JPH01255974A - Diagnostic image processing system - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image workstation in the medical field, and particularly to a diagnostic imaging method that is effective in improving diagnostic accuracy.

[Conventional technology]

Previously published literature (for example, Sueda et al.: Image Processing Expert System Toshiba Shivu (Vol. 40, No. 5).

In 1985 (P403-P406).

This paper discusses an expert system with built-in knowledge about image processing so that even non-experts in image processing can easily perform image processing. here.

Processing procedures, parameter settings, etc. are performed mainly for the purpose of pre-processing (noise removal, binarization, etc.) for pattern recognition.

Here, the knowledge and know-how regarding processing procedures that image processing experts have is incorporated into the computer's knowledge, and based on this, the computer itself takes the place of the image processing expert and provides guidance to the user about image processing procedures. It describes a system to do this.

This system enables even engineers outside the field of image processing to easily perform image processing, and aims to improve the efficiency of system development.

In the guidance method, the system inquires about the state of the image to be processed, makes inferences based on that information, and executes image processing.

[Problem to be solved by the invention]

The above conventional technology supports image processing related to preprocessing such as pattern recognition and 2-image analysis, and it is important to consider performing image processing suitable for 1 diagnosis in conjunction with the recognition process of 1 person, such as in medical image diagnosis. The operator had to interact with the system to give instructions on what kind of processing to perform.

An object of the present invention is to provide a diagnostic image processing method that does not require image processing instructions from the operator.

[Means for Solving the Problems] The above object is achieved by adding a function of storing diagnostic information to one image processing unit and inferring the most suitable image processing content using the collected diagnostic information.

[Operation] The operation of the technical means is explained in Figure 1.0 Image processing section 1
External storage device 1 from the diagnostic recording unit 100 added to 10
Based on the diagnostic findings 11 input in step 8, the inference mechanism 15 uses the rule-type process selection button m12 to select the most suitable process definition 14 from the process definition knowledge 13 that defines the content of image processing. The processing definition knowledge is comprised of a program name 140 for executing one image processing for boat fishing and parameters 141 necessary for executing the program. On the other hand, the image processing program execution section 16 in the image processing section 110 starts a necessary program based on the processing definition knowledge i14 received from the inference mechanism 15, and displays the processed image on the display 17.

Note that the diagnosis recording section 100 and the image processing section 110 perform processing in conjunction with each other to assist the doctor in image diagnosis.

〔Example〕

An embodiment of the present invention will be described below with reference to FIGS. 2 and 3.

FIG. 2 specifically shows the processing explained in FIG.
is a part corresponding to FIG.

Based on the diagnostic finding 11 "high absorption in the ventricles" inputted from the diagnostic recording unit 100, necessary processing is inferred using the built-in processing selection knowledge 12. In this example, the instruction is inferred to "perform process 1 for conditions where the site is the ventricle and the concentration is high absorption." Process 1 is processing definition knowledge 14
The contents are described in 240., where the first process is gradation conversion 240. to relatively increase the contrast of the high absorption region. As the second process, a process 241 is defined to enlarge the ventricle so that it can be easily observed by a doctor.
According to this content, the original image 210 is processed by the image processing unit to obtain the processed image 211. A flowchart of the above-mentioned processing procedure is shown in FIG.

Step 30: Collect diagnostic findings regarding the images in the diagnostic recording section.

Step 31: Based on the diagnostic findings, the image processing that best meets the conditions is inferred using rule-based image processing selection knowledge.

Step 32 Based on the processing definition that defines the content of the image processing selected in step 31, the image processing program to be executed and the parameters to be used are determined.

Step 33 The information regarding the processing content obtained in step 32 is transferred to the image processing section.

Step 34: The image processing section receives the image processing contents of step 33.

Step 35: Execute the specified image processing.

Step 36: Display the processed image on the display.

FIG. 4 shows an example of the hardware configuration of the system, in which a display 40. Printer 41. A processing bag [42] and an external storage device 43 storing the knowledge base are shown.

Further, as the image processing section 110, the image display 44
1, an image processing device 45 and an external storage device 46 for storing image data are shown. Furthermore, both systems are connected by a transmission line 47.

Note that in this configuration, one diagnosis recording section 100 and one image processing section 110
It goes without saying that an integrated configuration is also possible.

〔Effect of the invention〕

According to the present invention, as in medical image diagnosis, image processing suitable for diagnosis in conjunction with human recognition processes can be realized without issuing special image processing instructions, thereby improving diagnostic efficiency and improving diagnostic efficiency. It is possible to improve IR accuracy.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of the entire system, Fig. 2 is a specific example, Fig. 3 is a flowchart showing the processing procedure, and Fig. 4 is an explanatory diagram of the entire system.
The figure shows a system to which the present invention is applied. 14 Ha present meaning, Chi 0 knowledge to θ diagnosis record country 2) IbJ Riga I, Atmosphere bi ¥ J3 diagram

Claims (1)

[Claims] 1. A medical image workstation comprising an image display device, an image processing device, and an external storage device for images collects finding information related to image diagnosis and performs diagnostic image processing based on the finding information. A diagnostic image processing method characterized by: 2. The above-mentioned cross-sectional image processing is performed according to the mode in which the image processing content is automatically determined by a computer and the mode in which the operator selects from candidates extracted by the computer. Image processing method.