JPH0214360A - Diagnosis report preparation supporting device - Google Patents

Abstract

PURPOSE:To shorten the preparing time of a diagnosis report and further, to unite expression for the view of doctors by preparing questions, which are needed for the diagnosis report preparation of the human body, in advance, successively answering these questions and electronically executing the edition of the diagnosis report. CONSTITUTION:A question item storing area 13 is provided to store the question item, which is needed for preparing the diagnosis report of the human body, and a question item calling means 14 is equipped to selectively call the question item to a spot designated by a designating means 12, from the question index storing area 13. The question to be needed for diagnosis is selected by the question item calling means 14 and displayed on a question display means 15. The doctor, etc., to execute operation inputs the answer to the question, displayed by the question display means 15, with an answer input means 16. Here, it is not always necessary to input the answer in a sentence. Then, it can be also available the several answers are prepared to the question and one of the answers is selected by a number, etc. When the answer is inputted by the answer input means 16, this answer is edited by an editing means 17 and the diagnosis report is prepared.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a diagnostic report creation support device used when, for example, a radiologist creates a patient's diagnostic report based on images of an affected area.

"Prior Art" Various image diagnoses using X-rays, ultrasound, positrons, MRI (nuclear magnetic resonance tomography), etc. using magnetism have become widely used. The pioneer of such image diagnosis was simple photography using X-rays. Simple photographs are effective for general examination to determine whether there is a lesion within the body. However, plain X-ray photography involves the following problems because it projects the three-dimensional human body onto a two-dimensional film.

(i) The three-dimensional structure of the human body is unknown from a two-dimensional film.

(ii) Overlapping images occur in plain X-ray photographs,
This creates a shadow that differs from the original shape of the tissue or lesion image to be observed, causing confusion in image judgment.

In order to eliminate such defects, it is necessary to project a two-dimensional part of the human body onto a two-dimensional film for inspection.To do this, the human body is placed on a rotating table and an X-ray tube is A rotational imaging method was devised in which images were taken by rotating the film.With this rotational imaging method, differences in absorption between tissues had to be expressed as density differences on the film, so there was a problem with contrast in soft tissues. Ta.

Therefore, C′l” (Computed tomolo
gy), that is, computerized tomography was devised. This CT
In this method, a tube is rotated around the human body to irradiate a thin layer of X-ray flux, while the transmitted X-ray flux is detected by a detector and the X-ray absorption value is calculated by a computer.

Now, with CT, it is possible to clearly obtain X-ray images of lesions, organs, and tissues surrounded by membranes and fat.

It is also possible to directly depict tissues and lesions within organs with slight differences in absorption as images. in general,
The lesions are concentrated in soft tissues. Therefore, CT, which can obtain images of soft tissues without the use of ultra-soft X-rays or imaging agents, is being used for a wide range of diseases such as those of the brain and internal organs, and is being used for cerebrovascular diseases. It provides a lot of information on the treatment of diseases such as cancer and brain tumors.

The above explained CT using X-rays, but as mentioned above, image diagnosis using ultrasound, positron, etc. is also performed, and doctors can make findings regarding various diseases by observing these images. , creating diagnostic imaging reports.

[Problem to be solved by the invention] Incidentally, the creation of such an image diagnostic report has traditionally been done manually by the doctor himself, as in the creation of other diagnostic reports, or by recording the dictation and having a secretary write it down. It was done by For this reason, the following problems occurred.

(1) Since each image diagnosis report, such as a radiographic image diagnosis, was filled out manually on a prescribed form, it took a considerable amount of time to prepare.

(11) If no abnormality is seen in the images, etc., the same findings will be made for the patient, but even in this case, multiple reports with the same contents will not be created. It was not efficient.

(iii) Even if the patient has the same case, the wording of the image diagnosis report varies slightly depending on the doctor, and in some cases, the interpretation of the report may be completely different from that of the person reading the report.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a diagnostic report creation support device that can shorten the time required to create a diagnostic report and unify the expression of findings by doctors.

"Means for Solving the Problem" In the present invention, as shown in principle in FIG. 1, (i) a model display means 11 is prepared for displaying a human body model of a desired part including the part to be diagnosed; .

The model display means 11 is, for example, a high-definition display, and in the case of a radiation image diagnosis report, a modeled radiograph is displayed here. The diagnostic report creation support device is equipped with (ii) instruction means 12, which instructs the part to be diagnosed in the model displayed by the model display means 11. For example, the model display means 11 displays When a patient's radiograph exists that corresponds to the model, the doctor looks at the patient's radiograph, selects a location for diagnosis, and indicates its position on the human body model.

