JP2002032476A - Medical diagnosis support method and apparatus, and diagnosis support information recording medium - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To provide a medical diagnosis support device which enables an appropriate diagnosis without being affected by the experience of a doctor. SOLUTION: A storage means 10 storing a database K in which medical condition information on a medical condition of a patient obtained without analyzing image data and disease information on a disease corresponding to the medical condition information are stored, a diagnostic finding #k, Reading findings ＄ k,
A medical condition information input means 20 for inputting medical condition information J such as an inquiry result% k indicating a chief complaint and a history of the patient 9 and a pathological test result & k, and the medical condition information J corresponding to the medical condition information J when the medical condition information J is input. The medical diagnosis support apparatus 1 includes an output unit 30 that reads disease information such as a disease name and a representative image of the disease from the storage unit 10 and outputs the information to the image display unit 40.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a medical diagnosis support method and apparatus, and a computer-readable recording medium.

[0002]

2. Description of the Related Art It is well known that in medical diagnosis, it is not always easy to properly judge a disease name from a patient's condition or pathological test results.

[0003] For example, a patient complains of nausea or a heavy head (hereinafter referred to as the patient's chief complaint), a diagnosis result obtained by a doctor diagnosing a patient, or an image obtained by reading an X-ray image. The disease name corresponding to the interpretation result is not always one, but more than one. In other words, different diseases often have almost the same symptoms.

[0004] Including those with similar symptoms will further increase the number of suspected diseases.

In other words, when a patient is diagnosed, although it is known that the patient has an abnormality, it is difficult to accurately specify the disease name. Requires a lot of experience.

[0006]

On the other hand, from the viewpoint of the patient, it is inevitable that the doctor wants an appropriate treatment corresponding to his or her medical condition without depending on the experience of the doctor, and the doctor always responds to this demand. Failure to do so can lead to loss of trust.

On the other hand, even a physician with sufficient experience and skill is difficult to completely prevent oversight due to temporary forgetting. As mentioned above, the names of suspicious diseases vary widely, including those with similar symptoms, so at the time of diagnosis, if it is not possible to recall all candidate disease names corresponding to the patient's medical condition, The accuracy of diagnosis is also reduced.

[0008] The present invention has been made in view of the above circumstances, without being affected by the experience or forgetting of a doctor,
It is an object of the present invention to provide a medical diagnosis support method and apparatus that enables treatment suitable for a patient's condition, and a recording medium used in the method and apparatus.

[0009]

According to a first aspect of the present invention, a medical diagnosis support method is provided for correlating disease state information on a patient's condition obtained without analyzing image data with disease information on a disease corresponding to the disease state information. A prepared database is prepared in advance, and when the medical condition information is input, disease information corresponding to the medical condition information is read from the database,
The read disease information is output to a predetermined output device.

Reading the disease information corresponding to the disease information from the database means reading the disease information corresponding to the disease information from storage means such as a magnetic disk storing the database.

[0011] A second medical diagnosis support method of the present invention comprises a database which associates disease state information on a patient's condition obtained by analyzing image data with a representative image of a disease corresponding to the disease state information. Prepared in advance, and when the medical condition information is input, read a representative image of a disease corresponding to the medical condition information from a database, and output the read representative image to a predetermined output device. It is assumed that.

In obtaining the disease state information by analyzing the image data, it is preferable to use, for example, a computer-aided image diagnosis apparatus described in Japanese Patent Application Laid-Open No. 8-294479.

[0013] The first medical diagnosis support apparatus of the present invention stores a database in which disease state information on a patient's condition obtained without analyzing image data is associated with disease information on a disease corresponding to the disease state information. Storage means, medical condition information input means for inputting the medical condition information disease, and when medical condition information is input, reads disease information corresponding to the medical condition information from the storage means, and outputs the read disease information to a predetermined output device. Output means for outputting.

[0014] The second medical diagnosis support apparatus of the present invention is a database for associating medical condition information relating to a patient's medical condition obtained by analyzing image data with a typical image of a disease corresponding to the medical condition information. The stored storage means, the medical condition information input means for inputting the medical condition information, and when the medical condition information is input, a representative image of a disease corresponding to the medical condition information is read from the storage means, and the read representative image is displayed. Output means for outputting an image to a predetermined output device.

The first computer-readable first diagnosis support information recording medium of the present invention is a recording medium used in the first method and the first apparatus, and can be obtained without analyzing image data. It is characterized in that a database is stored in which disease state information relating to a disease state of a patient is associated with disease information relating to a disease corresponding to the disease state information.

The computer-readable second diagnosis support information recording medium of the present invention is a recording medium used in the second method and the second apparatus, and is a condition of a patient obtained by analyzing image data. A database in which medical condition information relating to the medical condition information is associated with a representative image of a disease corresponding to the medical condition information.

