JP2002200048A - Diagnosis support system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a diagnosis support system which can automatically analyze medical images of a patient, which are judged 'normal', even an operator does not operate the diagnosis support system without omission. SOLUTION: A doctor makes the final diagnosis on a diagnosis report image 41. Data relating to 'normal' or 'abnormal' are recorded in recording media 12, 88 based on the reported result. Only data recorded 'normal' are extracted from recording media 12, 88 and an automatic support process automatically analyze the medical image data. When the result of the diagnosis support process is 'abnormal', the result is informed to the doctor as an alarm image 43. The doctor confirms the informed medical image again.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a diagnosis support system for analyzing a medical image by a computer and assisting a doctor in making a decision, and more particularly to a diagnosis support system having an improved method for starting a diagnosis support process.

[0002]

2. Description of the Related Art Conventionally, shadows of medical images captured by a CT apparatus, an MRI apparatus, an ultrasonic apparatus, etc. are analyzed using a computer, and lesion candidates are narrowed down from the shadows and presented to a doctor. Diagnosis support for seeking judgment (CA
D: Computer Aided Detectio
n) The system is being put into practical use. The following has been proposed as an operation mode when this diagnostic support system is used for actual medical examination. At the time of normal lung cancer screening, first, two doctors perform independent interpretations in parallel, then bring the interpretation results, and both doctors finally make a comprehensive judgment of whether there is an abnormality or no abnormality . If the result of the comprehensive judgment is abnormal, a detailed examination is performed on the patient. On the other hand, if the result of the comprehensive judgment is no abnormality, the doctor has started the diagnosis support system by himself and analyzed the presence or absence of the abnormality. When the result of the analysis by the diagnosis support system indicates that there is an abnormality, the doctor performs the image interpretation for confirmation again.

[0003]

When the doctor comprehensively judges that there is no abnormality, the doctor has to start the diagnosis support system by performing operations such as key input by himself. For example, if both doctors make a comprehensive judgment and find that there is no abnormality, the diagnosis support system is first entered by inputting the ID number of the patient who is found to have no abnormality into the diagnosis support system using an operator such as a keyboard. Was analyzing the patient's medical image. That is, unless the patient's ID number or the like is input, the diagnosis support system does not analyze the patient. Therefore, when one of the two doctors is to input the patient's ID number or the like to the diagnosis support system, or based on a memo or the like describing the result of the comprehensive judgment by the two doctors, For example, when three persons are to enter, the patient's ID number or the like may not be inputted or an incorrect one may be inputted due to some mistake. In such a case, there is a problem that the diagnosis support system is not activated for a patient who has not been input, and does not function effectively despite the introduction of the diagnosis support system.

[0004] An object of the present invention is to provide a diagnosis support system capable of automatically performing analysis processing on a medical image of a patient determined to have no abnormality without operating the diagnosis support system one by one. Is to do.

[0005]

According to a first aspect of the present invention, there is provided a diagnostic support system comprising: data recording means for adding data relating to abnormality or no abnormality to medical image data diagnosed by a doctor and recording the data on a recording medium; Extracting means for extracting, from the medical image data recorded on the recording medium, data to which data relating to no abnormality has been added; and automatically executing a diagnosis support program on the medical image data extracted by the extracting means. It comprises an analysis means for starting and performing analysis, and a notifying means for notifying a doctor of the analysis result in accordance with the analysis result of the analysis means.

