Ultrasound Cross-Section Analysis With Thumbnail-Based Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ultrasound diagnostic systems face inefficiencies in selecting cross-sections for myocardial analysis, with existing acquisition and AI-based functions often requiring unnecessary image generation and analysis, leading to increased time and user burden.
Innovation Solution
An analysis apparatus and ultrasound diagnostic system that utilizes processing circuitry to efficiently select cross-sections for myocardial analysis by specifying and selecting cross-sections based on association degrees, using thumbnail images and highlighting relevant data for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If AI-based cross-section selection is performed by generating and analyzing images for all cross-sections, then cross-sections necessary for myocardial analysis can be automatically obtained, but time is required for image generation and analysis of unnecessary cross-sections
Solution Approach 1:
The patent extracts only the necessary cross-section information from the full dataset by using thumbnails as representative samples. Instead of processing all cross-section images, the system extracts thumbnail images from each cross-section and performs AI-based analysis only on these extracted representatives, thereby automatically selecting necessary cross-sections without the time penalty of analyzing all images.
Solution Approach 2:
The patent segments the cross-section selection process into two stages: first, AI-based analysis of thumbnails to identify candidate cross-sections; second, detailed analysis only of the selected candidates. This segmentation allows automated selection to occur efficiently at the thumbnail level, avoiding the time-consuming full-image analysis of all cross-sections.
2Ease of operation
If acquisition assisting function is used to manually designate cross-section order, then cross-sections can be acquired, but user burden increases due to non-optimal ordering
Solution Approach 1:
The patent implements self-service by enabling the system to automatically determine the optimal cross-section acquisition order through AI-based thumbnail analysis. The system serves itself by autonomously identifying which cross-sections are necessary and in what order they should be acquired, eliminating the need for users to manually designate the acquisition sequence while reducing operational complexity.
3Reliability
If all cross-section images are processed for analysis, then comprehensive data is obtained, but unnecessary image processing increases time consumption
Solution Approach 1:
The patent performs preliminary action by analyzing thumbnails before processing full cross-section images. The AI-based thumbnail analysis serves as a preliminary filtering step that identifies which cross-sections are worth further detailed analysis, ensuring that comprehensive data is obtained only for relevant cross-sections while maintaining high productivity through early elimination of unnecessary processing.
Data Source
AI summary
According to one embodiment, an analysis apparatus includes processing circuitry. The processing circuitry obtains a thumbnail of each of multiple pieces of cross-section image data related to first cross-section image data selected by a user and showing a cross-section of an organ, specifies, using the multiple thumbnails obtained, a cross-section of an organ included in each of the multiple thumbnails, and selects a cross-section for use as a target of analysis other than a cross-section selected by the user from the specified cross-sections based on a degree of association between the first cross-section image data and each of the multiple pieces of cross-section image data.