Gaze Detection for Developmental Disorder Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing developmental disorders in young children, such as those during medical check-ups at half-past-one-year-old, face challenges due to the shortage of psychiatrists and the need for more accurate and efficient diagnosis tools, particularly in detecting the point of regard which is crucial for early intervention.
Innovation Solution
A diagnosis assisting apparatus equipped with a display, imaging unit, line of sight detecting unit, point of view detecting unit, and output controller that displays diagnostic images to accurately detect and correct the point of view of a subject, using stereo cameras and infrared LED light sources to enhance diagnosis accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional interview methods are used for developmental disorder diagnosis, then diagnosis can be performed by psychiatrists, but the process takes too long and there is a shortage of psychiatrists
Solution Approach 1:
The patent replaces the mechanical interview process conducted by psychiatrists with an automated optical detection system. The line-of-sight detection apparatus uses imaging units and coordinate calculation algorithms to automatically detect and analyze the subject's gaze patterns, substituting human psychiatric evaluation with an automated visual analysis system that can process multiple subjects simultaneously without time constraints.
Solution Approach 2:
The system enables self-diagnosis capability by automatically capturing images, calculating line-of-sight coordinates, and generating diagnostic assessments without requiring psychiatrist intervention. The apparatus performs the entire diagnostic process independently, from image acquisition through coordinate detection to final assessment generation, making the diagnosis process self-sufficient and eliminating dependency on limited psychiatric resources.
2Measurement precision
If simple point of regard detection methods are used, then the system is easier to operate, but the detection accuracy is insufficient for reliable diagnosis
Solution Approach 1:
The patent transitions from simple two-dimensional image capture to three-dimensional spatial coordinate detection. By calculating line-of-sight directions and converting them into precise three-dimensional coordinates within the display region, the system adds a dimensional layer of accuracy that enables reliable point-of-regard detection while maintaining operational simplicity through automated coordinate transformation algorithms.
Solution Approach 2:
The patent introduces coordinate calculation as an intermediary process between image capture and point-of-regard detection. The line-of-sight direction calculation unit serves as a mediator that transforms raw imaging data into meaningful spatial coordinates, bridging the gap between simple image acquisition and accurate point-of-regard identification without requiring complex direct detection mechanisms.
3Measurement precision
If comprehensive line of sight detection is performed across all frames, then detection accuracy improves, but processing time and computational load increase
Solution Approach 1:
The patent applies partial action by detecting line-of-sight directions in selectively extracted frames rather than processing every frame. The system identifies and processes only the necessary frames that contain meaningful gaze information, performing sufficient detection to ensure diagnostic accuracy while avoiding the excessive processing burden of analyzing every single frame, thus achieving an optimal balance between precision and efficiency.
Data Source
AI summary
A diagnosis assisting apparatus includes a display, an imaging unit configured to image a subject, a line of sight detecting unit configured to detect a line of sight direction of the subject from a picked-up image imaged by the imaging unit, a point of view detecting unit configured to detect a point of view of the subject in a display region of the display based on the line of sight direction, and an output controller configured to display a diagnostic image including a natural image and a geometrical image, and the point of view detecting unit detects the point of view of the subject in a case where the diagnostic image is displayed.