[0006] The final diagnosis made by the doctor is reported and recorded. The data recording means records, on the recording medium, data on the presence or absence of the abnormality based on the report. The extracting means extracts only those recorded as having no abnormality from the recording medium, and causes the analyzing means to automatically analyze the extracted medical image data. If the result of the analysis by the analysis means is abnormal, the result is transmitted to the doctor, so that the doctor can reconfirm based on the medical image. Therefore, if there is a mistake in recording the report at the stage of report creation, or if there is a mistake
Since this is automatically notified to the doctor, the diagnosis support system can be effectively used without bothering the doctor.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a diagnosis support system according to the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram showing a hardware configuration of an entire diagnosis support system to which the present invention is applied. This diagnosis support system is based on a plurality of tomographic images collected for a target site of a subject by a medical image modality such as an X-ray CT apparatus, an MRI apparatus, and an ultrasonic apparatus, and performs image processing for extracting and narrowing down a lesion candidate shadow or the like. , And lesion candidate shadows, images during extraction, and the like are displayed. This diagnostic support system is disclosed in Japanese Patent Application No.
Since it is described in JP-A-00-314550, a detailed description thereof will be omitted.

The diagnosis support system includes a central processing unit (CPU) 10 for controlling the operation of each component, a main memory 11 storing a control program for the diagnosis support system, a plurality of tomographic image data and operation data for each patient. A magnetic disk 12 storing a program and the like, a display memory 13 for temporarily storing image data for display, a CRT display 14 as a display device for displaying an image based on the image data from the display memory 13, a screen, It is composed of a mouse 15 and a mouse controller 16 for operating the above software switches, a keyboard 17 having keys and switches for setting various parameters, a speaker 18, and a common bus 19 for connecting the above-mentioned components.

In this embodiment, only the magnetic disk 12 is connected as a storage device other than the main memory 11, but in addition, a floppy (registered trademark) disk drive, hard disk drive, CD-ROM drive, magneto-optical A disk (MO) drive, ZIP drive, PD drive, DVD drive, or the like may be connected. Furthermore, it can be connected to various communication networks such as a LAN (local area network), the Internet, and a telephone line via a communication interface (not shown) so that image data can be exchanged with other computers and databases. It may be.

Hereinafter, an example of operation of the diagnosis support system of FIG. 1 will be described with reference to the drawings. FIG. 2 is a diagram illustrating a main flow executed by the diagnosis support system. CP in FIG.
U10 operates according to this main flow. FIG.
FIG. 3A is a diagram showing how data on a medical image is processed according to this main flow. FIG. 3A shows data on a medical image before a doctor's diagnosis.
FIG. 3B shows data on a medical image after a doctor's diagnosis, and FIG. 3C shows data after a doctor's diagnosis and an analysis result by the diagnosis support system. FIG. 4 is a diagram showing an example of a display screen displayed on the display.
4A shows a screen at the time of report creation, and FIG. 4B shows a display screen at the time of warning. This main flow is a series of steps performed when two doctors finally make a comprehensive judgment of whether there is an abnormality or no abnormality with respect to the interpretation result after the two doctors independently perform the interpretation in parallel. Processing. Less than,
Details of the main flow will be described in the order of steps.

[Step S20] A patient ID input screen is displayed on the display 14, and input of a patient ID number is waited. When the ID number is input, a medical image as shown in FIG. 3A corresponding to the ID number of the patient to be diagnosed is read from the magnetic disk 12 from among the medical images taken in advance by the medical image modality, and the display 14 is displayed. To be displayed. The medical image shown in FIG. 3A is composed of a header information section in which various information related to the patient is recorded and a plurality of n pieces of medical image data 1 to n. The supported flag CAD shown in the header information section is a flag that becomes “1” when the analysis is performed by the diagnosis support system. FIG. 3 (A)
In the case of FIG. 3B, since the analysis by the diagnosis support system has not been performed yet, the supported flag CAD is “0”.

[Step S21] The doctor displays 1
Then, while displaying the medical images in sequence, a final comprehensive determination of whether there is an abnormality or not is made while checking the result of the interpretation performed by the user. [Step S22] When the comprehensive judgment in step S21 is completed, a diagnostic report screen 41 as shown in FIG. 4A is displayed on the display 14 so that the doctor can prepare the judgment result as a diagnostic report. The doctor sees Fig. 4 (A)
Enter the diagnosis result on the display screen. The diagnosis result may be input by using the keyboard 17 to input detailed contents of the diagnosis as text data, or an item created in advance may be checked. Among the diagnostic reports, the most important ones are the item "no abnormality is recognized" and the item "recognize the above abnormality". If one of these items is not checked, the next step will not proceed even if the end button 42 is operated. In FIG. 4A, the item "No abnormality is recognized" is checked, and the inner circle of the double circle on the left side is shown in black. The diagnostic report created by the doctor is recorded as additional information of the medical image data, as shown in FIG. The additional information is accompanied by detailed information such as detailed text data created by the doctor and information indicating "no abnormality is recognized" or "recognize abnormality".

