Correlating Radiology Features with Pathology Tissue Characteristics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radiology imaging methods lack direct correlation between image features and tissue biology, making it difficult to characterize lesions effectively and translate results across patient cohorts or cancer types.
Innovation Solution
A method that performs computer-based analysis of pathology images to recognize tissue and cellular characteristics, computes radiology features from radiology images, and determines correlations between these features and biological ground truth information to improve lesion characterization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radiology images are analyzed using quantitative image features to objectively describe tumor phenotypes, then the ability to characterize lesions is improved, but the correlation between image features and tissue biology remains weak due to lack of spatial resolution and lack of correlation with tissue biology
Solution Approach 1:
The patent introduces pathology images as an intermediary medium that bridges radiology images and tissue biology. Pathology images serve as a mediator that provides both spatial information and biological ground truth, enabling the correlation between radiology image features and tissue characteristics without requiring direct measurement of tissue properties from radiology images alone.
Solution Approach 2:
The patent segments the analysis into separate components: radiology image analysis for quantitative features, pathology image analysis for tissue characterization, and correlation determination between the two. This segmentation allows each component to be optimized independently while achieving comprehensive lesion characterization through their integration.
2Reliability
If pathology images are used to provide subcellular resolution and direct information about tissue biology, then the correlation with therapy response is improved, but the availability is limited to only biopsied lesions which may not be representative for the whole tumor
Solution Approach 1:
The patent creates a virtual copy of the biopsy sample's tissue characteristics by analyzing the pathology image and generating a derived pathology image that represents the recognized tissue and cellular characteristics. This derived image serves as a representative model that can be correlated with radiology features without requiring actual biopsy material from every lesion, thereby extending applicability to all lesions while maintaining biological relevance.
3Measurement precision
If a large number of quantitative image features are extracted from radiology images, then the objective description of tumor phenotypes is improved, but the biological meaning of the features remains unknown making translation to other patient cohorts difficult
Solution Approach 1:
The patent applies local quality by analyzing specific regions of interest in both radiology and pathology images, focusing on particular tissue characteristics and cellular features rather than treating the entire image uniformly. This localized analysis enables more meaningful correlations between image features and biological properties, improving the biological interpretability and facilitating translation across patient cohorts.
Data Source
AI summary
The invention relates to a method for use in characterizing lesions in radiology images, comprising performing a computer-based analysis of a pathology image of a sample of a lesion of a subject in order to recognize tissue and/or cellular characteristics of the lesion, wherein the analysis produces a derived pathology image that represents the recognized tissue and/or cellular characteristics, computing one or more radiology features for the lesion from a radiology image of the lesion, and determining correlations between the computed one or more radiology features and the recognized tissue and/or cellular characteristics. With this method, biological ground truth information can be used to identify radiology features that are indicative of certain tissue and/or cellular characteristics of lesions and that may therefore be better suited for characterizing the lesions. Such radiology features can then be used together with the corresponding correlations for an improved characterization of lesions in radiology images.