The diagnostic report creation support device includes a question item storage area 13 that stores question items necessary to create a human body diagnostic report, and a question item storage area 13 for selecting question items for locations specified by the instruction means 12. Question item calling means 14 is provided for calling up questions. Then, the question item calling means 14 selects a question necessary for diagnosis, and the selected question is displayed on the question display means 15. A doctor or the like who operates the diagnostic report creation support device inputs an answer to the question displayed by the question display means 15 using the answer input means 16. Here, the answer does not necessarily need to be input as text, and several answers to the question may be prepared in advance and selected by number or the like.

When the answer is input by the answer input means 16, the editing means 17 edits it to create a diagnostic report.

That is, according to the diagnostic report creation support device of the present invention, for example, when a radiologist selects a model necessary for diagnosis from among a plurality of human body models and further specifies a location to be diagnosed on this model, this region is Questions necessary to make a finding are displayed on question display means 15 such as a display.

Therefore, if a radiologist or the like inputs answers to these questions one after another, the editing means 17 edits the answers and creates a diagnostic report in a unified writing format.

"Example" The present invention will be described in detail below with reference to Examples.

FIG. 2 shows an overview of the configuration of the diagnostic report creation support device of this embodiment. The diagnostic report creation support device of this embodiment is connected to one coaxial cable 21 that constitutes a LAN (Local Area Network) called Ethernet developed by Xerox Corporation, and is connected to a data input/editing section. 22 and a laser printer 23. The data input/editing section 22 is further comprised of the following parts.

(i) CPU (Central Processing Unit) 24: This is a device that serves as the control center of this diagnostic report creation support device.

(11) Hard disk 25: This is a storage medium that stores programs for performing various controls and necessary contents of the created diagnostic report.

(iii) RAM 26 Random access memory for storing programs read from the hard disk 25 and for temporarily storing various data.

(iv) Bit map display 27: This is a display in which 1 bit on the screen corresponds to 1 bit in the display area of RAM 26, and allows multi-window display as a display method that displays multiple windows simultaneously on the screen. It is.

(v) Keyboard 28: This is a device that inputs data through key operations.

(vi) Mouse 29: It is a type of pointing device.

FIG. 3 shows the first stage of work when creating a report for the radiographic examination area 7 using this diagnostic report creation support device. The CPU 24 first displays an initial screen on the bitmap display 27 (step 2).

FIG. 4 shows this initial screen.

The initial screen includes (a) “Application” and (b)
Items for each work such as "Create findings", (C) "Edit dictionary", (D) "Edit X-ray anatomical chart", and (E) "Finish".
Items such as the following are displayed in the window 31. The operator (for example, a radiologist) uses the mouse 29 to move the illustrated arrow-shaped cursor to the desired selection item, and clicks a button (not shown) placed on the mouse 29 to select one of these options. You will be asked to select an item.

Here, when the item (a) "Application" is selected, the operator instructs various application programs to be selected and executed. (b) If the item "Create findings" is selected, support for creating a diagnostic report as described in this specification will be provided. (c) If the item "Dictionary Edit" is selected, the mode is set in which the dictionary used for creating findings is edited. (d) If the item "X-ray anatomical chart editing" is selected, the mode is set in which an anatomical chart based on X-ray photography is edited. (e) The "End" item is selected to end the diagnostic report creation support state.

It should be noted that a head model window 41 is displayed on the display screen shown in step (2) to indicate that data related to brain diagnosis will be handled. In this specification, the case where (b) "Create findings" is selected to create a diagnostic report will be described below.

Now, when the operator selects the item "Create findings" (
Step 2; Y), three windows 33 to 35 are newly displayed on the bitmap display 27 as shown in FIG. 5 (Step 2). Of these, the data input window 33 for creating findings is used to create findings.
(i) type of finding, (11) finding number, (c) finding creator, and (d) date are entered.

Here, (1) type of finding refers to whether or not a contrast medium has been injected, and here the words "head CT (simple only)" are input.

This means that no contrast medium was injected.

(ii) The finding number is the serial number of the radiographic image diagnosis report. The key input pad window 34 and virtual keyboard 35 are used to input these data, including (c) finding creator, and (d) date.