[Step S23] The process of constantly monitoring the doctor's diagnosis results looks at the diagnostic report as shown in FIG. 3B created in step S22, and the overall judgment is "abnormal. It is determined whether or not "no", and if "no abnormality" (yes), the process proceeds to the next step S24, and if "no abnormality" (no), the process returns and the next diagnostic report is created. Monitor until done. [Step S24] Since it is determined in step S23 that the comprehensive determination of the diagnostic report is "No abnormality detected",
Based on the medical image data to which the diagnostic report is added, an image diagnosis support process by a computer described in Japanese Patent Application No. 2000-314550 is started. It goes without saying that the image diagnosis support processing may be performed using another diagnosis support apparatus.

[Step S25] It is determined whether or not the result of the image diagnosis support processing in step S24 is abnormal. If there is an abnormality (yes), the process proceeds to step S26. If there is no abnormality (no), the process proceeds to step S27. Proceed to. [Step S26] Since the result of the image diagnosis support processing by the computer is abnormal, a warning window 43 as shown in FIG. 4B is displayed on the display 14. This warning window 43 means that a portion considered abnormal is found as a result of processing by the image diagnosis support system. When the doctor selects (clicks) the warning window 43, a window showing a detection result in which a lesion candidate shadow is shown by a marker as shown in FIG. 5 and an enlarged window showing the marker portion in an enlarged manner are displayed on the display 14. Are displayed at the same time.
The enlarged window is an enlarged display of the marker portion in the detection result window at a predetermined magnification. By enlarging and displaying in this manner, the doctor can more accurately observe the lesion shadow candidate.

The screen shown in FIG. 5 is a standard screen, and various display modes can be selected by the display mode selection buttons arranged in a line at the left end of the screen. The display mode selection button has seven types of modes: a standard display mode, a horizontal display mode, a vertical display mode, a distance display mode, a spiral display mode, and a registration display mode. FIG. 5 shows a standard mode in which images are displayed in the order in which the focus candidate shadows have been extracted, and the images are displayed in the order of the extracted images a to d as shown in FIG. Switching between the images a to d is performed by a screen switching button of an upper and lower triangle at the lower right of FIG. The screen switching button displaying a downward triangle is a next screen display button, and the screen switching button displaying an upward triangle is a previous screen display button. Note that since there are two lesion shadow candidates in the image c, a marker switching button is displayed in such a case. In addition, a button that allows the display magnification of the enlarged screen to be appropriately changed is also displayed, but is omitted here. Descriptions of the horizontal order display mode, the vertical order display mode, the distance order display mode, the spiral order display mode, and the registration order display mode other than the standard mode will be omitted.

[Step S27] Since the result of the diagnosis by the diagnosis support system is normal, here, in order to indicate that the diagnosis by the diagnosis support processing has been completed, FIG.
As shown in (1), "1" is set to the supported flag CAD in the header information section of the medical image data, and the routine returns. By setting “1” to the supported flag CAD, even if information on the medical image data is stored in a database described later, the supported flag CAD is set.
Thus, it is possible to recognize that the analysis processing by the diagnosis support system has been completed for the medical image data, and the analysis processing by the diagnosis support system is not performed twice. If the doctor overlooks despite the warning display in step S26, since the supported flag CAD is not set to "1", the analysis is again performed by the diagnosis support system and the result is Will be reported to the doctor later.

