Bone Age Assessment via Image Segmentation and Pixel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current bone age assessment methods, such as the Greulich-Pyle and Tanner-Whitehouse methods, rely on manual image comparison, leading to poor accuracy due to the subjective nature of image analysis.
Innovation Solution
An electronic apparatus and method for bone age assessment that divides input images into segmented images, processes pixels based on a reference value, and matches these segments with reference images to determine bone age grades, using a processor to automatically compare and prioritize image regions, adjusting reference values as needed for accurate matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual image comparison is used to determine bone age, then the assessment process is simple to perform, but the accuracy of bone age determination is poor
Solution Approach 1:
The patent divides the input medical image into multiple segmented images, each corresponding to a specific body part. This segmentation allows for focused analysis of individual bones and regions, improving measurement precision by enabling detailed comparison with reference images for each specific anatomical structure rather than treating the entire image as a single unit.
Solution Approach 2:
The patent processes pixel values by applying reference values to enhance specific features. By transforming the image data through parameter changes (pixel value processing based on reference values), the system improves the distinguishability of bone structures and characteristics, thereby enhancing the accuracy of bone age determination.
2Productivity
If manual image comparison is used for bone age assessment, then the system complexity is low, but the productivity of assessment is reduced due to time-consuming manual analysis
Solution Approach 1:
By segmenting the image into multiple parts corresponding to different body parts, the system can process and compare each segment independently and in parallel. This segmentation enables automated processing to efficiently handle multiple regions simultaneously, significantly improving productivity compared to manual whole-image analysis.
Solution Approach 2:
The patent uses reference images that represent standard bone appearances at different ages. By comparing segmented images against these reference copies, the system automates the assessment process, eliminating the need for manual expert analysis while maintaining or improving accuracy through consistent, repeatable comparisons.
3Measurement precision
If the entire input image is processed for bone age assessment, then comprehensive analysis is achieved, but computation time and resources are excessive
Solution Approach 1:
The patent divides the comprehensive input image into multiple segmented images, each representing a specific body part. This segmentation maintains comprehensive analysis by covering all relevant anatomical regions while reducing computation time by allowing parallel processing of smaller segments and enabling the system to focus computational resources on specific areas of interest rather than processing the entire image uniformly.
Solution Approach 2:
The patent determines priority levels for different segmented images based on their importance for bone age assessment. By processing high-priority segments first and potentially skipping or simplifying processing of lower-priority segments, the system achieves sufficient comprehensiveness for accurate bone age determination while significantly reducing overall computation time and resources.
Data Source
AI summary
The present disclosure proposes an apparatus for determining a bone age. The apparatus may divide an input image capturing a human body into a plurality of segmented images, determine a first segmented image having a highest priority for a first body part from the segmented images, process each of first pixels of the first segmented image based on a reference value, select a first reference image for the first body part from a reference image set, determine whether or not a partial region matching the first reference image exists in the first segmented image processed by the reference value, upon determining that the partial region exists, determine a bone age grade of the first body part based on the first reference image, and determine a bone age of the human body based on the bone age.