When the operator inputs all the data into the window 33 for inputting data for creating findings and selects the end key 39 in the window 38 (step ■), the window 33 for inputting data for creating findings is replaced by a window as shown in FIG. A patient data input window 38 is displayed (step ■)
.

Each item in the data input window 33 for creating findings can be selected from among the options prepared in advance, or by checking the content automatically displayed by the device, or modifying it as appropriate, and changing the CR in the keypad window 34. It is set by selecting it or by inputting a return key (not shown) on the virtual keyboard 35.

After setting all the items, the end key 36 can be pressed.

Returning to FIG. 6, the explanation will be continued. The patient data entry window 38 displays the following items: (a) ID number, (c) name, (c) date of birth, (d) phone number, (e) gender, and (f) address. It has become. The diagnostic report creation support device displays a selection of ID numbers of patient data registered in the hard disk 25 in advance. Other data is called up using the ID number specified from among the selections as a key, and other items are automatically displayed based on this. The operator checks the contents of each item and makes corrections as appropriate.

On the other hand, when setting a new ID number, input is made from the key input pad window 34, and other accompanying data is also input from the key input pad window 34.
and input using the virtual keyboard 35.

When the patient data has been completely input into the patient data input window 38, the operator selects the end key 39 in the window 31 (step 2).

FIG. 7 shows the state of the bitmap display after this. bit map display 27
, a judgment message ``Are there any findings observed in the X-ray anatomical diagram?'' is displayed (Step ■ in Figure 3).

In response to this, the operator operates the mouse 29, places the cursor indicated by an arrow on the "Yes" or "No" position, and clicks the above-mentioned button (Step 2).

As a result, if it is determined that there are findings (Y), a diagnostic report of "findings found", which will be explained next, will be created (step 2). On the other hand, if it is determined that there are no findings (step ■; N), a diagnostic report with no findings indicates that nothing particularly bad is seen in the patient's radiological image that the operator currently has. is created (step @).

Note that if any of the items "Application", "Dictionary editing", or "X-ray anatomical diagram editing" is selected instead of the "Finding creation" item in step ■ (Step 0; Y), the work of each selected item will be executed (step ■).

FIG. 8 shows an outline of the work performed when the operator determines that there is a finding in the radiation image of the patient that he currently has. That is, in this case, as shown in FIG.
is opened, and the words "Please select the tomographic plane of the CT scan" are displayed here (Step ■ in Figure 8).

The operator operates the mouse 29 while looking at the currently held radiographic image, aligns the cursor represented by a ``■'' with the corresponding tomographic plane in the head model window 41, and clicks a button. In this way, the tomographic plane of the CT scan is selected (step ■; Y). When a tomographic plane is selected, the CPLI 24 selects the applicable one from among the cross-sectional image models of the CT scan read out from the vertical screen 25, and displays this in the center of the bitmap display 27 (step 2). . Additionally, an instruction to ``select the area to be specified with the mouse'' is displayed (step ■).

FIG. 10 shows this display state. A window 44 of a CT scan cross-sectional image model is displayed in a large size on the bitmap display 27, and a window 45 with instructions for region selection is opened above the window 44.

At this point, the operator looks at the patient's current radiology images and identifies areas where there are any abnormalities.

Then, the user places the ■-shaped cursor at the position corresponding to that area in the window 44 and clicks the button (step ■).

As an example, assume that the operator recognizes an abnormality in the white matter part of the right temporal lobe located to the left of the center of the cross-sectional image model, places the ■-shaped cursor on this part, and clicks the button of the mouse 29.

Note that the position where the button of the mouse 29 is clicked in step (2) may be included in a plurality of areas. In that case, all the areas are displayed first, and the desired area is selected from among them by clicking the left button of the mouse 29.

FIG. 1I shows the contents displayed on the bitmap display when the right temporal lobe white matter region is selected.

The corresponding white matter portion of the right temporal lobe on the bitmap display 27 is displayed with halftone dots (step 2).

At the same time, the message ``Are you sure you want the specified position?'' is displayed in the window 45, and the specified area is confirmed using the mouse 29' displayed as an icon (step ■).In other words, if the operator confirms the Clicking the left button of the mouse 29 to indicate your intention (Y
), the work proceeds to the next step to create findings for that area (step ■).

On the other hand, when the right button of the mouse 29 is clicked to indicate a negative intention (step ■; N), the halftone dot display for the already specified area is cleared (step ■), and the process returns to step ■. The instruction "Please select the area to be specified with the mouse" will be redisplayed.

