Skip to content
Open access

Improving Diagnostic Sensitivity in Imbalanced Oral Cancer Image Classification: A Comparative Study of CNN and Transformer Architectures

Sep 2026 · Cancers · Vol 18 · 0 citations · 42 references
Medicine

TL;DR

This study investigates whether class imbalance can be mitigated through random under-sampling, random over-sampling, and medically informed image augmentation when training three representative deep learning architectures: EfficientNet, Vision Transformer, and Swin Transformer.

Abstract

Simple Summary Early detection of oral cancer is essential for improving patient survival, yet artificial intelligence models often struggle because available clinical image datasets are small and highly imbalanced, with relatively few malignant cases. This study investigates whether class imbalance can be mitigated through random under-sampling, random over-sampling, and medically informed image augmentation when training three representative deep learning architectures: EfficientNet, Vision Transformer, and Swin Transformer. Using a publicly available dataset of oral cavity photographs and a unified five-fold cross-validation framework, we compared these strategies using both multiclass classification and a clinically oriented malignant-risk evaluation. The results indicate that data augmentation generally improves minority-class detection while maintaining competitive overall classification performance. Nevertheless, external multicenter validation and prospective clinical studies are required before routine clinical application.

Read PDF

Similar papers

Review Open access Sep 2026

Transformer-based deep learning models for lung cancer detection and classification: a comprehensive review

A review of transformer-based architectures applied to lung cancer detection and classification highlights potential future directions, including model optimization, data standardization, and validation across diverse clinical environments, to enhance the reliability and integration of transformer-based models in healt...

Neha Thakur, Pradeep Chouksey, P. Sadotra et al. · 0 citations
Open access Aug 2026

Analysis of CNN-Based Deep Learning Architectures and Performance Enhancement Strategies for Pneumonia Classification Using Chest X-Ray Images

These findings demonstrate that reliable medical AI systems require systematic optimization of preprocessing techniques, model architecture, data augmentation strategies, and clinically meaningful evaluation metrics rather than maximizing a single performance indicator.

YongJun Kim, Ji-Yeoun Lee · 0 citations
Open access 2026

Towards Transparent Decisions: CNN Ensemble with XAI-Driven Interpretations

A comprehensive deep learning-based framework that leverages three benchmark datasets—PH2, ISIC (Benign vs Malignant), and HAM10000—using transfer learning and ensemble techniques is presented, demonstrating a notable increase in classification accuracy.

A. D. Hayder, J. Saeed · 0 citations
#graph neural networks Open access Sep 2026

STLC-Net: Skin Transfer Learning Classification Framework for Multi-Class Skin Cancer Detection and Clinical Decision Support

STLC-Net (Skin Transfer Learning Classification Network) is introduced, a robust, zero-data-leakage framework for automated seven-class skin lesion classification using the benchmark HAM10000 dataset and integrated into an interactive web deployment for real-time, interpretable clinical decision support.

Gaurav Bhabhar · 0 citations
Review Open access Sep 2026

Recent Advancements in Artificial Intelligence for Breast Cancer: Image Augmentation, Segmentation, Diagnosis, and Prognosis Approaches

Breast cancer remains the most frequently diagnosed malignancy and a leading cause of cancer death among women worldwide, with GLOBOCAN 2022 estimating 2.30 million new cases and 666,103 deaths globally (Bray et al., 2024; Gu et al., 2026). Artificial intelligence (AI), and deep learning (DL) in particular, has become...

Nitish Verma, Ayushmaan Sharma, Kelvin Luntsi et al. · 0 citations
#explainable ai Open access Sep 2026

Explainable ensemble transfer learning for skin lesion classification with multi-method explainability validation

Quantitative evaluation using TMAI and complementary XAI methods provides additional evidence that the model's predictions are grounded in clinically relevant lesion regions rather than background artefacts, supporting the potential of the framework for accurate and interpretable automated skin lesion classification.

NE. Sravani, Srinivas Koppu · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.