Gaze-Based Assessment Scoring for More Accurate ASD Evaluation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for assessing autism spectrum disorder (ASD) lack accuracy in identifying specific characteristics of individuals with ASD.

Innovation Solution

An assessment device and method that utilizes a display unit to show multiple unit images with specificity and sensitivity objects, detecting gaze points, and calculating object and entire gaze rates to determine assessment scores based on these rates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional gaze point detection methods are used to assess ASD, then the assessment process is simple, but the accuracy in identifying specific characteristics of ASD is insufficient

Engineering Contradiction:
Improveaccuracy of ASD assessmentVSAvoidcomplexity of assessment system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the assessment into multiple tasks (first task with geometric pattern, second task with person image) and divides gaze analysis into specific regions (geometric pattern region, person image region, other regions). This segmentation allows precise measurement of gaze characteristics in different contexts while maintaining a structured, manageable assessment framework.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces temporal dimension by analyzing gaze points across different time periods (during first task, during second task) and spatial dimension by dividing the display into multiple determination regions. This multi-dimensional approach enhances assessment accuracy by capturing dynamic gaze behavior patterns rather than static single-point measurements.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If multiple determination regions are used to track gaze points across different tasks, then the accuracy of gaze behavior quantification improves, but the complexity of region setting and data processing increases

Engineering Contradiction:
Improveprecision of gaze behavior measurementVSAvoiddifficulty of gaze point analysis
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent divides the display area into multiple determination regions (geometric pattern region, person image region, other regions) for each task. This segmentation enables precise tracking of gaze behavior in different contextual areas while organizing data processing into manageable regional units, reducing overall analytical complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent focuses measurement on specific determination regions relevant to each task rather than analyzing the entire display area uniformly. This partial action approach concentrates processing resources on critical areas, improving measurement precision while reducing unnecessary computational complexity in irrelevant regions.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4483816B1Evaluation device and evaluation program
Publication Date: 2026.01.21 SYSMEX CORP
  • EP4483816B1 patent drawingFigure 1
  • EP4483816B1 patent drawingFigure 2
  • EP4483816B1 patent drawingFigure 3

AI summary

An assessment device includes a display unit; a gaze-point detecting unit configured to detect a position of a gaze point of a subject; a display control unit configured to display multiple unit images including at least one of an object having specificity and an object having sensitivity on the display unit in different tasks; a region setting unit configured to set an object determination region corresponding to each of the objects and an entire determination region corresponding to an entirety of the display unit for each of the tasks; a determining unit configured to determine whether the gaze point is present in each of the object determination region and the entire determination region based on the detected position of the gaze point; an arithmetic unit configured to calculate an object gaze rate indicating a rate at which the gaze point is present in the object determination region in each of the tasks, and an entire gaze rate indicating a rate at which the gaze point is present in the entire determination region throughout all the tasks based on a determination result; and an assessing unit configured to calculate a score based on the object gaze rate and the entire gaze rate calculated by the arithmetic unit, and acquire assessment data based on the score.