FIG. 12 shows details of the finding creation work shown in step (2) in FIG. In addition, in FIG. 8, one region is selected from one radiation image, but it may be possible to select a plurality of regions. In this case, findings regarding the disease specified from the plurality of selected sites will be created.

Now, when the "region" for creating findings in FIG. 12 is specified (step ■; Y), a window for finding findings is displayed in the window 44 for the CT scan cross-sectional image model, as shown in FIG. 13. 47 is set, and options are displayed (step ■).In this FIG.
In order to make the inside of the window 44 easier to understand, it is enlarged and displayed outside the window 44 for the purpose of explaining this specification. In this window 47, findings to be made regarding the white matter portion of the right temporal lobe are restored. The operator looks at the radiographic image of the patient in his hand, moves the cursor to the finding to be selected, and clicks the button on the mouse 29 to select one of the findings (step ■: Y). Then,
A group of necessary questions regarding the finding is stored in the question item storage area 1.
The first question is selected from the area on the RAM 26 corresponding to No. 3 (step ■). Then, the question is displayed in the window 45, and the answer group to be selected is displayed in the window 47 (step 2). The operator looks at the patient's radiographic image and selects one of them (step ■).

When the answer to the question is selected (Y), the next question is selected (step ■) unless all questions have been completed (step ■; Y1■; N).

Then, the questions and answers are similarly displayed in an alternative format on the pit map display 27 (step 2). In this way, questions are asked one after another (steps ■ to ■), and when all questions are completed (step ■; Y)
The diagnostic report creation support device edits the text based on the above answers, creates findings, and outputs them on the bit map display 27, and if necessary, prints out the contents from the laser printer 23. (Step ■)

Figures 14 to 25 show the steps ■ in the above explanation.
This is a concrete example of the subsequent display control.

When the item "high absorption area" is selected as a finding in FIG. 13, the question "Where is the finding found?" is displayed in the window 45 as shown in FIG. 14. Further, a window 49-1 displaying a group of answers to this question is opened at the upper right of the window 44 for the CT scan cross-sectional image model. Here, the operator selects the corresponding answer in the window 49-1 using the mouse 29. In this example, the item "in and around area" is selected. Note that for the item "Other", the display color of the background part is different to distinguish it from other parts. When the operator selects the item "Others," a separate window may be expanded to allow selection of more items, or the operator may simply specify "Others." It may also be processed as if the In the diagnostic report creation support device of this embodiment, when "Others" is selected with the mouse 29, a separate window for inputting other items is displayed, and the designation of "Others" is simply made. When performing this, the mouse 29 is clicked on a part outside the area of the window 49 to be set. The meaning of the item "Others" explained above with respect to FIG. 14 applies similarly to FIGS. 15 and subsequent figures.

When the answer to the first question displayed in the window 49-1 is given, the window 4 is displayed as shown in FIG.
5: “What is the distribution of the findings?” 2
The second question will be asked. At the same time, a window 49-2 is opened that displays a group of answers to this question.

In this example, the operator selects the "multiple occurrence" item.

Similarly, FIGS. 16 to 25 show the questions and the responses to these questions from the third to the final 122nd time, which will be conducted later. Each question is displayed in window 45, and the options are displayed in window 49-3.
~49-12 will be displayed. In the case of this patient, the operator selects the item "indistinct border" in FIG. 16, "scratches" in FIG. 17, and selects "ellipse" in FIG. 18. In addition, in the selection shown in FIG. 19, the operator determines that there is no need to make a selection for this question, and clicks with the mouse 29 outside the area of the window 49-6. In addition, in Figure 20', the item "dense staining is observed in some or all parts" is selected, and in Figure 21, "
"Mottled" is selected, "uniform" is selected in FIG. 22, "extremely" is selected in FIG. 23, "yes" is selected in FIG. 24, and "uniform" is selected in FIG. Then, the item "No" is selected.

FIG. 26 shows an example of a report created based on the above answers. In this way, the necessary input data shown in Figures 5 and 6 are displayed in the upper part of the report, and the lower part shows the data created from the data specified by the selection operations shown in Figures 7 and onwards. The results will be displayed.

The sentence expressing the findings is not simply a combination of selected parts of the options, but is analyzed according to a predetermined grammar (context-free grammar) and reconstructed as a natural sentence.

