Differential Marker Generation for Skin Singularity Evolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing skin singularities, such as those used by dermatologists, lack comprehensive full-body imaging and skin maps that provide access to dermoscopy images at any point, making it difficult to monitor the evolution of skin singularities over time.
Innovation Solution
A computer-implemented method that generates differential markers for skin singularities by comparing dermoscopic images of the same body part taken at different times, using image processing and machine learning techniques to align and merge the images, and calculate evolution criteria for singularity descriptors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dermoscopic images are taken individually using a hand-held dermatoscope, then dermoscopic resolution is achieved for specific lesions, but full-body imaging with comprehensive skin mapping is not available
Solution Approach 1:
The body surface is divided into multiple image zones captured by a grid of dermoscopic probes. Each probe captures a localized high-resolution image, and these segmented images are then computationally assembled into a complete full-body skin map, resolving the contradiction between local precision and global coverage
Solution Approach 2:
Multiple levels of imaging are nested within a unified system: individual lesion images are nested within body region images, which are nested within the complete full-body skin map. This hierarchical nesting allows dermoscopic resolution at the lesion level while maintaining comprehensive body surface coverage
2Loss of time
If images are taken at different times to monitor evolution of skin singularities, then temporal monitoring capability is achieved, but alignment and association of images with body areas becomes complex and error-prone
Solution Approach 1:
The system captures temporal feedback by comparing skin singularity characteristics across multiple time points. Automated image registration algorithms provide feedback on alignment accuracy, and the system iteratively adjusts transformations to maintain precise correspondence of body areas across different acquisition times, reducing manual intervention
Solution Approach 2:
Virtual copies of the body surface map are created for each time point, allowing simultaneous display and comparison of multiple temporal snapshots. These digital copies can be overlaid, annotated, and compared without affecting the original images, simplifying the complexity of managing multiple time-point datasets
3Loss of information
If manual association of dermoscopic images with body areas is performed, then image labeling is achieved, but automation and efficiency are reduced
Solution Approach 1:
The system performs self-service through automated image processing pipelines that independently complete tasks previously requiring manual intervention. Algorithms automatically detect skin singularities, extract features, register images to body maps, and generate diagnostic reports, maintaining high association accuracy while dramatically increasing processing throughput
Solution Approach 2:
Manual mechanical processes of image labeling and body area mapping are replaced with automated computational systems. Machine learning models and computer vision algorithms substitute for manual annotation, performing image-to-body-area association with both high accuracy and automated efficiency
Data Source
AI summary
A method for generating at least one differential marker of the presence of a skin singularity of a human body, the method including acquisition of a first and a second set of dermoscopic images of singularities of the skin of a human body of a first individual at a first and respectively a second date; generation of a first and a second representation of a first image of a part of the human body and of a first symbol respectively a second symbol superimposed on the first image of each representation at a position in a first reference frame of the first image, the geometry and/or the color of the second symbol being different from the geometry and/or the color of the first symbol when the second class of the dermoscopic image of the second set is different from the first class of the dermoscopic image of the first set.