UGCANet: A Unified Global Context-Aware Transformer-based Network with Feature Alignment for Endoscopic Image Analysis

TL;DR

This paper introduces UGCANet, a Transformer-based neural network with feature alignment and global context modules, designed to improve the accuracy of endoscopic image analysis for detecting gastrointestinal lesions and polyps.

Contribution

It presents a novel multi-task deep neural network architecture that enhances endoscopic diagnosis by integrating global context awareness and feature alignment.

Findings

Outperforms existing methods on endoscopic image datasets

Achieves higher detection accuracy for gastrointestinal lesions and polyps

Demonstrates robustness across multiple diagnostic tasks

Abstract

Gastrointestinal endoscopy is a medical procedure that utilizes a flexible tube equipped with a camera and other instruments to examine the digestive tract. This minimally invasive technique allows for diagnosing and managing various gastrointestinal conditions, including inflammatory bowel disease, gastrointestinal bleeding, and colon cancer. The early detection and identification of lesions in the upper gastrointestinal tract and the identification of malignant polyps that may pose a risk of cancer development are critical components of gastrointestinal endoscopy's diagnostic and therapeutic applications. Therefore, enhancing the detection rates of gastrointestinal disorders can significantly improve a patient's prognosis by increasing the likelihood of timely medical intervention, which may prolong the patient's lifespan and improve overall health outcomes. This paper presents a novel Transformer-based deep neural network designed to perform multiple tasks simultaneously, thereby enabling accurate identification of both upper gastrointestinal tract lesions and colon polyps. Our approach proposes a unique global context-aware module and leverages the powerful MiT backbone, along with a feature alignment block, to enhance the network's representation capability. This novel design leads to a significant improvement in performance across various endoscopic diagnosis tasks. Extensive experiments demonstrate the superior performance of our method compared to other state-of-the-art approaches.