In the first medical diagnosis support method, apparatus, and diagnosis support information recording medium of the present invention, the medical condition information is displayed by a doctor or an interpreting doctor on an X-ray film image or an image displayed on a display monitor. At least one of the image interpretation result obtained by reading, the diagnosis result obtained by examining the patient, the interview result showing the patient's chief complaint and the history of the patient or his or her family, and the pathological test result desirable.

In the first medical diagnosis support method, apparatus and diagnosis support information recording medium of the present invention, the disease information is desirably at least one of a disease name and a representative image of the disease. .

[0019]

According to the first medical diagnosis support method, apparatus, and recording medium of the present invention, disease information relating to a patient's condition obtained without analyzing image data and a disease relating to a disease corresponding to the condition information are obtained. Since a database in which the information is associated is prepared in advance, and when the medical condition information is input, the disease information corresponding to the input medical condition information is read from the database and output to a predetermined output device. A doctor can use disease information such as a disease name and a representative image of the disease corresponding to the disease state information as an aid in diagnosis.

Further, the second medical diagnosis support method of the present invention,
According to the apparatus, and the recording medium, a database is prepared in advance that associates the disease state information on the patient's condition obtained by analyzing the image data with a representative image of a disease corresponding to the condition information, When the medical condition information is input, the representative image of the disease corresponding to the input medical condition information is read from the database and output to a predetermined output device. Image can be used as an aid in diagnosis.

Further, in the first medical diagnosis support method, apparatus and recording medium, if disease state information obtained by examining a patient or the like is input, there is a possibility (suspicious) disease name in relation to the disease state information. In any of the first and second medical diagnosis support methods, apparatuses, and recording media, for example, a suspected abnormal shadow part and a representative image of a disease can be compared and interpreted. Therefore, it is possible to prevent oversight regardless of whether or not there is experience, and it is possible to reduce the probability of being influenced by experience of diagnostic ability when diagnosing a patient, which is useful for improving diagnostic accuracy. In addition, diagnostic efficiency can be improved.

Further, it is not necessary to remember all the disease names or typical images of the diseases corresponding to the medical condition information, so that the burden on the doctor is reduced.

[0023]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the medical diagnosis support apparatus according to the present invention, and a diagram showing a first aspect of medical condition information input to the medical diagnosis support apparatus.

The medical diagnosis support apparatus 1 includes a storage unit 10 storing a database in which disease state information on a patient's condition obtained without analyzing image data and disease information on a disease corresponding to the disease state information are stored. , Patient 9
A medical condition information input means 20 for inputting medical condition information J obtained by examining or obtaining an image or performing a pathological examination, and when the medical condition information J is input, the input medical condition information J Output means 30 for reading out corresponding disease information from the storage means 10 and outputting the read disease information to the image display means 40 as an output device. The output means 30 incorporates a CPU (Central Processing Unit) which stores a program for performing processing described later.

The medical condition information input means 20 includes the medical condition information J
Any device may be used as long as the device 1 can acquire the same, for example, a device such as a keyboard for inputting numerical values or characters, or a pointing device such as a mouse may be used.

Incidentally, there are hospitals that create and store electronic medical records today. In such a hospital, the information on the created electronic medical record may be used as the medical condition information J as the medical condition information input means 20.

As the image display means 40, a CRT monitor, a liquid crystal panel or the like may be used.

The medical diagnosis support apparatus 1 is connected to command means 50 such as a keyboard or a mouse for inputting a command for controlling the processing in the apparatus 1.

When preparing the database, for example, a database in which the disease condition information is associated with disease information relating to the disease corresponding to the disease condition information is created by a database creation device 60 provided outside, and the created database is stored on a hard disk. The database may be stored in a magnetic recording device 11 such as a computer, or the created database may be recorded and stored in a computer-readable portable recording medium 12 such as a CD-ROM or MO (magneto-optical disk). You may make it. In the latter case, it goes without saying that the medical diagnosis support apparatus 1 is provided with a reading device 13 for reading the recording medium 12.

The medical condition information can be obtained by a doctor or a radiologist reading an X-ray film image or an image displayed on a display monitor or the like (image reading result), and can be obtained by a doctor examining a patient. At least one of a diagnostic finding (diagnosis result), an inquiry result showing the chief complaint of the patient, a medical history of the patient or family (history), and a pathological test result. In addition, other information that can be used for diagnosis, such as monitor information such as an electrocardiogram and a respiratory rate, may be used. Interpretation findings, diagnostic findings, interview results,
Hereinafter, information excluding any of the pathological test results is referred to as other information.

