Dermoscopic Image Diagnosis via Content-Based Retrieval and Deep Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer-aided diagnosis (CAD) systems for skin cancer diagnosis are non-interactive and lack the ability to explain their decision-making process, making them less beneficial for dermatologists, and existing methods for skin lesion diagnosis are not sufficiently reliable or easy to use.
Innovation Solution
A decision support system (DSS) that uses deep feature extraction from dermoscopic images through transfer learning in Convolutional Neural Networks (CNNs) and ensemble techniques, including Logistic Regression and Support Vector Machine models, for classification and content-based image retrieval, providing an interactive interface for dermatologists to aid in diagnosing skin cancer types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional CAD systems are used for skin cancer diagnosis, then automation is improved, but the ability to explain decision-making and interactivity deteriorates
Solution Approach 1:
The patent introduces an intermediary retrieval-based decision support system that acts as a bridge between the automated CAD system and the dermatologist. This system retrieves and displays similar cases from a database, providing explanatory context for the automated diagnosis while maintaining automation benefits. The intermediary system translates automated predictions into interpretable case comparisons.
2Measurement precision
If deep learning methods are used to improve classification accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-training convolutional neural networks on large datasets before deploying them for specific skin cancer classification tasks. This transfer learning approach allows the system to achieve high classification accuracy without requiring complex task-specific training procedures, thereby reducing operational complexity while maintaining precision.
3Measurement precision
If more features are extracted from dermoscopic images to improve discrimination, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent extracts only the most discriminative features from dermoscopic images using pre-trained CNNs, rather than processing all possible image features. This selective extraction approach maintains high discriminative power for cancer classification while significantly reducing processing time by focusing computational resources on the most informative features.
Data Source
AI summary
Disclosed is a content-based image retrieval (CBIR) system and related methods that serve as a diagnostic aid for diagnosing whether a dermoscopic image correlates to a skin cancer type. Systems and methods according to aspects of the invention use as a reference a set of images of pathologically confirmed benign or malignant past cases from a collection of different classes that are of high similarity to the unknown new case in question, along with their diagnostic profiles. Systems and methods according to aspects of the invention predict what class of skin cancer is associated with a particular patient skin lesion, and may be employed as a diagnostic aid for general practitioners and dermatologists.