JPH04314177A - Routine reading system - Google Patents

Abstract

PURPOSE:To improve reading efficiency by changing output route according to an image when the image stored in every kind of storage medium is fetched and is displayed on a work station. CONSTITUTION:The image to be stored in every kind of storage medium is collated with criterion set in advance, and frequency is decided, and for example, the image with the highest frequency is stored in high-speed memory (for example, image memory 5), the image with the frequency in the second place in low-speed memory (for example, hard disk 6), and other images in a data base 3. The criterion is set in such a way that learning or estimation is performed corresponding to, for example, the past image use experience of a user, or it is set so as to be changed appropriately corresponding to the request of the user. The fetch of the image is performed in sequence of the image with higher frequency based on the control of a CPU 8.

Description

[Detailed description of the invention]

[Purpose of the invention]

0001

[Industrial Application Field] The present invention is a routine interpretation method that stores medical images retrieved from a database via a network into high-speed memory and low-speed memory of a workstation, and displays desired medical images on the workstation as needed. Regarding the system.

0002

[Background Art] Medical images (hereinafter simply referred to as images) taken with various medical devices are stored in a database via a network, and when necessary, desired images are retrieved and displayed on a workstation for interpretation. A routine image interpretation system based on this method is known. FIG. 4 shows an outline of such a routine image interpretation system. 1 is an image capturing device, which consists of various medical equipment, and the images captured by this device are sequentially transferred to L.
It is transmitted via a network 2, such as an AN (Local Area Network), to a database 3 and stored there. The database 3 is equipped with a large-capacity storage device and can store all of the captured images.

Reference numeral 4 designates a workstation that includes an image memory 5 made of high-speed memory such as a semiconductor memory, a hard disk 6 made of low-speed memory such as a magnetic disk or an optical disk, and a display 7 for displaying images for interpretation.
, CPU (Central Processing Unit) 8, which controls the entire control operation.
It is composed of etc. Images retrieved from the database 3 can be stored in the image memory 5 and the hard disk 6.

[0004] With this configuration, in order for a user such as an interpreting doctor to observe images on the workstation 4, the patient I
When a specific image is designated by inputting D, the time required from when the designated image is retrieved to when it is displayed differs depending on which storage medium the designated image is stored in. Image memory 5 is a high-speed memory that is displayed in the shortest time.
The second case is when the image is stored on the hard disk 6, which is a slow memory, and the one that is displayed for the longest time is the one connected to the workstation 4 via the network 2. This is a case where the image is stored in the database 3 that is being used. Conventionally, images to be stored in each of these storage media have not been particularly distinguished.

[0005]

[Problems to be Solved by the Invention] However, in conventional routine radiology interpretation systems, images are stored in each storage medium without any reference, so when displaying images on a workstation, the three extraction routes are the same. Since this happens with probability, there is a problem in that even images that you want to display quickly can take a very long time. For example, in the case of images that are frequently used, even though we would like to improve the efficiency of image interpretation by shortening the time it takes to display them, there are cases where images that are stored in the database are retrieved and displayed, resulting in a reduction in the efficiency of image interpretation. begins to decline.

The present invention has been made in response to the above-mentioned problems, and provides a routine image interpretation system in which frequently used images are stored in a storage medium with a short retrieval time, giving priority to other images. The purpose is to [Structure of the invention]

[0007]

[Means for Solving the Problems] In order to achieve the above object, the present invention stores medical images retrieved from a database via a network in a high-speed memory and a low-speed memory of a workstation, and stores them in a desired medical image as necessary. In a routine image interpretation system for displaying images on a workstation, means for presetting and storing judgment criteria for medical images retrieved from a database, and judgments read from the storage means for determining the frequency of retrieved medical images. a control means for automatically storing the most frequently occurring medical images in the high speed memory, the next most frequently occurring medical images in the low speed memory, and the other medical images in the database. and a display control means for extracting and displaying the corresponding image from a storage location in response to a user's image display request.

[0008]

[Operation] The frequency of images to be stored in each storage medium is determined by comparing them with preset criteria, and accordingly, the most frequent image is stored in the high-speed memory, and the next most frequent image is stored in the high-speed memory. Each other image is automatically stored in a database in slow memory. Further, the determination criteria can be learned and estimated based on, for example, the user's past image usage results. Alternatively, the criteria can be changed arbitrarily according to the user's request. This reduces the time it takes to retrieve the most frequently used images from the storage medium and display them on the workstation.
Image interpretation efficiency can be improved.

[0009]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Examples of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a first embodiment of the routine image interpretation system of the present invention. Reference numeral 1 denotes an image capturing device, which is comprised of various medical devices such as an X-ray CT device, an MRI device, and an ultrasound diagnostic device. , images taken by these are sent to a database 3 via a network 2 such as a LAN.
is transmitted and stored. This database 3 is equipped with a large-capacity storage device, and can store all of the captured images.

Reference numeral 4 is a workstation having an image memory 5 consisting of a high-speed memory such as a semiconductor memory, a hard disk 6 consisting of a low-speed memory such as a magnetic disk or an optical disk, and a plurality of displays 7 for displaying images for interpretation. ing. The plurality of displays 7 are MDS (
It is connected to the image memory 5 and hard disk 6 via a multi-display system) 9. 10 is a bus for image data, and 11 is a data bus for control signals. Further, the plurality of displays 7 are provided with an input device such as a touch panel 12. 13 is a communication interface for connecting to the database 3; Reference numeral 8 denotes a CPU (central processing unit), which is equipped with a memory 8M and is in charge of overall control operations.

