Detecção da Psoríase Utilizando Visão Computacional: Uma Abordagem Comparativa Entre CNNs e Vision Transformers
This work addresses automated detection of psoriasis for medical diagnosis, but it is incremental as it applies existing methods to a new medical dataset.
The paper compared CNNs and Vision Transformers for multi-class classification of psoriasis and similar lesion images, finding that Vision Transformers, particularly DaViT-B, achieved superior performance with an f1-score of 96.4% using smaller models.
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.