Medical Image Region of Interest Indicator System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In medical image diagnosis using three-dimensional medical images, it is challenging to visually recognize the location of lesions detected by CAD systems within displayed images, leading to increased workload in verifying the presence and significance of detected objects.
Innovation Solution
An image processing apparatus that acquires medical images with low-dimensional images, automatically detects regions of interest, generates axis information representing space or time axes, and outputs a display signal showing low-dimensional images, axis information, and additional information indicating the presence and content of regions of interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CAD automatically detects regions of interest in medical images, then diagnostic accuracy is improved, but the workload for verifying detected objects increases
Solution Approach 1:
The patent introduces an intermediary visual indicator system that bridges the CAD detection results and the radiologist's verification process. The indicators provide intermediate information about detected regions, allowing radiologists to verify findings more efficiently without having to manually search through all images.
Solution Approach 2:
The patent creates visual copies or representations of the detected regions of interest through indicators displayed on the image. These visual copies allow radiologists to quickly identify and verify detected areas without having to mentally track or search for the actual regions in the medical images.
2Loss of information
If multiple low-dimensional images are displayed to show region of interest positions, then diagnostic information completeness is improved, but visual recognition of lesion locations becomes more difficult
Solution Approach 1:
The patent adds a new visual dimension by superimposing indicators on the displayed images. Instead of requiring radiologists to mentally map positions across multiple separate images, the indicators provide an additional visual layer that directly marks lesion locations on the images themselves, transforming the information presentation from spatial coordination to direct visual identification.
Solution Approach 2:
The patent merges the region of interest information with the visual display of medical images by superimposing indicators directly on the images. This combination allows radiologists to see both the anatomical context and the detected regions simultaneously, eliminating the need to cross-reference multiple separate displays.
3Loss of information
If axis information is added to indicate space or time axes in low-dimensional images, then spatial orientation understanding is improved, but device complexity increases
Solution Approach 1:
The patent extracts only the essential spatial or temporal orientation information needed for diagnosis and represents it through simple visual indicators. Rather than implementing complex axis systems, the solution extracts and displays only the critical orientation cues that help radiologists understand the context of detected regions.
Data Source
AI summary
An image processing apparatus, an image display system, an image processing method, and a program by which it is possible to grasp, on a screen for observing a given low-dimensional image having a lower level of dimensions than a medical image, information of the low-dimensional image in which a region of interest is present and information of the region of interest. A medical image including two or more low-dimensional images (102) is acquired, region-of-interest information representing information of a region of interest (108) automatically detected from the medical image is acquired for each of the low-dimensional images, axis information (104) representing a space axis or a time axis is generated, additional information (106) associated with the axis information is generated, the additional information including presence information indicating that the region of interest is present and content information indicating content of the region of interest, and the low-dimensional images, the axis information, and the additional information are displayed on a display.