Criteria for determining the frequency of images to be stored in each storage medium are preset in the CPU 8 and stored in the memory 8M. The factors used to determine this criterion include the imaging modality (imaging device), the imaging site, the patient name (patient ID), the examination name (examination ID), the requesting department name (ID), the imaging date and time, the name of the interpretation room, etc. can be used alone or in combination. Each of the above-mentioned elements is used as ancillary data in pairs with image data. Next, the operation of this embodiment will be explained.

[0013] A user such as an image interpreting doctor uses the touch panel 12.
When a patient ID is input via the workstation 4, a corresponding image is requested from the database 3 via the network 2. The database 3 stores images for a plurality of tests, for example, 10, including the latest test for the patient, as shown in Figure 2.
It is transmitted to the workstation 4 in the form of incidental data + image data and stored in the hard disk 6 as shown in FIG.

[0014] The CPU 8 looks at the incidental data of the latest examination, compares it with the criteria stored in advance in the memory 8M, and determines which images are frequently used. For example, the latest test is
When the head is photographed using a line CT, the CPU 8 reads the accompanying data stored in the hard disk 6, and also the database 3.
CT head images taken in the past, for example, for five examinations, and images from the latest examination are stored in the image memory 5 via the network 2. As a result, images differentiated as follows are stored in the image memory 5, hard disk 6, and database 3. Image memory: CT, head, 5 examinations, hard disk: 10 examinations, database: all examinations. ) is created and stored in memory 8M.

Next, the user presses the touch panel 1 for image interpretation.
When the image ID is input from 2, a control signal based on this interrupts the CPU 8, and the CPU 8 takes in the ID information into a register. Subsequently, by searching the correspondence table, the desired image is displayed on the display 12 via the MDS 9. This allows the user to interpret the desired image.

[0016] The CPU 8 has a function of learning and inferring the judgment criteria by storing the past image use results of the user and finding common denominators among these elements. For example, as a criterion for determining how many past examinations to view frequently, the most frequent value of n times for past users is determined and this value is stored as the criterion.

As described above, according to this embodiment, the CPU 8 is set in advance with criteria for determining the frequency of images, and
The system is configured so that the judgment criteria can be changed by learning and inferring based on the user's past image usage record, so the frequency of images to be stored in each storage medium is constantly determined and the optimal storage medium is automatically selected. Can be stored. This reduces the time it takes to retrieve images from storage media and display them on a workstation in the order of frequency.
Image interpretation efficiency can be improved.

FIG. 3 shows a second embodiment of the present invention, in which the same reference numerals as in FIG. 1 indicate the same parts. Note that the image capturing device 1, network 2, and database 3 are omitted. Reference numeral 14 denotes an input device, which includes, for example, a keyboard, an ID reader, etc., and is used by the user to arbitrarily input judgment criteria. This judgment criterion is based on the memory 8 of the CPU 8.
Stored in M.

Criteria for determining the frequency of images to be stored in each storage medium are set in advance in the CPU 8 in response to a user's request, and are stored in the memory 8M. Elements for determining this judgment criterion include those similar to those in the above embodiment. Each of these elements is used as ancillary data in pairs with image data. Next, the operation of this embodiment will be explained.

[0020] It is assumed that a user such as an image interpreting doctor inputs judgment criteria in advance from the input device 14 in response to this request, and the criteria are stored in the memory 8M of the CPU 8. In this state, when the user inputs a patient ID via the touch panel 12 in order to perform image interpretation, the workstation 4 requests the database 3 via the network 2 for the corresponding image. The database 3 transmits images for a plurality of, for example, 10 examinations, including the latest examination of the patient, to the workstation 4 in the form of incidental data + image data, as shown in FIG. 2, and stores them in the hard disk 6. Thereafter, the same operations as those in the above embodiment are continued, thereby allowing the user to interpret the desired image.

[0021] The user can arbitrarily input judgment criteria via the input device 14 whenever necessary. The input criteria are sequentially rewritten and stored in the memory 8M of the CPU 8.

As described above, according to the second embodiment, the CPU
8, the criteria for determining the frequency of images are set in advance according to the user's request, and the criteria can be changed according to the user's needs, so the frequency of images to be stored in each storage medium can be adjusted. can be constantly judged and automatically stored in the most suitable storage medium. As a result, effects similar to those of the embodiment described above can be obtained.

[0023]

Effects of the Invention As described above, according to the present invention, the judgment criteria for medical images retrieved from the database are set in advance so as to be changeable as appropriate, and the images to be stored are checked against the judgment criteria. Since images are stored in the optimal storage medium after determining the frequency of images, images with high frequency can be stored in a storage medium with a short retrieval time, giving priority to other images, improving image interpretation efficiency. be able to.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a first embodiment of a routine image interpretation system of the present invention.

FIG. 2 is a schematic diagram necessary for explaining the embodiment of the present invention.

FIG. 3 is a block diagram showing a second embodiment of the invention.

FIG. 4 is a block diagram of a conventional example.

[Explanation of symbols]

1 Image interpretation device 3 Database 4 Workstation 5 Image memory (high speed memory) 6 Hard disk (low speed memory) 7 Display 8 CPU (Central Processing Unit) 8M memory 12 Touch panel 14 Input device

Claims (1)

[Claims] Claim 1: In a routine image interpretation system that stores medical images retrieved from a database via a network in high-speed memory and low-speed memory of a workstation, and displays desired medical images on the workstation as needed, means for presetting and storing judgment criteria for medical images retrieved; means for determining the frequency of retrieved medical images by comparing them with the judgment criteria read from the storage means; a control means for automatically storing images in the high-speed memory, frequently-frequent medical images in the low-speed memory, and other medical images in the database; 1. A routine interpretation system comprising: a display control means for displaying images taken out from the image display;