Gaze Tracking Medical Image Diagnosis Assistance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing complexity and volume of medical images due to an aging population and advances in medicine lead to a higher burden on radiologists, increasing the risk of missed or misinterpreted abnormal shadows during image diagnosis, as existing systems struggle to accurately identify observed and unobserved regions, especially when images are improperly displayed.
Innovation Solution
An image diagnosis assistance apparatus that detects a user's gaze position on a display screen, determines the observed region in a tomographic image, and identifies the corresponding region in a three-dimensional medical image, using a combination of display means, detection means, and determination means to provide accurate identification of observed and unobserved areas, even when images are not properly displayed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sight line information is used to identify observed regions, then the system can track where the radiologist is looking, but it cannot determine whether the displayed image quality is sufficient for proper interpretation
Solution Approach 1:
The patent introduces display state information as an intermediary factor between sight line detection and observed region identification. The determination unit uses both sight line information and display state information (window level, window width, luminance) to comprehensively determine whether a region has been properly observed, rather than relying solely on sight line data
Solution Approach 2:
The system provides feedback by comparing the display state against predetermined criteria for proper image display. When the display state does not meet the criteria (e.g., improper window level or luminance), the system identifies this as a potential interpretation omission risk, allowing the radiologist to adjust the display before making a diagnosis
2Ease of operation
If the system relies on sight line information alone to determine observed regions, then the operation is simple, but it fails to account for improper display conditions that may hide abnormal shadows
Solution Approach 1:
The system performs preliminary evaluation of display state before finalizing the observed region identification. By checking window level, window width, and luminance values in advance against predetermined criteria, the system ensures that only regions displayed under appropriate conditions are counted as properly observed, preventing misidentification due to poor display quality
3Reliability
If multiple display parameters are monitored to ensure proper image display, then interpretation accuracy improves, but system complexity increases
Solution Approach 1:
The determination unit serves multiple functions: it evaluates sight line information, assesses display state parameters (window level, window width, luminance), and determines observed regions based on the combination of these factors. This multi-functional approach consolidates what would otherwise require separate systems into a single integrated unit
Data Source
AI summary
An image diagnosis assistance apparatus displays, on a display screen, a tomographic image obtained from a three-dimensional medical image, and detects a gaze position of a user on the display screen. The image diagnosis assistance apparatus determines an observed region in the tomographic image based on the detected gaze position, and identifies the observed region in the three-dimensional medical image based on the region determined to have been observed in the tomographic image.