FIG. 7 is a diagram showing a schematic configuration of another embodiment of the diagnosis support system according to the present invention. This shows a state where medical image modalities such as an X-ray CT apparatus 71, an MRI apparatus 72, and an ultrasonic apparatus 73, a report creation apparatus 74, and a diagnosis support apparatus 75 are connected via a network 70. Medical image information output from each medical image modality is sequentially stored in the database 76. The image information stored in the database 76 is as shown in FIG. 3A, and the supported flag CA
D.

The report creation device 74 and the diagnosis support device 7
The configuration of 5 is basically the same as that of the diagnosis support system of FIG. 1, and is entirely controlled by a central processing unit (CPU). The difference between the diagnosis support system of FIG. 1 and the report creation device 74 and the diagnosis support device 75 is that the diagnosis support system of FIG. 1 performs both the report creation process and the diagnosis support process, whereas the report creation device 74 Only the report creation processing is performed, and the diagnosis support device 7
5 is that only the diagnosis support processing is performed. Therefore, in the diagnosis support system according to this embodiment, the diagnosis report created by the report creation device 74 is sequentially stored in the database 76 together with the image data. Then, the diagnosis support apparatus 75 sequentially extracts, from the database 76, images for which the overall judgment of the diagnosis report is “No abnormality is recognized”, executes a diagnosis support process on the extracted images, and In this case, a warning is displayed to the doctor. In some cases, a device equipped with a diagnosis support system as shown in FIG.

FIG. 8 is a flowchart for explaining an operation example of the network-connected diagnosis support system of FIG. Hereinafter, details of this flowchart will be described in the order of steps. Note that, in FIG.
The flowchart consisting of steps S82 to S82 is executed by the report creation device 74,
The flowchart consisting of 90 to step S96 is executed by the diagnosis support device 75.

[Step S80] Step S20 in FIG.
ほ ぼ Executes substantially the same processing as in step S22, and a doctor creates a diagnostic report. That is, since a patient ID input screen is displayed on the display 14, the doctor inputs the patient ID number. When the ID number is entered,
An image as shown in FIG. 3A corresponding to the ID number of the patient to be diagnosed is read from the database 76, stored in the storage device of the report generating device 74, and displayed on the display 14. The doctor makes a final overall judgment of whether there is an abnormality or no abnormality while sequentially displaying the images displayed on the display 14 and checking the result of the interpretation performed by the doctor. Then, the doctor inputs the diagnosis result according to the diagnosis report screen 41 as shown in FIG. 4A and creates a diagnosis report. The format of the diagnosis report is the same as that of FIG. The diagnostic report created by the doctor is recorded as additional information of the image data, as shown in FIG. The additional information is accompanied by detailed information such as detailed text data created by a doctor, and information of "No abnormality is recognized" or "No abnormality is recognized".

[Step S81] Information on a diagnosis report, which is a diagnosis result of a doctor, is stored in the database 88 as additional information of the corresponding image data. [Step S82] It is determined whether or not the diagnosis by the doctor has been completed. If it is determined that the diagnosis has not been completed (no), the process returns to step S80 to make a diagnosis on the next patient and determine that the diagnosis has been completed (yes). If so, the process ends.

[Step S90] The header information section of each image data stored in the database 88 is read. [Step S91] Analyzed flag CAD of header information section
Is determined to be “0”. If “0” (yes), the process proceeds to the next step S92, and if “1” (no), the process returns. That is, when the analyzed flag CAD is “1”, it is not necessary to perform the analysis. The order in which the header information is read from the database 88 is arbitrary, but the reading process may be performed for the past one week based on the date of creation of the diagnostic report.

[Step S92] Since the analysis flag CAD is determined to be "0" in the previous step S91, FIG.
As shown in (B), the diagnostic report added to the image data is examined, and it is determined whether or not the comprehensive determination is “no abnormality is recognized”, and if “no abnormality is recognized” (yes) Proceeds to the next step S93, and “accepts an abnormality”
In the case of (no), the process jumps to step S96. [Step S93] Since it is determined in step S92 that the comprehensive determination of the diagnostic report is “no abnormality recognized”,
Based on the image data to which the diagnostic report has been added, the image diagnosis support processing is started as described above.

