Dual Determiner Neural Network for Lesion Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical image diagnosis systems face challenges in accurately distinguishing between serrated lesions (SSL) and hyperplastic polyps (HP) during endoscopic examinations, often leading to erroneous classifications due to similarities in appearance, which can result in misdiagnosis.
Innovation Solution
An image processing device employing a neural network with a first and second determiner, where the first determiner classifies lesions into categories of non-neoplasticity and neoplasticity, and the second determiner further differentiates between serrated lesions and hyperplastic polyps using intermediate feature amounts, outputting signals for accurate classification and display.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single determiner is used to classify lesions, then the device complexity is reduced, but the measurement precision of lesion type discrimination deteriorates
Solution Approach 1:
The patent divides the lesion classification task into two separate determiners: a first determiner that classifies lesions into broad categories (neoplastic vs. non-neoplastic) and a second determiner that specifically distinguishes between serrated lesions and hyperplastic polyps. This segmentation allows each determiner to specialize in specific classification tasks, improving overall measurement precision without requiring a single overly complex determiner to handle all classification challenges.
2Measurement precision
If detailed differentiation between serrated lesions and hyperplastic polyps is performed, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The classification system is segmented into two sequential determination processes. The first determiner handles the broad classification (neoplastic vs. non-neoplastic), while the second determiner specifically addresses the differentiation between serrated lesions and hyperplastic polyps. This segmentation allows detailed differentiation to be performed using a dedicated second determiner that is optimized for this specific task, rather than requiring the entire system to be overly complex.
Solution Approach 2:
The first determiner acts as an intermediary that performs preliminary classification before the second determiner performs detailed differentiation. By using the first determiner to filter and prepare the data, the second determiner can focus specifically on distinguishing between serrated lesions and hyperplastic polyps with higher precision, without the complexity of handling all possible lesion types simultaneously.
Data Source
AI summary
A processor of an image processing device acquires a medical image including a lesion region, and causes a determiner to perform a determination process of determining a type of a lesion on the basis of the medical image. A first determination process is a process of determining whether the type belongs to a first group including a first type classified as a first category and a second type classified as a second category or a second group including a third type classified as the second category. A second determination process is a process of determining whether the type is the first type or the second type. The processor outputs a first signal for specifying whether the type is classified as the first category or the second category on the basis of a determination result of the first determination process and a determination result of the second determination process.