Exemplary embodiments of the present invention will be described in detail based on the following figures, wherein:
A description will be hereinafter given of an exemplary embodiment of the invention with reference to the drawings.
Operations of the document search system will be hereinafter described with reference to the flowchart.
By an operation of the operating portion 105 (keyboard or the like) by a medical doctor, the medical finding input portion 12 inputs a medical finding written with a free expression (S101). For example, when a radio surgeon interprets an MRI image, an input medical finding such as “In a FLAIR image T2 emphatic image, nodule or porphyritic high signals are widely shown in the pons, the bilateral thalamus, the radiate crown region, the bilateral white matter and the like” is input. The input medical finding is output to the similar medical finding searching portion 14.
The similar medical finding searching portion 14 executes natural sentence search for past report information retained in the past report information retaining portion 24 by using the input medical finding as a key, and thereby acquires past report information including a past medical finding (second medical finding) similar to the input medical finding (first medical finding) (S102).
The natural sentence search is realized as follows. That is, morphological analysis of the input medical finding is performed, and the input medical finding is divided in unit of word. Further, only important words are extracted. Full text search is made for the past medical findings in the past report information by using these important words as a key. The search target is not limited to the full text of the past medical findings, but may be part thereof. The past report information acquired by the search is output to the diagnosis prospect extracting portion 16.
The diagnosis prospect extracting portion 16 extracts the diagnosis in the input past report information as a prospect (S13). The extracted diagnosis is output to the keyword extending portion 18. The extracted diagnosis is the diagnosis given by the medical doctor based on the past medical finding similar to the input medical finding. Therefore, such an extracted diagnosis can be regarded as a diagnosis based on experiences of the medical doctor. However, at this point, it is not possible to determine that the extracted diagnosis has a medical evidence. Therefore, in the document search system, the document search is made by using the acquired diagnosis prospect as a key, and thereby document information as a medical evidence of the medical finding is acquired.
That is, the keyword extending portion 18 searches for a word related to the input diagnosis prospect (related word) and a word translated into a given language from the input diagnosis prospect (translated word) by using a thesaurus and a bilingual dictionary that are retained in the thesaurus and bilingual dictionary retaining portion 26 (S104).
The related word and the translated word that are acquired by the keyword extending portion 18 are output to the document searching portion 20 together with the diagnosis prospect. The document searching portion 20 performs full text search for the document information retained in the document retaining portion 28 by using the diagnosis prospect, the related word, and the translated word as a keyword, and acquires document information having one of the diagnosis prospect, the related word, and the translated word as the document information as a medical evidence of the input medical finding (S105). The document information acquired by the search is output to the document display controller 22.
The document retaining portion 28 is a database in which document information related to the medical science is registered. The document retaining portion 28 does not necessarily exist inside the document search system, but may be connected to the document searching portion 20 via a network such as the Internet. The document retaining portion 28 desirably includes registered document information that is up to date as much as possible.
The document display controller 22 controls displaying the input document information on the monitor 106 (S106). The medical doctor reads the displayed document information and operates the operating portion 105, and thereby identifies the document information as a medical evidence of the medical finding, identifies the diagnosis, and adds a medical finding.
A description will be given of screen transition of the monitor 106 in the foregoing operations.
Then, when the operating portion 105 (mouse or the like) is operated by the medical doctor, and the diagnosis search button region 202 is selected by, for example, being clicked, the medical finding is input (S101 of
Further, the operating portion 105 (mouse or the like) is operated by the medical doctor, and one of the diagnosis prospects displayed in the diagnosis prospect display space 203 is selected by, for example, being clicked. Then, the related word and the translated word of the selected diagnosis prospect are created (S104 of
Next, the operating portion 105 (mouse or the like) is operated by the medical doctor, and one of the document information displayed in the document information display space 204 is selected by, for example, being clicked. Then, as shown in
Further, if the operating portion 105 (mouse or the like) is operated by the medical doctor, and one of check spaces 206 in front of the document information in the document information display space 204 is selected by, for example, being clicked, it is possible that, as shown in
As above, in the document search system, when the medical doctor prepares the medical finding with the free expression (natural language sentence or the like), the diagnosis prospect in the past report information including the past medical finding similar to such a medical finding is extracted, the document information including one of the diagnosis prospect, the related word of the diagnosis prospect, and the translated word of the diagnosis prospect is searched, and thereby such document information is acquired as documented information as a medical evidence of the medical finding. As a result, the document information as a medical evidence of the medical finding can be appropriately searched.
As shown in
Further, the document information acquired by the search by the document searching portion 20 may be displayed in descending order of relevance degrees in relation to the input medical finding. In this case, the document searching portion 20 as a calculating portion acquires the medical finding input by the medical finding input portion 12 (input medical finding) via the similar medical finding searching portion 14, the diagnosis prospect extracting portion 16, and the keyword extending portion 18. Next, the document searching portion 20 calculates the relevance degree in relation to the input medical finding for every piece of document information acquired by searching for the document information retained in the document retaining portion 28. For example, the relevance degree is calculated by extracting a keyword of the input medical finding and a keyword of the document information respectively, and comparing these keywords to each other by spatial vector method. Then, the document searching portion 20 adds the calculated relevance degree to each piece of document information acquired by the search, and outputs the document information added with the relevance degree to the document display controller 22.
The document display controller 22 controls sorting out the document information added with the relevance degree in descending order of relevance degrees and display the sorted result on the monitor 106. Thereby, it is possible to display the document information according to the relevance degree. For example, the document information highly related to the input medical finding can be displayed preferentially. As a result, the medical doctor can efficiently acquire the document information as a medical evidence of the medical finding.
In a case where the past report information retained in the past report information retaining portion 24 is prepared by a plurality of medical doctors, expressions for the same diagnosis may be largely varied according to the preparer. This results in the possibility that appropriate document information is not able to be acquired. In this case, it is desirable that the input medical finding be input by using various expressions.
The medical finding extending portion 30 creates a medical finding with a different expression in relation to the input medical finding (differently expressed medical finding). As a method of changing the expression, for example, a keyword in the input medical finding is extracted, and the keyword is substituted with a synonym or a paraphrastic word of the keyword. To acquire the synonym or the paraphrastic word of the keyword, it is necessary to prepare the similarity degree between the keyword and the synonym or the paraphrastic word. The similarity degree may be previously defined. Otherwise, the similarity degree may be realized by scoring the similarity degree between words included in medical findings of past report information that include the same diagnosis, out of the past report information retained in the past report information retaining portion 24. After that, the similar medical finding searching portion 14 executes natural sentence search for the past report information retained in the past report information retaining portion 24 by using the differently expressed medical finding as a key, and acquires past report information including a past medical finding similar to the differently expressed medical finding. As described above, by using various medical findings that have the same content with a different expression, the search range of the document information is widened, and thereby the document information as a medical evidence of the medical finding can be more appropriately searched for.
Further, the following method may be used. The document retaining portion 28 has the past report information or typical report information that has the same format as the past report information and serves as a model of document information. The past report information or the typical report information is acquired by search, and a portion thereof is further acquired as the input medical finding.
The medical finding input portion 12 inputs the input medical finding by a medical doctor's operation of the operating portion 105 (keyboard or the like). After that, as in the foregoing description, the similar medical finding searching portion 14 searches the past report information including the past medical finding similar to the input medical finding, the diagnosis prospect extracting portion 16 extracts the diagnosis prospect in the past report information, and the keyword extending portion 18 acquires the related word and the translated word.
The document searching portion 20 performs full text search for the past report information or the typical report information as the document information retained in the document retaining portion 28 by use of the diagnosis prospect, the related word, and the translated word as a keyword, and acquires the information having one of the diagnosis prospect, the related word, and the translated word. The document display controller 22 controls displaying the input past report information or the input typical report information on the monitor 106, and outputs the past report information or the typical report information to the text selecting portion 32.
The text selecting portion 32 outputs a section selected by an operation of the operating portion 105 by the medical doctor (for example, a partial text of the medical finding) out of the input past report information or the input typical report information to the medical finding input portion 12. The medical finding input portion 12 adds the selected section to the input medical finding. After that, document search is performed by using the input medical finding added with the selected section. As described above, at least a portion of the past report information or a portion of the typical report information is acquired and added to the input medical finding. Thereby, the past report information or the typical report information acquired by the search can be reused for creating a new medical finding, and efficiency of creation of the medical finding can be improved.
In the foregoing exemplary embodiment, the document information is searched by using the diagnosis prospect in the past report information, the related word of the diagnosis prospect, and the translated word of the diagnosis prospect as a keyword. However, it is also possible that the document information is searched by using a word other than the diagnosis prospect included in the past report information, the related word of the word, and the translated word of the word as a keyword.
A document search method employed according to an aspect of the present invention is performed with a Central Processing Unit (CPU), Read Only Memory (ROM), Random Access Memory (RAM), and the like, by installing a program from a portable memory device or a storage device such as a hard disc device, CD-ROM, DVD, or a flexible disc or downloading the program through a communications line. Then the steps of program are executed as the CPU operates the program.
The foregoing description of the exemplary embodiments of the present invention has been provided for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Obviously, many modifications and variations will be apparent to practitioners skilled in the art. The exemplary embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, thereby enabling others skilled in the art to understand the invention for various embodiments and with the various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the following claims and their equivalents.
This application is based on and claims priority under 35 USC 119 from Japanese Patent Application No. 2006-194923 filed Jul. 14, 2006.
Number | Date | Country | Kind |
---|---|---|---|
2006-194923 | Jul 2006 | JP | national |