Skin Lesion Measurement Using Correlated Ruler Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dermoscopes suffer from parallax errors in measuring skin lesions due to fixed rulers etched on the lens, leading to inaccurate and inconsistent size measurements, which are critical for monitoring and diagnosing skin conditions like melanoma.
Innovation Solution
A compute system utilizing a pixel-to-millimeter algorithm and AI-based modules for accurate skin lesion measurement, including a ruler segmentation model, skin lesion segmentation model, and area calculation, to generate a correlated ruler and precise skin lesion area report.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a fixed ruler is etched on the dermoscope lens, then the device structure is simplified and easier to manufacture, but parallax errors occur leading to inaccurate measurements
Solution Approach 1:
The patent replaces the mechanical fixed ruler system with a digital image processing system. Instead of relying on physical rulers etched on the lens, the system captures images and uses software algorithms to detect, segment, and measure skin lesions, eliminating parallax errors while maintaining ease of manufacture through digital rather than mechanical solutions.
Solution Approach 2:
The patent introduces an intermediary digital processing layer between the dermoscope lens and the measurement output. The fixed ruler images are captured and then processed through multiple intermediary steps including image segmentation, contour detection, and area calculation algorithms, which mediate the measurement process to eliminate direct parallax errors.
2Device complexity
If manual measurement methods are used, then device complexity is reduced, but measurement consistency and reproducibility deteriorate
Solution Approach 1:
The patent implements a self-service measurement system where the software automatically detects, segments, and measures skin lesions without requiring manual intervention. The system performs self-calibration and self-measurement through automated algorithms, ensuring consistent and reproducible results while maintaining relatively simple device architecture.
Solution Approach 2:
The patent incorporates feedback mechanisms in the form of automated validation and verification steps within the image processing pipeline. The system provides feedback loops that verify measurement accuracy, adjust parameters, and ensure reproducibility, all while maintaining a straightforward device structure that doesn't require complex additional hardware.
3Device complexity
If fixed rulers are used for measurement, then the device design is simplified, but measurement accuracy deteriorates due to parallax errors
Solution Approach 1:
The patent replaces the mechanical fixed ruler system with a digital image processing system. Instead of relying on physical rulers etched on the lens, the system captures images and uses software algorithms to detect, segment, and measure skin lesions, eliminating parallax errors while maintaining ease of manufacture through digital rather than mechanical solutions.
Solution Approach 2:
The patent changes the measurement parameter approach from direct physical ruler reading to digital pixel-based area calculation. By converting the measurement from linear ruler scale to digital image pixel area, the system eliminates parallax errors inherent in fixed ruler systems while maintaining simple device design.
Data Source
AI summary
A method of operation of a compute system includes: receiving a patient image including a skin lesion and a fixed ruler, generating a correlated ruler by segmenting the fixed ruler including translating pixel-to-millimeter to the patient image, calculating a skin lesion area by segmenting the skin lesion, and generating a skin lesion output report, including an image of the skin lesion, the correlated ruler and the skin lesion area, for displaying on a device.