Radioactive ray image processing method, apparatus and system for same, image diagnosis supporting method and system
An image processing and radiation technology, which is applied in the fields of radiological diagnostic equipment, image data processing, image data processing, etc., can solve the problem of low detection accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0118] Figure 23 is a flowchart showing the processing of this embodiment. In step F001, moving images representing breathing dynamics are obtained by the above-mentioned method. Here, a plurality of image frames can be obtained. In step F002, image analysis is performed for each frame to detect suspicious points of a disease in a conventional still image, and a site suspicious of a disease (also called an area of interest) is detected from each image. In the judging step of F003, when receiving the result of step F002 and detecting a suspicious point of disease in at least one frame, the program proceeds to step F004, and outputs (displays) the meaning of suspicious point of disease in step F004. In step F003, if there is no suspicion of disease at all, the procedure proceeds to step F005 and is output as no doubt of disease.
[0119] In this embodiment, even if there are few suspicious cases, a report is made to the doctor as the diagnoser that the disease is suspiciou...
Embodiment 2
[0123] The present embodiment is the same as the first embodiment, but is different in that a site suspected of a disease (also referred to as a region of interest) is presented as an output result for all frames. exist Figure 22 shows an example where candidate diseases are detected in frames F1 and F18, but if based on Figure 7 By labeling an image with such motion vectors, it is possible to easily specify the region of the same part in other frames. That is, if Figure 24 As shown, the region of another frame corresponding to the region of the candidate disease found in F1 can be easily specified based on this flag.
[0124] In this case, even in other frames, although signs of the disease can be seen, they may not be detected by the detection capability of conventional CAD processing for still images.
[0125] Therefore, in the present embodiment, as a suspicious region of a disease, it is also displayed in the corresponding region of another frame as a suspicious reg...
Embodiment 3
[0128] In this embodiment, the accuracy of image analysis for detecting suspicious points of a disease is variable, or it is assumed that a plurality of image analysis algorithms can be used. The candidate disease region recognized in Example 2 cannot be detected in the first image frame, but it may be detectable if the detection accuracy of CAD processing (image analysis) is improved, and it may be possible if other analysis algorithms are used. detection. Therefore, in this embodiment, when a suspicious point of a disease is detected in at least one frame, it can be judged that a high-precision image analysis algorithm is more suitable for corresponding areas corresponding to other frames. If the suspicious point of the result cannot be detected, then, Suspects of disease initially detected will not be disease.
[0129] Figure 26 is a flowchart showing the processing of this embodiment. Since steps F001-F003 are related to Figure 23 are the same, so its description is ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More