Dual Energy X-ray Image Analysis Using Temporal Pixel Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current evaluation methods for 2-dimensional projection images of vascular systems are complex and require manual expertise to assess myocardial perfusion, often leading to artifacts and inefficiencies in identifying vascular structures and perfusion areas.
Innovation Solution
An automated evaluation method that generates a 2-dimensional evaluation image by analyzing the time characteristic of pixel values in combination images, allowing for the identification of vascular system pixels, perfused areas, and non-perfused areas without user specification, using a computer to assign types and display perfusion extent in a color-coded format.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual expertise is used to assess myocardial perfusion, then diagnostic accuracy is improved, but device complexity and time consumption increase
Solution Approach 1:
The system performs automated classification of pixels into vascular structures, perfusion areas, and non-perfused areas using algorithmic analysis of time characteristics. The computer automatically assigns types to pixels based on temporal patterns without requiring manual intervention, making the system self-sufficient while maintaining diagnostic accuracy
Solution Approach 2:
The invention transforms the assessment from manual visual evaluation to automated temporal parameter analysis. By analyzing time characteristics of pixel values across multiple projection images, the system extracts quantitative features that enable automatic classification, changing the evaluation parameters from subjective visual assessment to objective temporal measurement
2Reliability
If manual assessment methods are used, then diagnostic reliability is improved, but productivity decreases
Solution Approach 1:
The invention replaces the manual mechanical assessment process with an automated computational system. The computer analyzes time characteristics of pixel values and automatically classifies regions, substituting human manual evaluation with algorithmic processing that maintains reliability while dramatically increasing productivity
Solution Approach 2:
The system processes multiple projection images continuously to extract temporal information. By analyzing the continuous time characteristics of pixel values across the image sequence, the system maintains diagnostic reliability while enabling rapid automated assessment without interruption or manual intervention
3Productivity
If automated methods are used, then productivity is improved, but measurement precision may worsen
Solution Approach 1:
The invention adds the temporal dimension to the analysis by examining time characteristics of pixel values across multiple projection images. This fourth dimension (time) provides additional information that enables accurate automated classification, allowing the system to distinguish vascular structures from perfusion areas with high precision while maintaining productivity
Solution Approach 2:
The system uses the temporal evolution of pixel values as feedback to refine its classification. By analyzing how pixel intensities change over time across the image sequence, the algorithm receives feedback about the actual perfusion dynamics, enabling it to accurately identify perfusion areas and adjust its classification accordingly
4Measurement precision
If complex evaluation methods are used, then diagnostic accuracy is improved, but loss of time increases
Solution Approach 1:
The system performs preliminary analysis by extracting time characteristics from the sequence of projection images before final classification. By pre-processing the temporal information and identifying key temporal patterns early in the evaluation process, the system prepares the data for rapid automated classification, reducing the overall evaluation time while maintaining accuracy
Data Source
AI summary
A sequence of groups of projection images shows an object under examination comprising a vascular system and its environment. A computer determines a 2-dimensional evaluation image having a plurality of pixels based on combination images determined from the projection images of a group. The combination images have a plurality of pixels with pixel values. The sequence of the combination images shows the time characteristic of the distribution of a contrast medium in the object. The pixels of the evaluation image correspond to those of the projection images. The computer assigns each pixel, at least in a part area of the evaluation image, a type that is characteristic of whether the respective pixel corresponds to a vessel of the vascular system, a perfusion area or a background. It performs the assignment of the type on the basis of the time characteristic of the pixel values of the combination images.