GI Tract Image Segmentation for Mucosal Disease Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for assessing and monitoring mucosal diseases in the gastrointestinal tract are inadequate due to lack of standardization, sensitivity, specificity, and comprehensive evaluation, often relying on invasive or expensive radiological methods that fail to accurately assess mucosal inflammation and disease severity.
Innovation Solution
A computerized method and system for parsing images of the gastrointestinal tract into segments, assigning severity values to each segment, and generating a representation of mucosal disease location and severity, allowing for standardized and comprehensive assessment of mucosal diseases using capsule endoscopy images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radiological methods (CTE or MRE) are used to assess mucosal disease, then comprehensive evaluation of disease extent and complications is improved, but the cost and accessibility of the assessment deteriorates
Solution Approach 1:
The patent segments the gastrointestinal tract into multiple defined regions (e.g., proximal small bowel, distal small bowel, colon) and assigns specific scoring criteria to each segment. This segmentation allows for standardized assessment that can be performed on capsule endoscopy images, making the evaluation more accessible while maintaining comprehensive disease extent assessment through systematic regional evaluation.
Solution Approach 2:
The patent introduces specific quantitative parameters for mucosal assessment including lesion type classification (aphthous, linear, serpiginous, cobblestone), severity grading (mild, moderate, severe), and involvement percentage (0-100%). These parameter changes transform subjective visual assessment into objective, standardized metrics that can be reliably applied across different practitioners and settings.
2Measurement precision
If invasive procedures are used to assess mucosal inflammation, then measurement precision is improved, but patient comfort and procedure complexity worsens
Solution Approach 1:
The patent uses capsule endoscopy to create visual copies (images) of the mucosal surface, allowing for detailed assessment of mucosal inflammation without direct physical contact or tissue sampling. The system captures multiple images throughout the GI tract that can be reviewed and scored, providing a non-invasive alternative to endoscopic biopsy while maintaining assessment precision through systematic imaging and scoring of mucosal features.
3Measurement precision
If standardized scoring systems are implemented for mucosal assessment, then measurement precision and comparability are improved, but the time and complexity of assessment worsens
Solution Approach 1:
The patent divides the assessment into discrete, pre-defined segments of the GI tract with specific scoring criteria for each. This segmentation structure allows assessors to systematically work through regions without missing areas, and the standardized format enables efficient documentation. The segmented approach actually reduces time by providing a clear workflow compared to unstructured assessment.
Solution Approach 2:
The patent establishes specific parameter thresholds and grading criteria (e.g., lesion involvement of 0-10%, 11-30%, 31-50%, 51-70%, 71-100%) that enable rapid categorization and scoring. These parameter changes transform complex visual assessment into discrete choice selections, reducing the time required for evaluation while maintaining precision through consistent application of defined criteria across all assessments.
Data Source
AI summary
A computerized-method for a mucosal assessment of a mucosal disease in a Gastrointestinal Tract (GIT) of a subject, including receiving a stream of images of at least a portion of the GIT, parsing the stream into a plurality of segments, wherein each segment corresponds to a region of the at least portion of the GIT, obtaining a set of values for each segment, wherein the set of values refers to the pathological involvement of the segment in the mucosal disease and to severity of mucosal manifestation of the mucosal disease in the segment, and based on said set of values for each segment, generating a representation indicating the location and severity of the mucosal manifestation of the mucosal disease in the entirety of the at least portion of the subject's GIT, thereby allowing to assess the condition of the mucosal disease in the at least portion of the GIT.