[Step S94] It is determined whether or not the result of the image diagnosis support processing in step S93 is abnormal. If there is an abnormality (yes), the process proceeds to step S95, and if there is no abnormality (no), the process proceeds to step S96. Proceed to. [Step S95] Since the result of the image diagnosis support processing is abnormal, as in step S26 described above, FIG.
A warning window 43 as shown in (B) is displayed on the display 14 of the report creation device 74. Thereafter, when the doctor selects (clicks) the warning window 43, the window showing the detection result in which the lesion candidate shadow is shown by a marker as shown in FIG. The enlarged window shown is displayed at the same time. If the report creation device 74 is not in the activated state or the confirmation by the doctor is not performed in spite of the warning display, the supported flag CAD remains “0” and is stored in the database 88. Then, the header information section is read again from the database 88, and the above-described series of processing and analysis by the diagnosis support device 75 are performed, and the result is later reported to the doctor.

[Step S96] Since the result of the diagnosis by the diagnosis support system is normal, in order to indicate that the diagnosis by the diagnosis support processing has been completed, FIG.
As shown in (1), "1" is set to the supported flag CAD in the header information section of the medical image data, and the routine returns. By setting "1" to the supported flag CAD in this manner, the database 88 does not perform the diagnostic support processing on this image data.

In the above-described embodiment, the case of adding data indicating that there is no abnormality or the presence of an abnormality has been described.
If there is no abnormality, the data may not be added. In this case, extract the data without abnormal data,
The diagnosis support processing may be executed in response thereto. Instead of adding data, the data may be recorded in a flag format.

[0027]

As described above, according to the diagnosis support system of the present invention, a medical image of a patient determined to have no abnormality is automatically analyzed even if the operator does not operate the diagnosis support system one by one. There is an effect that processing can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a hardware configuration of an entire diagnosis support system to which the present invention is applied;

FIG. 2 is a diagram showing a main flow executed by the diagnosis support system of FIG. 1;

FIG. 3 is a diagram showing how data relating to a medical image is processed according to the main flow of FIG. 2;

FIG. 4 is a view showing an example of a display screen displayed on a display, FIG. 4 (A) is a screen at the time of report creation, FIG.
(B) is a display screen at the time of warning

FIG. 5 is an example of a screen displayed when the warning window of FIG. 4 is selected, showing a standard mode in which images in which a focus candidate shadow extraction process has been performed are displayed in order.

FIG. 6 is a diagram showing an example of a standard mode in which extracted images a to d are displayed in order.

FIG. 7 is a diagram showing a schematic configuration of another embodiment of the diagnosis support system according to the present invention;

8 is a flowchart for explaining an operation example of the diagnosis support system connected to the network of FIG. 7;

[Explanation of symbols]

10 central processing unit (CPU), 11 main memory, 12
... magnetic disk, 13 ... display memory, 14 ... CRT display, 15 ... mouse, 16 ... controller, 17 ...
Keyboard, 18 speaker, 19 common bus, 41 ...
Diagnostic report screen, 42 ... end button, 43 ... warning window, 70 ... network, 71 ... X-ray CT apparatus, 7
2: MRI apparatus, 73: Ultrasound apparatus, 74: Report creation apparatus, 75: Diagnosis support apparatus, 76: Database

Claims (1)

[Claims] 1. A data recording means for adding data relating to abnormality or no abnormality to medical image data diagnosed by a physician and recording the data on a recording medium, and from among the medical image data recorded on the recording medium Extracting means for extracting data to which data relating to no abnormality is added; analyzing means for automatically activating a diagnosis support program for medical image data extracted by the extracting means and performing analysis; and A diagnosis support system comprising: a notifying unit that notifies a doctor of the analysis result according to the analysis result.