Automated BSA Score Calculation for Psoriasis Lesions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for evaluating disease severity in skin diseases like Psoriasis are inexact and time-consuming, relying on human estimations and lacking objective, quantitative measures, especially for Guttate Psoriasis with numerous small inflammatory lesions.
Innovation Solution
An image processing method using the Felzenszwalb segmentation algorithm and Convolutional Neural Networks (CNNs) to automatically calculate the Body Surface Area (BSA) score, enhancing segmentation accuracy and objectivity by filtering out false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If human estimation methods are used to calculate BSA score, then the process is simple and quick, but the measurement precision is low and subject to human bias
Solution Approach 1:
The patent replaces the manual visual estimation process with an automated image processing system using Convolutional Neural Networks (CNNs). The system captures images of skin lesions and automatically calculates BSA scores through computational algorithms, eliminating human subjectivity and significantly improving measurement precision while reducing time consumption.
Solution Approach 2:
The patent creates a digital copy of the skin lesion assessment process through photographing and image processing. Instead of direct human visual estimation, the system captures images and processes them through CNNs to generate BSA scores, providing an objective quantitative measure that replicates and improves upon manual assessment.
2Productivity
If manual classification of lesions is performed, then the process is straightforward, but the productivity is low and measurement precision is coarse
Solution Approach 1:
The patent replaces manual lesion classification with automated CNN-based image recognition. The system processes multiple images simultaneously, rapidly identifying and classifying lesions with high precision, thereby dramatically increasing productivity while maintaining or improving measurement accuracy compared to manual methods.
Solution Approach 2:
The system enables self-service automated assessment where the CNN model independently performs lesion detection, classification, and BSA calculation without requiring manual intervention. The algorithm automatically processes images and generates results, making the assessment process both faster and more precise.
3Measurement precision
If visual inspection is used for Guttate Psoriasis with numerous small lesions, then the method is simple, but the measurement precision is insufficient due to difficulty in measuring by eyesight
Solution Approach 1:
The patent replaces visual inspection with automated image processing using CNNs. The system captures images of skin areas with numerous small Guttate Psoriasis lesions and uses computational algorithms to detect, segment, and count lesions that are too small and numerous for accurate visual assessment, thereby dramatically improving measurement precision.
Solution Approach 2:
The patent applies image segmentation techniques to divide the captured images into distinct regions, identifying individual lesions even when they are small and numerous. This segmentation approach allows the system to accurately measure each lesion and calculate the total BSA score, overcoming the limitations of visual inspection for Guttate Psoriasis.
Data Source
AI summary
An image processing method is provided that automatically calculates Body Surface Area (BSA) score using machine learning techniques. A Felzenszwalb image segmentation algorithm is used to define proposed regions in each of a plurality of training set images. The training set images are oversegmented, and then each of the proposed regions in each of the plurality of oversegmented training set images are manually classified as being a lesion or a non-lesion. A Convolutional Neural Network (CNN) is then trained using the manually classified proposed regions in each of the plurality of training set images. The trained CNN is then used on test images to calculate BSA scores.