The disease information includes at least one of a disease name (disease name) and a representative image of the disease (hereinafter also referred to as a disease representative image).
The disease representative image only needs to be an image representing the characteristics of the disease, and is not limited to a radiation image, and may be an image showing an overview of the affected part. The number of disease representative images is not limited to one.
For example, when it is preferable to use one set of two images, two images may be set as one set.

In this embodiment, in order to obtain interpretation findings, CR (Computed Radiography) is used.
A radiation image (CR image) is obtained using the system 90. That is, as shown in FIG. 1, in an imaging device 91, an X-ray emitted from an X-ray tube 91a is irradiated toward a patient 9 as a subject, and an X-ray transmitted through the subject is transmitted to an imaging plate having a stimulable phosphor. The transmitted radiation image is made incident on the IP and stored and recorded as energy information. Then, the imaging plate IP on which the transmitted radiation image is accumulated and recorded is used as a CR device 92 as an image reading device.
To read the transmitted radiation image to obtain digital image data S carrying the transmitted radiation image. Next, the digital image data S is input to the image recording device 93, an image is recorded on the CR film 94 based on the digital image data S, and then the CR film is developed to output a hard copy on which the CR image is recorded. Get 95. The hard copy output 95 is similar to a conventional X-ray film. For example, the hard copy output 95 is composed of an image 95a having a gradation close to a screen / film X-ray image and an image 95b in which a predetermined spatial frequency component is emphasized. It may be output side by side on one CR film 94.

The digital image data S obtained by the reading device 92 is also input to an image diagnostic device 80 connected to a display monitor 81 and used for image diagnosis, and further is input to an image file device 97 and input to an optical disk 98. The data is recorded and stored.

Next, a first embodiment of the operation of the medical diagnosis support apparatus 1 having the above configuration will be described.

First, a method of creating a database will be described. 2 and 3 are diagrams for explaining an example of a method for creating a database.

In FIGS. 2 and 3, uppercase letters (hereinafter simply referred to as uppercase letters) are the name of the disease, and # followed by lowercase letters (hereinafter simply referred to as lowercase letters) and numbers of the disease name are diagnostics generally appearing in the disease. Findings, ＄ followed by lowercase letters and numbers in the disease name are commonly found in the disease, and% followed by lowercase letters and numbers in the disease name are the interview results commonly found in the disease, & followed by the disease name Lowercase letters and numbers indicate pathological test results that commonly appear in the disease,
And the lowercase letter and number following the disease name indicate other information that commonly appears for the disease.

Diagnostic findings (# + small letters and numbers in disease name),
Interpretation findings (＄ + lowercase letters and numbers in disease name), interview results (%
(Lower case and number of disease name), pathological test result (lower case and number of disease name), and other information (lower case and number of disease name) are disease state information corresponding to the disease name (upper case). #, ＄,%, &, ＠ indicate the type of medical condition information,
The number at the end of each medical condition information indicates a rough classification of the medical condition information. Thereby, even if the disease name is different,
When the numbers at the end of the same medical condition information are the same, it indicates that similar information (for example, similar diagnostic findings and pathological test results) can be obtained, and subtle differences in the medical condition information are distinguished by lowercase letters related to the disease name.

For example, irrespective of the disease name, 1 in the diagnostic finding # indicates swelling of the throat, 2 indicates fever, and 1 in the interpretation finding ＄ indicates a ground-glass shading pattern, 2 indicates a lumpy shading pattern, and 1 in% indicates numbness of the limbs, 2 indicates laxity, 1 in the pathological test result & indicates that the blood glucose level is higher than the standard value, 2 indicates hematuria, and the like.

Then, as shown in FIG. 2, a disease name (uppercase alphabetic character) as disease information, a diagnostic finding (# + lowercase and numeral of disease name), and a radiological finding (読 + lowercase and numeral of disease name) , Medical examination results (% + lowercase letters and numbers of disease name), pathological test results (& + lowercase letters and numbers of disease name), and other information (＠ + lowercase letters and numbers of disease name) Create data associated with information.

The image data (D + shown by uppercase letters of the disease name) carrying a disease representative image (shown by uppercase letters of the disease name), which is a representative image of the disease, also includes the disease state information and the disease information as one of the disease information. Correspond (link). This image data may be associated with a disease name.

For example, when preparing data on the disease A, the diagnosis findings # a1, # a2,
# A3,..., #An generally appear, and when a radiation image of a patient with the disease A is read, a radiological finding is made.
a1, ＄ a2, ＄ a3,..., ＄ an are generally obtained, and as a result of the interview,% a1,% a2,% a3,.
,% # An is generally obtained, & a1, & a2, & a3,..., & An are generally obtained as pathological test results, and other information is $ a1, $ a2,.
When ＠an is generally obtained, data in which the disease name A as the disease information is associated with the diagnostic findings, interpretation findings, interview results, pathological examination results, and other information as the disease state information, and the disease A The image data DA carrying the representative image AA is created.

