Open Set Recognition for Endoscopic Image Classification: A Deep Learning Approach on the Kvasir Dataset

2025-06-23

AI Summary

To address the challenge of identifying unknown pathologies in open-world clinical endoscopy—where conventional closed-set classifiers fail—this work pioneers the systematic application of open set recognition (OSR) to medical imaging, establishing the first OSR benchmark for endoscopic images. Leveraging the Kvasir dataset, we comparatively evaluate ResNet-50, Swin Transformer, and a novel ResNet–Transformer hybrid architecture under both closed-set and open-set settings, with OpenMax as the unified baseline method. Experimental results demonstrate that our approach significantly improves discrimination between known and unknown lesions, thereby enhancing model reliability for clinical deployment. The hybrid architecture achieves superior OSR performance, outperforming both pure CNN and pure transformer baselines. This study establishes the first standardized OSR benchmark for endoscopic image analysis, providing a reproducible technical framework and empirical foundation to advance safe, real-world deployment of AI-assisted endoscopic diagnosis.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Data transparency and provenance
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Endoscopic image classification plays a pivotal role in medical diagnostics by identifying anatomical landmarks and pathological findings. However, conventional closed-set classification frameworks are inherently limited in open-world clinical settings, where previously unseen conditions can arise andcompromise model reliability. To address this, we explore the application of Open Set Recognition (OSR) techniques on the Kvasir dataset, a publicly available and diverse endoscopic image collection. In this study, we evaluate and compare the OSR capabilities of several representative deep learning architectures, including ResNet-50, Swin Transformer, and a hybrid ResNet-Transformer model, under both closed-set and open-set conditions. OpenMax is adopted as a baseline OSR method to assess the ability of these models to distinguish known classes from previously unseen categories. This work represents one of the first efforts to apply open set recognition to the Kvasir dataset and provides a foundational benchmark for evaluating OSR performance in medical image analysis. Our results offer practical insights into model behavior in clinically realistic settings and highlight the importance of OSR techniques for the safe deployment of AI systems in endoscopy.

Problem

• Addressing open-world clinical challenges in endoscopic image classification
• Evaluating deep learning models for Open Set Recognition on Kvasir dataset
• Benchmarking OSR performance to enhance AI reliability in endoscopy

Innovation

• Open Set Recognition for endoscopic image classification
• Deep learning models like ResNet-50 and Swin Transformer
• OpenMax baseline for distinguishing unseen categories