Pylung: a supporting tool for comparative study of ViT and CNN-based models used for lung nodules classification

Marques, Felipe; Pestana, Pedro; Filipe, Vitor; Pestana, Pedro Duarte

http://hdl.handle.net/10400.2/20709

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
Pylung_pub.pdf		2.87 MB	Adobe PDF	Download

Send Feedback

Authors

Marques, Felipe

Pestana, Pedro

Filipe, Vitor

Pestana, Pedro Duarte

Abstract(s)

Lung cancer is a significant global health concern, and accu- rate classification of lung nodules plays a crucial role in its early detec- tion and treatment. This paper evaluates and compares the performance of Vision Transformer (ViT) and Convolutional Neural Network (CNN) models for lung nodule classification using the Pylung tool proposed in this work. The study aims to address the lack of research on ViT in lung nodule classification and proposes ViT as an alternative to CNN. The Lung Image Database Consortium and Image Database Resource Ini- tiative (LIDC-IDRI) dataset is utilized for training and evaluation. The Pylung tool is employed for dataset preprocessing and comparison of models. Three models, ViT, VGG16, and ResNet50, are analyzed, and their hyperparameters are optimized using Optuna. The results show that ViT achieves the highest accuracy (99.06%) in nodule classifica- tion compared to VGG16 (98.71%) and ResNet50 (98.46%). The study contributes by introducing ViT as a model for lung nodule classification, presenting the Pylung tool for model comparison, and suggesting further investigations to improve the accuracy.

Keywords

Lung Cancer ViT CNN Nodule Classification

URI

http://hdl.handle.net/10400.2/20709

Citation

Marques, F., Pestana, P. & Filipe, V. (2023). Pylung: a Supporting Tool for Comparative Study of ViT and CNN-based Models Used for Lung Nodules Classification. 23rd International Conference on Intelligent Systems Design and Applications (ISDA)