In the same manner, each of the disease names B, C,..., Z as one of the disease information, and general medical conditions, radiological findings, interrogation results, pathological test results,
.., And DZ carrying representative images BB, CC,... Of each disease.

Such condition information and image data for each disease can be created using a medical dictionary or the like.

Depending on the type of disease, it is not always possible to obtain diagnostic findings, interpretation findings, interview results, pathological examination results, other information, and all information on typical images. Only those that can be obtained need to be converted into data.

Next, as shown in FIG. 3, focusing on common disease condition information (for example, the first symbol # a1 and the last numeral 1) of common disease, the common symbol and numeral For each condition information having the condition information, list data of disease names presenting the condition information is created.

For example, among the medical condition information (the ones with #) indicating the diagnostic findings, the medical condition information of the number 2 is grouped as # 2, and list data of the disease names exhibiting the diagnostic findings of # 2 is created. I do.

Similarly, information indicating the findings (marked with ＄), the results of the interview (marked with%), the results of the pathological examination (marked with &), and other information (marked with ＠) Regarding (1), list data of disease names exhibiting the same numerical condition information is created.

Thus, # 1 is a group of diseases in which swelling of the throat appears in the diagnostic findings, # 2 is a group of diseases exhibiting fever, # 1 is a group of diseases presenting ground glass, and ＄
2 is a group of diseases having a lumpy shape,% 1 is a group of diseases having numbness of limbs,% 2 is a group of diseases having numbness, & 1 is a group of diseases having high blood sugar level,
& 2 can be classified as a group of diseases presenting hematuria.

Since the name of the disease is not known at the first stage when the doctor examines the patient, the medical condition information J obtained by performing a medical examination and the like is information independent of the name of the disease. #N, radiological findings {n, interview result% n, pathological test results & n, and other information}
n.

For each group, an image of the most representative disease among the diseases included in the group (characters representing the group; hereinafter also referred to as disease state representative images) is extracted from the disease representative images. Then, the image data carrying the extracted disease representative image is assigned as disease state representative image data (D + group name). For example, # 1 is a group of diseases exhibiting a ground glass state. If a disease A should be first suspected as a disease presenting a ground glass state, image data DA carrying a disease representative image AA of the disease A is transferred to this group. May be assigned as the disease state representative image data D # 1.

Then, table data in which the list data is associated with the disease condition representative image data of each group of disease condition information is created. The table data is data for associating the disease state information with the disease name or a representative image of the disease, and the data amount is extremely small as compared with the raw data such as the disease state information and the image data.

The medical condition information and the image data of FIG. 2 and the table data of FIG. 3 created as described above are recorded on the magnetic recording device 11 as a database K or a CD.
-Recording on a portable recording medium 12 such as a ROM. At this time, in order to ensure the easiness of the processing described later, the disease state information and the image data are stored in the hierarchical level 2 deeper than the hierarchical level 1 of the table data as in the example of the data structure shown in FIG. Note that detailed information on a disease may be stored at a hierarchical level 3 deeper than the disease state information and the image data. When the database is used instead of a dictionary, the medical condition information and image data stored in the hierarchical level 2 or the detailed information stored in the hierarchical level 3 can be used.

Next, a method of utilizing the created database in diagnosing a patient will be described.

When the patient 9 comes to the consultation, the patient 9 is examined, and the diagnostic finding #k obtained by the consultation is
An inquiry result% k indicating the chief complaint of the patient 9 such as arm pain or the past history of the patient or family, and the other information ＠k are input to the medical condition information input means 20 as an example of the medical condition information J. . Note that k is an arbitrary positive integer and 1
Not necessarily one. The same applies hereinafter.

Further, the patient 9 is used by using the CR system 90.
The hard copy output 95 is obtained, and the hard copy output 95 is interpreted to obtain an interpretation finding ＄ k as one of the disease state information J such as a ground glass-like shadow pattern, a honeycomb-like shadow pattern, and a bumpy shadow pattern. An input is made to the medical condition information input means 20.

Further, a pathological test such as a blood test or a urine test is performed, and the pathological test result & k as one of the medical condition information J is obtained.
Is input to the medical condition information input means 20.

It should be noted that not all of these medical condition information is always obtained, so that only the information obtained when diagnosing the patient 9 needs to be input.

The medical condition information input means 20 outputs the diagnosis finding #k,
An output unit 30 that outputs the medical condition information J including the interpretation finding ＄ k, the interview result% k, the pathological examination result & k, and other information ＠k.
To enter.

