Wound Assessment Imaging With Color Calibration and Tissue Classification
Solution Overview
Problem
Current wound assessment methods rely heavily on subjective human judgment, leading to delayed and inaccurate diagnosis and treatment, particularly for complex wounds like those in diabetic patients, which can result in severe complications and amputation risks.
Innovation Solution
A wound assessment method and system utilizing an L-shaped color calibration card for image preprocessing, including distortion correction and color calibration, followed by image segmentation and tissue classification into granulation, slough, and eschar tissues, using machine learning models like FPN for precise wound evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If human doctors manually assess and diagnose wounds, then diagnostic flexibility and clinical judgment are improved, but diagnostic time increases and measurement precision deteriorates
Solution Approach 1:
The patent replaces the mechanical system of manual visual inspection and subjective judgment by doctors with an automated image processing system. The system uses computer vision algorithms to automatically detect, segment, and classify wound tissues, eliminating the need for manual measurement and assessment while achieving consistent and precise results.
Solution Approach 2:
The patent creates a digital copy of the wound through image capture and processing. Instead of relying on direct manual examination, the system uses captured images as representations of the wound, processes these digital copies through algorithms, and derives assessment results from the processed images, thereby separating the assessment process from manual intervention.
2Reliability
If multiple doctors subjectively assess wounds, then diagnostic coverage is improved, but diagnostic consistency deteriorates due to subjective opinions
Solution Approach 1:
The patent transforms the subjective assessment parameters into objective digital parameters. Instead of relying on doctors' subjective interpretations, the system uses quantifiable image parameters such as color values, texture features, and spatial distributions to consistently characterize wound tissues. This parameter transformation ensures reproducible and consistent assessment results.
Solution Approach 2:
The system performs self-assessment without requiring multiple doctors. The automated image processing system independently completes the entire assessment process from image capture to tissue classification, eliminating the need for multiple human assessors and their varying subjective opinions, thereby ensuring consistent results.
3Measurement precision
If manual wound classification is performed, then clinical judgment flexibility is improved, but measurement precision and objectivity deteriorate
Solution Approach 1:
The patent segments the wound image into distinct tissue regions based on visual characteristics. The system divides the wound area into different tissue types (e.g., granulation tissue, slough, eschar) by analyzing color, texture, and morphological features, enabling precise classification while maintaining operational simplicity through automated segmentation algorithms.
Data Source
AI summary
A wound assessment method is provided. The wound assessment method includes receiving an image that includes an L-shaped color calibration card and a wound; performing an image preprocessing on the image, based on the L-shaped color calibration card, to obtain an adjusted image, where the image preprocessing includes a distortion correction and a color calibration; obtaining a wound image excluding the L-shaped color calibration card, based on the adjusted image, where the wound image includes a plurality of regions; and classifying each of the plurality of regions of the wound image into one of wound tissue types, to obtain a wound tissue type distribution of the wound, where the wound tissue types include a granulation tissue, a slough tissue, and an eschar tissue.