Localizer Image Plane Selection via AI Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current non-invasive imaging technologies face challenges in acquiring high-quality localizer or scout images, which are often of lower quality and resolution than diagnostic images, making them difficult to interpret and relate to the proposed scan process and patient anatomy.
Innovation Solution
A method involving the acquisition of a plurality of localizer or scout images, which are then processed using a localizer network to select a subset of images for detection and visualization of anatomic landmarks-of-interest. These images are further processed by a scan plane network to determine image scan planes or parameters that contain regions of the anatomic landmark-of-interest, allowing for the acquisition of diagnostic images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If localizer or scout images are acquired to help relate scanner geometry with patient anatomy, then the ability to plan scan processes is improved, but the image quality and resolution deteriorate
Solution Approach 1:
The patent introduces an artificial intelligence-based image processing system as an intermediary between the low-quality localizer images and the scan planning process. The AI system enhances the localizer images by selectively improving regions containing anatomical landmarks while preserving the original low-quality nature of non-landmark areas, thereby enabling accurate scan planning without compromising overall image quality
Solution Approach 2:
The patent applies local quality enhancement by selectively improving only the regions of localizer images that contain anatomical landmarks of interest. The AI system identifies landmark regions and applies enhancement algorithms specifically to those areas, leaving other regions unchanged. This allows the images to serve their dual purpose of maintaining diagnostic quality where needed while preserving the fast-acquisition characteristics of localizer images overall
2Difficulty of detecting and measuring
If the resolution of localizer images is increased to improve interpretability, then the ease of interpretation is improved, but the acquisition time and complexity increase
Solution Approach 1:
The patent applies partial action by selectively enhancing only the necessary portions of localizer images - specifically the regions containing anatomical landmarks - rather than processing the entire image at high resolution. The AI system identifies and enhances only the landmark-containing regions, reducing the overall processing burden and maintaining fast acquisition times while improving interpretability where it matters most
Solution Approach 2:
The patent changes the quality parameters of localizer images selectively using AI-based processing. The system adjusts resolution and enhancement parameters specifically for regions containing anatomical landmarks, while maintaining original parameters for other regions. This selective parameter modification improves interpretability of critical areas without requiring a complete increase in overall image acquisition resources
Data Source
AI summary
The present disclosure relates to use of a workflow for automatic prescription of different radiological imaging scan planes across different anatomies and modalities. The automated prescription of such imaging scan planes helps ensure contiguous visualization of the different landmark structures. Unlike prior approaches, the disclosed technique determines the necessary planes using the localizer images itself and does not explicitly segment or delineate the landmark structures to perform plane prescription.