The output means 30 outputs the input medical condition information J,
That is, diagnostic finding #k, interpretation finding ＄ k, interview result% k,
Based on the pathological test result & k and other information ＠k, and the processing method input from the command means 50, a disease name or an image of a disease suitable for the specified processing of the input various disease state information J is displayed. The image is read from the database (storage means 10) and the read disease name or disease image is displayed on the image display means 40.

FIGS. 5 to 7 are views showing examples of a disease name and a disease image display layout to be displayed on the image display means 40. FIG. An example of the processing method according to the first embodiment will be described with reference to these drawings.

In the first processing method having the display layout shown in FIG. 5, a disease name and an image corresponding to the medical condition information desired by a doctor who diagnoses the medical condition of the patient 9 in the input medical condition information J are determined. This is a mode of automatic display.

For example, when a patient is examined, medical condition information J
# 1 to # 5, $ 1 to $ 5,% 1 to% 5, & 1 to
When & 5, $ 1 to $ 5 are obtained, the doctor
When # 1, # 2, $ 2, $ 3,% 2,% 5, & 4, $ 1 are commanded, referring to table data stored in a shallow hierarchical level as shown in FIG. Medical condition information #
1, # 2, $ 2, $ 3,% 2,% 5, & 2, $ 1, a representative disease image presenting the disease information, that is, a disease representative image and the disease name (representative disease name) are displayed. At the same time, a list of disease names corresponding to the disease state information, that is, all disease names (candidate disease names) presenting the disease state information is displayed near the disease state representative image. It should be noted that the candidate disease names are selected and displayed in a predetermined priority order from the higher priority order by the number corresponding to the size of the display area, and all of them can be scroll-displayed when an instruction is received. It may be.

As a result, all the disease names that may exhibit the medical condition information J input by the doctor are listed for each medical condition information J, so that the doctor pays attention to the listed disease names and makes a diagnosis. You can do it. In other words, the doctor does not need to remember all the disease names corresponding to the medical conditions, and can also prevent beliefs. In addition, since a typical image corresponding to the disease name is also displayed, diagnosis (for example, comparative image reading) is performed with reference to this image, so that the diagnosis efficiency can be improved. As a result, an appropriate judgment can be made without being affected by the lack of experience or forgetting of the doctor, and the patient can be subjected to the most appropriate treatment for the medical condition.

Further, since a database in which the disease state information is associated with the disease name is prepared in advance, the burden of data analysis and sorting is reduced, and if the disease state information is input, the corresponding disease name or typical disease name is displayed. The image can be output immediately, and the diagnostic efficiency can be improved.

Here, for example, regarding the interpretation finding # 3,
In some cases, the doctor desires not the disease state representative image presenting the image finding # 3 but the image of another disease presenting the image finding # 3. In this case, by designating a favorite one of the disease names displayed in the list by using the command means 50, it is possible to switch and display the image of the designated disease.

For example, each specified medical condition information # 1, # 2,
$ 2, $ 3,% 5 & 2, & 4, the disease state representative image of each condition information automatically displayed based on the disease A
In the case of the image AA, there is a high possibility that the patient 9 has the disease A. In this case, the image CC of the disease C, which is the disease state representative image, is displayed with respect to the interpretation finding # 3, which is not preferable for diagnosis. In such a case, if the interpretation finding # 3 is switched to the image AA of the disease A, the diagnostic performance can be improved.

It is preferable that the disease name corresponding to the displayed image among the disease names displayed in the list is displayed so that it can be easily distinguished from other disease names. For example, a pop-up display such as a color change or an arrow finger may be performed, or a disease name may be superimposed on a display image.

Further, it is preferable to display such that it is possible to determine whether the displayed image is a disease state representative image of the disease state information or an image of a disease designated by a doctor. For example, a disease name corresponding to the disease state representative image and its name are displayed. It is preferable to display a pop-up display in a different form from the disease names other than the above.

In the embodiment shown in FIG. 5, when the number of specified medical condition information is large, if all the medical condition information are to be displayed side by side, the display area of each medical condition information becomes narrow, so that the image becomes small. If the multi-window display is used to secure a certain size, there is a possibility that the diagnostic performance may be reduced such that the number of images that can be viewed at a time is reduced.

The second processing method which has the display layout shown in FIG. 6 is an embodiment which solves the problem of the embodiment shown in FIG. 5, and in which the patient 9 out of the medical condition information J obtained by the medical condition information input means 20. All disease names (candidate disease names) that may present the disease condition information J desired by the doctor who diagnoses the disease condition
Is automatically displayed in the list display area, and then an image (medical condition representative image or favorite image) is displayed in the image display area based on the selection designation by the doctor. When displaying the candidate disease names for the first time, the representative disease name of each medical condition is displayed, or one medical condition representative according to a predetermined condition (which may be a reference by a third method described later). The image may be displayed in the image display area.

Also in this second method, since all the disease names which may exhibit the medical condition information J inputted by the doctor are displayed for each medical condition information J, they are influenced by the lack of experience and forgetting of the doctor. The patient can make an appropriate decision without taking any action, and can take the most appropriate treatment for the patient.

Since a large image display area can be secured, the performance of comparative image reading can be improved.

Next, the third layout having the display layout shown in FIG.
Will be described. This third processing method
Based on the combination of the disease state information J acquired by the disease state information input means 20, a highly probable disease name and a disease representative image which is a representative image of the disease are extracted, and the extracted disease name and disease representative image are automatically extracted. This is a mode of displaying the information.

Not only the disease name and the disease representative image but also either one may be displayed.

The disease name and the disease representative image having a high possibility are not limited to one having the highest possibility as shown in FIG.
As shown in FIG. 7 (B), several items may be displayed from the most likely ones. At this time, for example, as shown by a numeral in the figure, it is preferable to display such that the level of possibility is recognized. Alternatively, the image of the most likely disease may be displayed large, while the images of the remaining diseases may be displayed small.

As shown in FIG. 7 (C), not only the disease name and the disease representative image, but also the disease condition information of the disease (including those other than the disease condition information J obtained by the disease condition information input means 20) and the details. Information may also be displayed.

In any of the display modes (A) to (C), all the disease names which may exhibit the disease state information J obtained by the disease state information input means 20 or the disease state of the patient 9 are diagnosed. All disease names that may present the medical condition information J desired by the doctor may be displayed in the list display area.

Here, when extracting a disease name or a disease representative image having a high possibility based on the combination of the disease condition information J, the judgment level may be different depending on the disease or condition information. For example, a point map in which points indicating importance as shown in FIG. 8 are assigned to each disease or condition information is prepared, and points corresponding to the condition information J acquired by the condition information input means 20 are added for each disease. It is preferable to use a method of determining that the larger the sum is, the higher the possibility is. Note that a blank portion in the map indicates that the condition information does not appear for the disease or has very low importance.

When the patient is examined, the medical condition information J is
# 1 to # 5, $ 1 to $ 5,% 1 to% 5, & 1 to & 5, $
When 1 to $ 5 are obtained, only these portions are to be added, and in the point example shown in FIG. 8, since the addition result of the disease A is the largest, the disease having the highest possibility is the disease A. Suppose there is.

According to the mode using the third processing method, a disease name and a disease representative image having a high possibility (high suspicion) corresponding to the combination of the acquired medical condition information J are automatically displayed. The physician is less burdened with judgment and less affected by experience and forgetting.

Further, if not only the names of the diseases having a high possibility but also the names of all the diseases having the possibility of presenting the disease state information J are displayed, a considerable amount of attention can be paid to those diseases.

Next, a description will be given of a second embodiment of the operation of the medical diagnosis support apparatus having the above configuration.

FIG. 9 is a block diagram showing the configuration of an embodiment of the medical diagnosis support apparatus according to the present invention (same as that in FIG. 1) and a second aspect of the medical condition information input to the medical diagnosis support apparatus. FIG.

In the second embodiment, in order to acquire medical condition information, the image diagnostic apparatus 80 is replaced by a computer-assisted image diagnostic apparatus 82.
It is assumed that it is used in place of, and prepares in advance a database in which disease state information relating to a patient's disease state obtained by analyzing image data and a disease representative image corresponding to the disease state information are prepared in advance. When disease state information is obtained by analyzing image data with the computer-aided image diagnostic apparatus 82, a disease representative image corresponding to the disease state information is read from a database, and the read disease representative image is output to a predetermined output device. This is to output.

As will be described later, from the viewpoint of performing comparative image interpretation, the image display means 4 of the medical diagnosis support apparatus 1
It is preferable that the display monitor 81 of the computer-aided image diagnostic apparatus 82 also serves as 0.

The computer-aided image diagnostic apparatus 82 is connected to a display monitor 81 such as a CRT which has both functions of the whole image display means and the local image display means, as disclosed in, for example, Japanese Patent Application Laid-Open No. 8-294479. In addition, an abnormal shadow detection algorithm based on morphology processing detects shadow candidates of microcalcifications (microcalcification shadow candidates), and an abnormal shadow detection algorithm based on iris filter processing detects shadow candidates of tumor parts (candidate tumor shadows). An apparatus for detecting, including a whole image representing a whole radiographic image of a subject, a whole image with a label, and digital calcification shadow candidates detected based on digital image data (whole image data) representing the whole image, A predetermined image enhancement process is performed on an image (local image) of a local region including an abnormal shadow candidate such as a tumor shadow candidate. And the processed local image is a device that displays on the display monitor 81 in a predetermined layout.