Incidentally, if it is determined that there are no findings in step (3) in FIG. 3, a report as shown in FIG. 27 will be created.

FIG. 28 shows the integrated analysis semantic network generated by the system for this sentence generation. This integrated analysis semantic network was generated using free context grammar. The question items and answer groups shown in Figs. 14 to 24, as well as the control of the question priority order, etc., can be controlled by the operator by instructing the item "Dictionary Edit" on the initial screen shown in Fig. 4. However, it is also possible to modify these as appropriate to change the structure of the sentence.

In the embodiments described above, the human body model is displayed as a simple figure, but for example, a realistic image obtained by reading a typical radiation image with an image sensor may be displayed on the bitmap display 27. Thereby, the operator can easily recognize the difference from the image currently in hand, and can perform diagnosis more objectively.

In addition, although the mouse was used to input answers etc. in the example, similar operations may be performed using other pointing devices, and the options can be selected by adding a symbol to each option. The code may be input from a keyboard or the like.

Further, in the embodiment, a case has been described in which findings are created on a radiographic image of the head, but the present invention can be similarly applied to image diagnosis of other parts. Furthermore, it goes without saying that images used for diagnosis are not limited to radiographic images.

"Effects of the Invention" As described above, according to the present invention, since the question items necessary for creating a human body diagnostic report are prepared in advance, it is possible to support the creation of an objective diagnostic report by answering the questions one by one. can. Furthermore, since the diagnostic report is edited electronically using the compiling means, it is not only possible to create a standardized report, but also the created report can be processed as is on a computer.

[Brief explanation of the drawing]

Figure 1 is a block diagram showing the principle of the present invention, Figures 2-2
Figure 7 is for explaining one embodiment of the present invention, of which Figure 2 is a block diagram showing an overview of the circuit configuration of the diagnostic report creation support device, and Figure 3 is a diagram for creating a radiographic image diagnosis report. A flowchart showing the first stage of work,
Figure 4 is a plan view showing an example of the initial screen displayed on the bitmap display of this device, and Figure 5 is the initial screen displayed on the bitmap display when "Create Findings" is selected on the screen in Figure 4. 6 is a plan view showing the screen displayed on the bitmap display after data input on the screen shown in FIG. 5 has been completed. FIG. 7 is a plan view showing the screen shown in FIG. 6. Fig. 8 is a plan view showing the screen displayed on the bitmap display after data entry has been completed; 9
The figure is a plan view showing the first screen displayed on the bitmap display when you select the branch that shows findings on the screen in Figure 7, and Figure 1O shows the predetermined line of the CT scan in Figure 9. Figure 11 is a plan view showing the first screen displayed on the bit map display when selected.
Fig. 12 is a flowchart showing the details of the finding creation work shown in step A plan view of a bitmap display showing the screen state for selecting findings for a position, Fig. 14~
Figure 25 is a plan view of the bitmap display showing a series of questions and options for their answers, and Figure 26 is a radiographic image diagnosis report created when findings are found. A plan view of a bit map display showing an example, Fig. 27 is a plan view of a bit map display showing an example of a radiographic diagnostic report created when no findings are found, and Fig. 28 shows the creation of this report. FIG. 11... Model display means, 12... Instruction means, 13... Question item storage area, 14... Question item calling means, 15...・Question display means, 1st...Answer input means, 17...Editing means, 24...
・cpu. 25...Hard disk, 26...R
A.M. 27...Bit map display, 29...
····mouse. Applicant: Fuji Xerox Co., Ltd. Representative
Patent Attorney Ume Yu Yamauchi 1st
Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 1 0 Figure 1 1 Figure 1 3 Figure j#1 4 Figure 15 Figure 1 6 Figure 7 Figure 18 Figure 1 9 Figure 20 Figure 2 1 Figure 22 Figure 23 'M24 Figure' M25 Figure 26 Figure 27 Figure 28

Claims (1)

[Scope of Claims] Model display means for displaying a human body model at a desired part including a part to be diagnosed; instruction means for indicating a part to be diagnosed in the model displayed by the model display means; and a diagnosis of the human body. a question item storage area storing question items necessary for creating a report; a question item calling means for selectively calling up a question item for a location specified by the instruction means from the question item storage area; and this question item calling unit. a question display means for displaying a question called up by the means; an answer input means for inputting an answer to the question displayed by the question display means; and an editing means for editing the answer input by the answer input means as a diagnostic report. A diagnostic report creation support device comprising: