Saved in:
Bibliographic Details
Main Authors: Lucena, Natanael, da Silva, Fábio S., Rios, Ricardo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.10119
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909646276526080
author Lucena, Natanael
da Silva, Fábio S.
Rios, Ricardo
author_facet Lucena, Natanael
da Silva, Fábio S.
Rios, Ricardo
contents This paper presents a comparison of the performance of Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) in the task of multi-classifying images containing lesions of psoriasis and diseases similar to it. Models pre-trained on ImageNet were adapted to a specific data set. Both achieved high predictive metrics, but the ViTs stood out for their superior performance with smaller models. Dual Attention Vision Transformer-Base (DaViT-B) obtained the best results, with an f1-score of 96.4%, and is recommended as the most efficient architecture for automated psoriasis detection. This article reinforces the potential of ViTs for medical image classification tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2506_10119
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Detecção da Psoríase Utilizando Visão Computacional: Uma Abordagem Comparativa Entre CNNs e Vision Transformers
Lucena, Natanael
da Silva, Fábio S.
Rios, Ricardo
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
This paper presents a comparison of the performance of Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) in the task of multi-classifying images containing lesions of psoriasis and diseases similar to it. Models pre-trained on ImageNet were adapted to a specific data set. Both achieved high predictive metrics, but the ViTs stood out for their superior performance with smaller models. Dual Attention Vision Transformer-Base (DaViT-B) obtained the best results, with an f1-score of 96.4%, and is recommended as the most efficient architecture for automated psoriasis detection. This article reinforces the potential of ViTs for medical image classification tasks.
title Detecção da Psoríase Utilizando Visão Computacional: Uma Abordagem Comparativa Entre CNNs e Vision Transformers
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2506.10119