The pathological condition information input means 20 is a feature quantity * 1 quantitatively representing a microcalcification shadow candidate and a feature quantitatively representing a tumor shadow candidate obtained by analyzing the image data S by the computer-aided image diagnostic apparatus 82. The amount * 2 is acquired as the medical condition information J.

The database K to be prepared in the second embodiment
Are image data D * 1 and D * 2 that carry a representative image (disease representative image) of a disease according to the level of the feature amount * 1, * 2 and each feature amount * 1, * 2. The difference from the first embodiment is that the data is associated. Note that image data carrying a representative image of a disease corresponding to a combination of the two feature amounts * 1 and * 2 as the disease state information J may be prepared.

Then, the table data in which the levels of the respective feature quantities * 1 and * 2 are associated with the image data D * 1 and D * 2 are located at a position shallower than the hierarchical level of the image data D * 1 and D * 2. Construct the stored database K.

On the other hand, at the time of image diagnosis, the image data S is analyzed by the computer-assisted image diagnostic apparatus 82, and the characteristic amount * 1 quantitatively representing the microcalcification shadow candidate and the characteristic amount * 2 quantitatively representing the tumor shadow candidate are determined. The medical condition information input means 20 obtains the information, and inputs the characteristic quantities * 1 and * 2 to the output means 30.

The output means 30 receives the input characteristic quantity * 1,
The image data corresponding to each level of * 2 is read from the database, and a disease representative image based on the read image data is displayed on the image display means 40.

The diagnostician makes a diagnosis by comparing the disease representative image displayed on the display monitor 81 of the computer-aided image diagnostic apparatus 82 with the image displayed on the image display means 40 of the medical diagnostic assistance apparatus 1. Therefore, the diagnostic performance can be improved.

The image display means 4 of the medical diagnosis support apparatus 1
Since the disease representative image displayed as 0 can be used as an aid in the determination, the degree of experience on the determination result can be reduced. In other words, even an inexperienced radiologist can make a decision close to that of an experienced radiologist.

Although the preferred embodiments of the medical diagnosis support method and apparatus and the diagnosis support information recording medium of the present invention have been described above, the present invention is not necessarily limited to the above-described embodiments.

For example, in the first embodiment, as the disease information to be stored in the database, two names, ie, the name of the disease and the representative image of the disease, are prepared, but only one of these is used. You may. Again, in this case,
When the medical condition information is input, the doctor reads the disease name or the disease representative image corresponding to the input medical condition information from the database and outputs the image. The representative image can be used as an aid in diagnosis. This allows
This makes it possible to prevent oversight, reduce the probability of being affected by the experience of diagnostic ability when diagnosing a patient, and improve diagnostic accuracy. Further, it is not necessary to memorize all the disease names or representative images of the disease corresponding to the medical condition information, so that the burden on the doctor is reduced.

Further, in the above embodiment, the point map is used when extracting the disease name and the disease representative image which are highly likely to correspond to the disease state information J. A database in which the disease names and the disease representative images are sequentially associated is constructed (see FIG. 10), and when the disease condition information J is input, the disease names and the disease representative images up to the order specified by the operator are output. You may do so. When the order is not specified, all disease names and disease representative images corresponding to the disease state information J are output, or
It is preferable to output up to a rank set as a standard value in advance.

By linking the database with the electronic medical record, an image of a disease acquired in the past in a hospital can be used as a disease representative image. Hereinafter, this embodiment will be described with reference to FIG.

A database is established in which disease names and disease representative images are associated in ascending order of probability for each medical condition information J. For example, in the case of a ground-glass shading pattern, the possibility is high in the order of disease names A, B, and C, and in the case of a honeycomb shading pattern, the possibility is high in the order of disease names α, β, and γ.
And so on.

An electronic medical record is prepared in which a disease name and a disease image of the patient obtained by diagnosing the patient are recorded.
Note that only the image number corresponding to the disease name may be recorded in the electronic medical record, and the image itself may be stored in the image filing apparatus 100 in association with the electronic medical record number or the patient number.

After diagnosing the patient, the doctor who is the operator inputs the medical condition information J. For example, a radiographic image of the patient is interpreted to determine a shadow pattern, and the determined shadow pattern is input to the medical diagnosis support apparatus as medical condition information J. Then, for example, when information such as a ground-glass shading pattern is input and up to the third priority is designated, the medical diagnosis support apparatus extracts disease names A, B, and C from the database as candidate diseases.

Next, each extracted disease name A,
An electronic medical record having the same disease name as B and C is extracted. Next, a disease image corresponding to the extracted electronic medical record is extracted from the image filing apparatus 100. Then, the extracted disease names and images are displayed in descending order of possibility.

When there are many patients with the same disease name, a large number of disease images can be extracted for each disease name.
In this case, all the extracted images may be displayed as images corresponding to the disease, but according to a predetermined condition, only images of a patient most suitable as the disease may be displayed. Good.

In the latter case, the image of the patient most suitable as the disease is registered in advance in the database as a disease representative image, so that the disease name and disease representative image corresponding to the input disease state information J can be stored in the database. Since the extracted disease name and disease representative image can be output in the order of probability, it is not necessary to link the database with the electronic medical record.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an embodiment of a medical diagnosis support apparatus according to the present invention, and a diagram illustrating a first aspect of medical condition information input to the medical diagnosis support apparatus.

FIG. 2 is a diagram for explaining an example of a database creation method

FIG. 3 is a diagram for explaining an example of a database creation method;

FIG. 4 shows an example of a data structure in a database.

FIG. 5 is a diagram illustrating an example of a display layout of a disease name and a disease image;

FIG. 6 is a diagram illustrating an example of a display layout of a disease name and a disease image.

FIG. 7 is a diagram showing an example of a display layout of a disease name and a disease image;

FIG. 8 is a diagram showing a point map used when extracting a disease name or a disease representative image having a high possibility based on a combination of medical condition information.

FIG. 9 is a block diagram illustrating a configuration of an embodiment of a medical diagnosis support apparatus according to the present invention, and a diagram illustrating a second aspect of medical condition information input to the medical diagnosis support apparatus.

FIG. 10 is a view for explaining another embodiment of the medical diagnosis support method of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Medical diagnosis support apparatus 9 Patient 10 Storage means 11 Magnetic recording device 12 Recording medium 13 Reading device 20 Medical condition information input means 30 Output means 40 Image display means 50 Command means 60 Database creation apparatus 80 Image diagnosis apparatus 81 Display monitor 82 Computer support image Diagnostic device 90 CR system 91 Imaging device 91a X-ray tube 92 CR device 93 Image recording device 94 CR film 95 Hard copy output 97 Image file device 98 Optical disk IP Imaging plate

Claims (12)

[Claims] 1. A database in which disease state information on a patient's condition obtained without analyzing image data and disease information on a disease corresponding to the disease state information is prepared in advance, and the disease state information is input. And reading the disease information corresponding to the medical condition information from the database, and outputting the read disease information to a predetermined output device. 2. The medical diagnosis support method according to claim 1, wherein the medical condition information is at least one of an image interpretation result, a diagnosis result, an inquiry result, and a pathological examination result. 3. The medical diagnosis support method according to claim 1, wherein the disease information is at least one of a disease name and a representative image of the disease. 4. A database in which medical condition information on a medical condition of a patient obtained by analyzing image data and a typical image of a disease corresponding to the medical condition information are prepared in advance, and the medical condition information is stored in a database. When input, a representative image of a disease corresponding to the medical condition information is read from the database, and the read representative image is output to a predetermined output device. 5. A storage unit storing a database in which medical condition information on a medical condition of a patient obtained without analyzing image data and disease information on a disease corresponding to the medical condition information are stored, and the medical condition information is input. Disease state information input means; and when the disease state information is input, read out disease information corresponding to the disease state information from the storage means, and output means for outputting the read disease information to a predetermined output device. A medical diagnosis support device characterized by the above-mentioned. 6. The medical diagnosis support apparatus according to claim 5, wherein the medical condition information is at least one of an image interpretation result, a diagnosis result, an inquiry result, and a pathological examination result. 7. The medical diagnosis support apparatus according to claim 5, wherein the disease information is at least one of a disease name and a representative image of the disease. 8. A storage unit storing a database in which medical condition information on a medical condition of a patient obtained by analyzing image data and a typical image of a disease corresponding to the medical condition information are stored, and Medical condition information input means for inputting, when the medical condition information is input, a representative image of a disease corresponding to the medical condition information is read from the storage means, and the read representative image is output to a predetermined output device. A medical diagnosis support device, comprising: 9. A computer-readable database storing a database in which medical condition information relating to a medical condition of a patient obtained without analyzing image data and disease information relating to a disease corresponding to the medical condition information are recorded. Possible diagnostic support information recording medium. 10. The computer-readable diagnosis support information record according to claim 9, wherein the medical condition information is at least one of an image interpretation result, a diagnosis result, an inquiry result, and a pathological examination result. Medium. 11. The computer-readable diagnosis support information recording medium according to claim 9, wherein the disease information is at least one of a disease name and a representative image of the disease. 12. A database in which medical condition information relating to a medical condition of a patient obtained by analyzing image data and a typical image of a disease corresponding to the medical condition information are recorded. Computer-readable diagnostic support information recording medium.