Hybrid ResNet-ViT Model for Lung Cancer Classification from Histopathology Images

Document Type

Conference Proceeding

Source of Publication

2024 25th International Arab Conference on Information Technology (ACIT)

Publication Date

12-12-2024

Abstract

Lung cancer remains one of the leading causes of cancer-related deaths globally. This study proposes a hybrid deep learning model that classifies lung cancer from histopathological images using Residual Networks (ResNet) and Vision Transformers (ViTs). The model uses ResNet for local feature extraction and ViT for global dependencies on the LC25000 dataset, which contains images of adenocarcinoma, benign tissue, and squamous cell carcinoma. The model achieved 99.31% classification accuracy after 10 epochs of training. Diagnostic precision and recall improved significantly, suggesting real-time clinical uses. Future work will explore model optimization, the inclusion of more lung disorders, and the development of multimodal data systems to further enhance diagnostic performance.

ISBN

979-8-3315-4001-2

Publisher

IEEE

Volume

00

First Page

1

Last Page

5

Disciplines

Medicine and Health Sciences

Keywords

Lung cancer, Histopathology, Deep learning, ResNet, Vision Transformers

Indexed in Scopus

no

Open Access

no

Link to Full Text

Share

COinS