阅读说明：本技术 一种基于模糊匹配的医疗事件编码方法 (Medical event coding method based on fuzzy matching ) 是由 霍红建 孟凡强 陈阶 于 2020-06-19 设计创作，主要内容包括：本发明提供了一种基于模糊匹配的医疗事件编码方法,该基于模糊匹配的医疗事件编码方法包括：对需要匹配的医疗事件的数据进行抽取关键词的标签；选择对应版本的专业词典；将步骤一中抽取的关键词的标签与专业词典进行模糊匹配,计算相似度；将计算的相似度与阈值进行比较；对相似度高于阈值的关键词进行编码。本发明充分考虑了药品的临床试验中发生医疗事件数据用语无法统一,格式复杂,人工编码误差大,没有统一标准的特点,采用计算机辅助实现了对医疗事件的标准化编码,工作量低、工作效率高、准确度高,为后续数据的统计分析工作做好充分的前期准备工作,节省临床试验成本,并加快整个临床试验的完成速度。(The invention provides a medical event coding method based on fuzzy matching, which comprises the following steps: extracting a label of a keyword from the data of the medical event needing to be matched; selecting a professional dictionary of a corresponding version; carrying out fuzzy matching on the labels of the keywords extracted in the step one and a professional dictionary, and calculating the similarity; comparing the calculated similarity with a threshold; and coding the keywords with the similarity higher than a threshold value. The invention fully considers the characteristics of non-uniform medical event data wording, complex format, large manual coding error and no uniform standard in the clinical test of the medicine, realizes the standardized coding of the medical event by adopting the computer assistance, has low workload, high working efficiency and high accuracy, makes full early-stage preparation work for the statistical analysis work of subsequent data, saves the cost of the clinical test and accelerates the completion speed of the whole clinical test.)

1. A medical event coding method based on fuzzy matching is characterized by comprising the following steps:

the method comprises the following steps: extracting a label of a keyword from the data of the medical event needing to be matched;

step two: selecting a professional dictionary of a corresponding version;

step three: carrying out fuzzy matching on the labels of the keywords extracted in the step one and a professional dictionary, and calculating the similarity;

step four: comparing the calculated similarity with a threshold;

if the calculated similarity reaches 100%, the similarity is completely matched, and the extracted keywords are used as codes of the series of medical events; if the similarity is higher than the threshold value but not higher than 100%, outputting the first three matching words with high similarity in the professional dictionary as reference, and carrying out manual interference encoding; and if the threshold value is lower than the threshold value, repeating the steps from the first step to the fourth step.

2. The method according to claim 1, wherein the corresponding version of the specialized dictionary in the second step is MedDRA, which is a set of medical standard terms.

3. The method as claimed in claim 1, wherein the medical event is an adverse event collected from a clinical trial on a medicine and a medical history of a patient.

4. The method according to claim 1, wherein the threshold is 50% -60%.

5. The medical event coding method based on fuzzy matching as claimed in claim 1, wherein the similarity calculation method is an edit distance algorithm, a cosine theorem algorithm of a vector space model.

6. The medical event coding method based on fuzzy matching as claimed in claim 1, wherein said step one of extracting keywords is to extract main stems, remove interfering words and wrongly written words.

7. The medical event coding method based on fuzzy matching as claimed in claim 1, wherein said computer system of said medical event coding method based on fuzzy matching comprises: the label extraction module is used for extracting the labels of the keywords after combing the medical event data needing to be matched; the matching label judging module is used for calculating the similarity between the label of the extracted keyword and the professional dictionary data according to a set similarity calculation method; the calculated similarity is compared with a threshold to determine whether the keyword can be used as a code for the series of medical events.

Technical Field

The invention belongs to the technical field of medical information management, and particularly relates to a medical event coding method based on fuzzy matching.

Background

In the clinical trial process of drugs, many medical events are generated, and in order to complete the statistics and summarization of the same and similar events, the medical events occurring in the clinical trial of drugs need to be counted and summarized according to certain rules, and are managed in a standardized manner.

At present, the coding mode for medical events is that coding personnel manually codes according to own experience or uses a tool provided by a coding dictionary to inquire keywords and search corresponding units. The workload of the two modes is huge, and subjective consciousness of encoding personnel can be brought in the encoding process, and because individual differences of different encoding personnel are large, the encoded content has large difference factors, and data analysis of clinical tests of later-stage medicines can be seriously influenced.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention provides a medical event coding method based on fuzzy matching, which is low in workload, high in working efficiency and high in accuracy and is used for coding by means of a computer.

The technical scheme for solving the technical problems is as follows:

a medical event coding method based on fuzzy matching comprises the following steps:

the method comprises the following steps: extracting a label of a keyword from the data of the medical event needing to be matched;

step two: selecting a professional dictionary of a corresponding version;

step three: carrying out fuzzy matching on the labels of the keywords extracted in the step one and a professional dictionary, and calculating the similarity;

step four: comparing the calculated similarity with a threshold;

if the calculated similarity reaches 100%, the similarity is completely matched, and the extracted keywords are used as codes of the series of medical events; if the similarity is higher than the threshold value but not higher than 100%, outputting the first three matching words with high similarity in the professional dictionary as reference, and carrying out manual interference encoding; and if the threshold value is lower than the threshold value, repeating the steps from the first step to the fourth step.

Further, the professional dictionary of the corresponding version in the second step is MedDRA, which is a set of medical standard terms.

Further, the medical events are information such as adverse events and medical history of patients collected when a clinical trial is performed on a certain medicine.

Further, the threshold value is 50% -60%.

Furthermore, the similarity calculation method is an edit distance algorithm and a cosine theorem algorithm of a vector space model.

Further, the extraction of the keywords in the first step is to extract a main stem and remove interfering words and wrongly written characters.

Further, the computer system of the medical event coding method based on fuzzy matching comprises:

the label extraction module is used for extracting the labels of the keywords after combing the medical event data needing to be matched;

the matching label judging module is used for calculating the similarity between the label of the extracted keyword and the professional dictionary data according to a set similarity calculation method; the calculated similarity is compared with a threshold to determine whether the keyword can be used as a code for the series of medical events.

After the coded data of the medical events of a certain system is obtained by the method, the coded data is used for data management and statistical analysis of clinical tests of the medicines and is submitted to the Chinese and foreign clinical medicine supervision departments.

Compared with the prior art, the invention has obvious beneficial effects, and the technical scheme can show that:

according to the medical event coding method based on fuzzy matching, the medical event information is identified and coded by adopting automatic operation matching of a computer, so that the manual workload is greatly reduced, the working efficiency is high, and the method is easy to inquire and convenient to use; the algorithm is simple and easy to realize in programming; by extracting the keywords of the medical events, carrying out fuzzy matching on the keywords and the professional dictionary, then coding, and coding more accurately and standardizing, the problems that the existing manual entry is expressed in various modes, is not described in a standard way, has errors in entry, adopts abbreviative words or common names and the like are avoided, the accuracy of the result arrangement is improved, the efficiency is improved, sufficient preliminary preparation work is well done for the statistical analysis work of subsequent data, the cost of clinical tests is saved, and the completion speed of the whole clinical test is accelerated.

Detailed Description

The present invention will be described in further detail with reference to the following examples and examples, but it should not be construed that the scope of the above subject matter of the present invention is limited to the following examples.