Lightweight Relational Embedding in Task-Interpolated Few-Shot Networks for Enhanced Gastrointestinal Disease Classification

2024-06-25

Conference on Algebraic Informatics

AI Summary

To address the challenges of poor endoscopic image quality, scarce annotated samples, and high inter-frame similarity in early colorectal cancer (CRC) screening—leading to suboptimal diagnostic accuracy—this paper proposes a lightweight few-shot learning framework. Methodologically, we introduce a novel task interpolation strategy to enhance multi-view image generation; design a lightweight relational embedding module to jointly model intra-frame salient features and inter-frame dynamic evolution; and incorporate a two-level routing attention mechanism to strengthen focus on lesion regions. The model employs a lightweight CNN backbone, integrating few-shot learning, task interpolation, relational embedding, and two-level routing attention. Evaluated on the Kvasir dataset, it achieves 90.1% accuracy, 0.845 precision, 0.942 recall, and 0.891 F1-score—significantly surpassing current state-of-the-art methods.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Traditional diagnostic methods like colonoscopy are invasive yet critical tools necessary for accurately diagnosing colorectal cancer (CRC). Detection of CRC at early stages is crucial for increasing patient survival rates. However, colonoscopy is dependent on obtaining adequate and high-quality endoscopic images. Prolonged invasive procedures are inherently risky for patients, while suboptimal or insufficient images hamper diagnostic accuracy. These images, typically derived from video frames, often exhibit similar patterns, posing challenges in discrimination. To overcome these challenges, we propose a novel Deep Learning network built on a Few-Shot Learning architecture, which includes a tailored feature extractor, task interpolation, relational embedding, and a bi-level routing attention mechanism. The Few-Shot Learning paradigm enables our model to rapidly adapt to unseen fine-grained endoscopic image patterns, and the task interpolation augments the insufficient images artificially from varied instrument viewpoints. Our relational embedding approach discerns critical intra-image features and captures inter-image transitions between consecutive endoscopic frames, overcoming the limitations of Convolutional Neural Networks (CNNs). The integration of a light-weight attention mechanism ensures a concentrated analysis of pertinent image regions. By training on diverse datasets, the model’s generalizability and robustness are notably improved for handling endoscopic images. Evaluated on Kvasir dataset, our model demonstrated superior performance, achieving an accuracy of 90.1%, precision of 0.845, recall of 0.942, and an F1 score of 0.891. This surpasses current state-of-the-art methods, presenting a promising solution to the challenges of invasive colonoscopy by optimizing CRC detection through advanced image analysis.

Problem

• Improving early colorectal cancer detection using endoscopic images
• Overcoming limitations of invasive colonoscopy with Few-Shot Learning
• Enhancing diagnostic accuracy via lightweight relational embedding

Innovation

• Few-Shot Learning for rapid adaptation
• Task interpolation augments insufficient images
• Light-weight attention mechanism improves focus