 1, where the combined decision of the audit result output module is: if the sample labels of the sub-identification results are all '0', the audit result is '0'; and if at least one sample label of the sub-identification result is '1', the auditing result is '1'.
5. The medical laboratory clinical biochemical inspection automatic auditing system according to claim 1, characterized in that the feature recognition module inputs the sample feature set into each decision tree in a random forest model, obtains a classification result output by each decision tree in the random forest model, and the random forest model is obtained by training with historical auditing data as a training sample;
and determining a sub-recognition result according to the classification result output by each decision tree in the random forest model and the weight corresponding to each decision tree in the random forest model.
6. The medical laboratory clinical biochemical inspection automatic auditing system according to claim 5, characterized in that the feature recognition module adds weights of decision trees with the same classification result in the random forest model, determines the classification weight of each classification result, and takes the classification result with the largest classification weight as a sub-recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210148648.8A CN114496196A (en) | 2022-02-18 | 2022-02-18 | Automatic auditing system for clinical biochemical inspection in medical laboratory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210148648.8A CN114496196A (en) | 2022-02-18 | 2022-02-18 | Automatic auditing system for clinical biochemical inspection in medical laboratory |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114496196A true CN114496196A (en) | 2022-05-13 |
Family
ID=81481674
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210148648.8A Pending CN114496196A (en) | 2022-02-18 | 2022-02-18 | Automatic auditing system for clinical biochemical inspection in medical laboratory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114496196A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116843986A (en) * | 2023-09-04 | 2023-10-03 | 四川省医学科学院·四川省人民医院 | Image recognition-based automatic auditing method for immunostationary electrophoresis detection data |
-
2022
- 2022-02-18 CN CN202210148648.8A patent/CN114496196A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116843986A (en) * | 2023-09-04 | 2023-10-03 | 四川省医学科学院·四川省人民医院 | Image recognition-based automatic auditing method for immunostationary electrophoresis detection data |
| CN116843986B (en) * | 2023-09-04 | 2023-12-08 | 四川省医学科学院·四川省人民医院 | Image recognition-based automatic auditing method for immunostationary electrophoresis detection data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111325103B (en) | Cell labeling system and method | |
| CN108595657B (en) | Data table classification mapping method and device of HIS (hardware-in-the-system) | |
| CN110141219A (en) | Myocardial infarction automatic testing method based on lead fusion deep neural network | |
| CN110141220A (en) | Myocardial infarction automatic testing method based on multi-modal fusion neural network | |
| CN115064215B (en) | Method for tracing strains and identifying attributes through similarity | |
| CN113392894A (en) | Cluster analysis method and system for multi-group mathematical data | |
| CN115312183A (en) | Intelligent interpretation method and system for medical inspection report | |
| CN115691722A (en) | Quality control method, apparatus, device, medium and program product for medical data detection | |
| CN114496196A (en) | Automatic auditing system for clinical biochemical inspection in medical laboratory | |
| CN108447047A (en) | Acid-fast bacilli detection method and device | |
| CN112767349B (en) | Reticulocyte identification method and system | |
| CN114707608A (en) | Medical quality control data processing method, apparatus, device, medium, and program product | |
| CN114220542A (en) | Physical examination information management method and device, storage medium and computing equipment | |
| CN112599250A (en) | Postoperative data analysis method and device based on deep neural network | |
| Lin et al. | Automated malaria cells detection from blood smears under severe class imbalance via importance-aware balanced group softmax | |
| CN113539475A (en) | Disease screening and diagnosis method using blood routine test data only | |
| CN111863135A (en) | False positive structure variation filtering method, storage medium and computing device | |
| CN116129182A (en) | Multi-dimensional medical image classification method based on knowledge distillation and neighbor classification | |
| CN115700557A (en) | Method, device and storage medium for classifying nucleic acid samples | |
| CN113889274B (en) | Method and device for constructing risk prediction model of autism spectrum disorder | |
| CN115274091A (en) | Medical information analysis method and system | |
| CN116994647A (en) | Method for constructing model for analyzing mutation detection result | |
| CN117708569B (en) | Identification method, device, terminal and storage medium for pathogenic microorganism information | |
| CN114252370A (en) | Cell reclassification method of reader, sample analysis method and related equipment | |
| CN111079420B (en) | Text recognition method and device, computer readable medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |