JP2630405B2 - Document input method - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document input method for inputting diagnostic information while displaying a medical image on, for example, a medical image workstation and creating an image diagnostic report and the like, and in particular, The present invention relates to a document input method that can easily and quickly perform a document input operation.

[Conventional technology]

Recently, with the rapid development of computer technology, image workstations with various image processing functions,
The development of workstations for office automation (OA), which have a function of editing figures and integrate various databases, etc., is in progress. In such a situation, attribute information about a specific image is compiled as a document,
The demand for technology to be stored together with images has been increasing mainly in the medical field. For example, a technique of displaying a medical image such as an X-ray image and a CT image on a medical image workstation and inputting diagnostic information to create an image diagnostic report has attracted attention.

Conventionally, in order to facilitate the operation of inputting the diagnostic information and efficiently create an image diagnosis report, a table in which medical knowledge is coded or abbreviated is created in advance, and this table is created in advance. Is displayed on a document editing display terminal or the like, and the operator inputs necessary diagnostic information using a code or an abbreviation attached to the diagnostic information to be input while watching this.

[Problems to be solved by the invention]

However, in such a conventional document input method of creating an image diagnostic report or the like, it is impossible to display all codes or abbreviations of diagnostic information to be input on a screen of a document editing display terminal. It is difficult for an operator to search for a code or an abbreviation of diagnostic information that the operator wants to input, and it takes time to input. Further, in order to increase the input efficiency (input speed) of the diagnostic information, the above code or abbreviation must be stored to some extent by the operator, and the operator's input operation requires skill. Therefore, it was not something that anyone could easily input.

Therefore, an object of the present invention is to provide a document input method capable of solving such a problem.

[Means for solving the problem]

Means for solving the above-mentioned problems according to the present invention is to use a document processing apparatus having an input device, a computer, and a display terminal for document editing, and to input a required document by a character input operation on the input device. In the input method, the characters input by the input device are inspected in a knowledge database prepared in advance, and a prediction data table of words that can be inferred to be input from now is created. The inference result is extracted from the character string of at least one character input by the input device, and displayed on the display terminal for document editing as an input candidate word. If the candidate word matches the word to be input next, it is displayed. By inputting the word as it is, and if it does not match, by continuously inputting with the input device, the above candidate word is deleted and the input character is displayed. It is.

(Operation)

The document input method configured as described above prepares a knowledge database in which information for each highly relevant knowledge is prepared in advance, inspects the knowledge database while inputting characters one after another with an input device, Infers words related to the information and displays them as input candidate words. Depending on whether this candidate word matches or disagrees with the next word to be input, it may or may not be used as an input word and normal input continues. Or something to do.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a flowchart showing a processing procedure of a document input method according to the present invention, and FIG. 2 is a block diagram showing a processing device used for implementing the document input method of the present invention. FIG. 2 shows a processing device for inputting diagnostic information and creating an image diagnostic report or the like while displaying medical images on a medical image workstation, for example, an input device 1 such as a keyboard, a computer 2, and a CRT display. And a document processing device having a document editing display terminal 3 such as a document storage device and an image display terminal 5 for displaying a medical image such as a head X-ray CT image. A knowledge database 6 relating to, for example, image diagnosis is prepared in advance by collecting information for each highly knowledgeable. In FIG. 2, reference numeral 7 denotes a parameter input device called "mouse" for inputting parameters in the X and Y directions of the screen, for example.

Here, the knowledge database 6 has a format as shown in FIG. 3, for example. The third diagram shows the knowledge database 6 head- X-ray CT image as an example, the data D ₁ of the for making inferences, and data D ₂ of the smallest string to be inferred, the inference result data consisting of D ₃ Metropolitan. Data D ₁ of the order to perform the above reasoning, for example, the head X-ray
The words that represent medical terms and symptoms used in CT image imaging are listed. Data D ₂ of the smallest string to be inferred, for example be a minimal string represented by the minimum 1 character for distinguishing them against various disease name. Words representing the medical terms and symptoms of the data D ₁ for performing the above inference,
For example, character strings such as "cerebral infarction", "encephaloma", "prolapse", and "in the brain" are arranged for each of the words "low absorption", "high absorption", and "iso-absorption". Data D ₃ of the inference result, the term representing the various disease name which is distinguished by the data D ₂ of the smallest character string to be the reasoning are arranged to correspond to the data D ₂ of the minimal string. For example, "cerebral infarction" corresponds to "cerebral infarction", "brain tumor" corresponds to "brain tumor", "demyelinating disease" corresponds to "prolapse", and the like.

In FIG. 2, the original image is recorded in the image storage device 4, and the data of the original image is read out from the image storage device 4 in accordance with an instruction from the input device 1 under the control of the computer 2, and the image is read. The image is displayed on the image display terminal 5. Next, while viewing the displayed original image, related information is selected from the knowledge database 6 under the control of the computer 2 and displayed on the document editing display terminal 3. After that, the character to be input is input from the input device 1.

Next, the processing procedure of the document input method of the present invention will be described with reference to the flowchart shown in FIG. First, the operator inputs characters to be input using the input device 1 one after another (step). Next, the input sentence thus input is checked in the knowledge database 6 (step). In other words, a 1: 1 correspondence is established between the document input so far and the data D ₁ for inferring the knowledge database 66 shown in FIG. 3, and “low absorption”, “high absorption”, or “iso-absorption” Check for the word Now, if you had term "low absorption" in the already input sentence are denoted by reference numerals d ₁ among the data D ₁ of the order making inferences shown in words and Fig. 3 of the "low absorption" It can be seen that the data corresponds. Next, a prediction data table of words that can be inferred to be inputted from now on the data having a 1: 1 correspondence is created (step). That is, as shown in FIG. 4, a portion of d ₁ shown in FIG. 3 data bring entirety and prediction data table 8. This prediction data table 8
Since the medical term "low absorption" in the already input sentence is present, as the data D ₂ of the smallest string to be inferred, "brain infarction", "brain tumor", "de", " It shows that there are four words of "in the brain", and the data D _{3 of the} inference results are words of "cerebral infarction", "brain tumor", "demyelinating disease", "cerebral hemorrhage" I have.

In this state, it is assumed that characters are input one after another by the input device 1 and one character is input in the middle of a sentence. Then, take out the inference result data D ₃ corresponding to the data D ₂ of the smallest string to be inferred by searching the prediction data table 8 shown in FIG. 4 (step). That is, the word “demyelinating disease” corresponding to “de-” in the data D ₂ of the minimum character string is extracted as an inference result. At this time, the data D ₂ of the smallest character string to be reasoning prediction data table 8 in the case of a single character, such as "de", search as compared to the last character entered. In the case of two characters such as "cerebral infarction", the character is searched for by comparing it with the last input character and the character immediately before it. Similarly,
In the case of n characters, a search may be made by comparing the last input character with the character string up to the character (n-1) before.

Next, the word "demyelinating disease" of the inference result extracted as described above is used as a candidate word for the next input.
(Step). Then, the operator determines whether or not the displayed candidate word matches the word to be input next, and operates the adoption key (not shown) provided on the input device 1 if it matches. Then, the step proceeds to the “YES” side, and the displayed candidate word “demyelinating disease” is directly input (step). afterwards,
Returning to the step, the operator successively inputs the character to be input with the input device 1, and thereafter repeats the above procedure.

On the other hand, if the candidate word displayed by the operator's judgment does not match the word to be input next, the above-mentioned adoption key is not operated, and the step proceeds to "NO". And
For example, a character to be input following "de" is input as usual using the input device 1 (step). For example, an input such as “dehydration symptom” is input. Then, the candidate word “demyelinating disease” previously displayed on the document editing display terminal 3 is erased, and the word “dehydration symptom”, for example, input continuously in the step is displayed (step). This returns to the normal input state, and then returns to the step before. Thereafter, the above procedure is repeated while branching to the “YES” side or the “NO” side according to the determination of the step.

FIG. 2 shows a case of a processing device that inputs diagnostic information while displaying a medical image on a medical image workstation and creates an image diagnostic report and the like.
The present invention is not limited to this, and can be similarly applied to a normal document processing apparatus that does not include the image storage device 4 and the image display terminal 5. Also, the knowledge database 6
Not limited to medical knowledge, knowledge in any field may be accumulated. Thus, the method of the present invention can be applied to a case where a general document is input.

〔The invention's effect〕

Since the present invention is configured as described above, a knowledge database 6 in which information is summarized for each highly relevant knowledge is prepared in advance, and the input device 1 is used to input characters one after another to check the knowledge database 6. In addition to inferring a word related to the already input information and displaying it as an input candidate word, the candidate word may or may not be used as an input word depending on whether it matches or disagrees with the next word to be input. Or continue typing. Therefore, the operator only has to input the characters to be input one after another, and the input operation of the document can be performed easily and quickly without searching for the input information code or the abbreviation as in the related art. Further, the operator does not need to memorize the codes or abbreviations to some extent in order to increase the information input speed unlike the related art, and the input operation by the operator does not require any particular skill. Therefore, anyone can easily input a required document.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a processing procedure of a document input method according to the present invention, FIG. 2 is a block diagram showing a processing device used for implementing the document input method of the present invention, and FIG. 3 is an example of a format of a knowledge database. FIG. 4 is an explanatory diagram showing an example of a prediction data table. 1 ... input device, 2 ... computer, 3 ... display terminal for document editing, 6 ... knowledge database, 8 ... prediction data table.

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Yasushi Miyazaki, 2-1 Shinjuyo, Kashiwa-shi, Chiba Prefecture Inside Hitachi Medical Research Laboratory, Inc. (72) Inventor Makie Shibayama 2-1 Shinjuyo, Kashiwa-shi, Chiba Stock (72) Inventor Osamu Sasahara 2-1 Shinjyuyo, Kashiwa-shi, Chiba Hitachi Medical Research Institute, Inc. (56) References JP-A-61-74062 (JP, A) JP-A-61 −145677 (JP, A)

Claims (1)

(57) [Claims] 1. A document input method for inputting a required document by performing a character input operation on the input device using a document processing device having an input device, a computer, and a display terminal for document editing. The input characters are inspected in a knowledge database prepared in advance, and a predicted data table of words that can be inferred to be input is created. In addition, the predicted data table is searched and a minimum of one character to be input by the input device is next searched. Extract the inference result from the string of
It is displayed on the display terminal for document editing as a candidate word for input. If the candidate word matches the word to be input next, enter it as it is, and if it does not match, continue inputting with the input device. A document input method for erasing the candidate word and displaying the input character.