Atlas-Based Candidate Element Identification for Disease State Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current quantitative medical imaging methods face challenges in accurately determining disease states due to large uncertainties in the relationship between intensity values and patient properties, making automated diagnosis time-consuming and experience-dependent, especially in whole-body imaging without a known limited region of suspected disease.
Innovation Solution
The method involves providing quantitative images with corresponding atlases, determining a correspondence map to compare element values, localizing candidate elements, classifying them based on properties, and analyzing the classes to determine disease states, utilizing a system with a processing unit and computer-readable media to automate the process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated diagnosis methods are used, then productivity is improved, but reliability deteriorates due to uncertainties in intensity values
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing expected intensity value ranges for normal tissue in an atlas during the offline phase. This allows the automated diagnosis system to quickly compare patient data against pre-established benchmarks, improving diagnosis speed while maintaining reliability through accurate reference comparisons.
Solution Approach 2:
The patent uses copying by creating a reference atlas from healthy patient data that replicates normal tissue intensity patterns. This copied reference model enables automated systems to diagnose new patients by comparing against the replicated normal patterns, improving both speed and reliability of automated diagnosis.
2Measurement precision
If visual searching through large medical imaging datasets is performed, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent applies taking out by extracting only the most relevant intensity value ranges from the complete medical imaging dataset that indicate potential disease states. The system identifies and extracts abnormal intensity patterns compared to the atlas, allowing rapid diagnosis without manually reviewing entire large datasets, thus reducing time loss while maintaining detection precision.
Solution Approach 2:
The patent uses parameter changes by transforming the diagnostic approach from visual inspection of raw intensity values to comparison against normalized atlas-based reference ranges. This parameter transformation enables automated systems to quickly identify deviations from normal patterns, significantly reducing diagnosis time while maintaining or improving detection accuracy.
3Measurement precision
If quantitative imaging techniques are used, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces an intermediary atlas component that mediates between the complex quantitative imaging data and the diagnosis process. The atlas serves as an intermediate reference layer that simplifies the interpretation of quantitative intensity values by providing pre-computed normal ranges, thereby maintaining measurement precision while reducing the effective complexity of the diagnostic system.
4Adaptability or versatility
If whole body imaging is performed, then adaptability is improved, but loss of time increases due to lack of limited region focus
Solution Approach 1:
The patent applies taking out by extracting and highlighting only the regions of interest from whole-body imaging data that show abnormal intensity patterns. The system automatically identifies and extracts suspicious areas by comparing against the atlas, allowing comprehensive whole-body coverage while reducing diagnosis time by focusing attention only on extracted abnormal regions rather than reviewing entire body scans.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods for determining a disease state of image elements are disclosed herein. An example method may include providing quantitative and non-quantitative images, determining a correspondence map of the quantitative images to an atlas, and determining candidate elements of the quantitative images by comparing the elements of the quantitative images to atlas elements of the atlas. The example method may also include localizing the candidate elements in the non- quantitative images, classifying the candidate elements based on properties of the non-quantitative images, and determining a disease state of the candidate elements.