Cet article est une Preprint
Les preprints sont des rapports de recherche préliminaires qui n'ont pas été certifiés par l’évaluation par les pairs. Ils ne devraient pas être considérés comme guidant la pratique clinique ou les comportements liés à la santé et ne devraient pas être rapportés dans les médias comme des informations établies.
Les preprints publiées en ligne permettent aux auteurs de recevoir des commentaires rapidement, et toute la communauté scientifique peut évaluer indépendamment le travail et répondre en conséquence. Ces commentaires sont publiés avec les preprints que quiconque peut lire et servir d’évaluation post-publication.
HiFuse: Hierarchical Multi-Scale Feature Fusion Network for Medical Image Classification (preprint)
arxiv; 2022.
Preprint
Dans Anglais
| PREPRINT-ARXIV | ID: ppzbmed-2209.10218v1
ABSTRACT
Medical image classification has developed rapidly under the impetus of the convolutional neural network (CNN). Due to the fixed size of the receptive field of the convolution kernel, it is difficult to capture the global features of medical images. Although the self-attention-based Transformer can model long-range dependencies, it has high computational complexity and lacks local inductive bias. Much research has demonstrated that global and local features are crucial for image classification. However, medical images have a lot of noisy, scattered features, intra-class variation, and inter-class similarities. This paper proposes a three-branch hierarchical multi-scale feature fusion network structure termed as HiFuse for medical image classification as a new method. It can fuse the advantages of Transformer and CNN from multi-scale hierarchies without destroying the respective modeling so as to improve the classification accuracy of various medical images. A parallel hierarchy of local and global feature blocks is designed to efficiently extract local features and global representations at various semantic scales, with the flexibility to model at different scales and linear computational complexity relevant to image size. Moreover, an adaptive hierarchical feature fusion block (HFF block) is designed to utilize the features obtained at different hierarchical levels comprehensively. The HFF block contains spatial attention, channel attention, residual inverted MLP, and shortcut to adaptively fuse semantic information between various scale features of each branch. The accuracy of our proposed model on the ISIC2018 dataset is 7.6% higher than baseline, 21.5% on the Covid-19 dataset, and 10.4% on the Kvasir dataset. Compared with other advanced models, the HiFuse model performs the best. Our code is open-source and available from https//github.com/huoxiangzuo/HiFuse.
Texte intégral:
Disponible
Collection:
Preprints
Base de données:
PREPRINT-ARXIV
Sujet Principal:
COVID-19
langue:
Anglais
Année:
2022
Type de document:
Preprint
Documents relatifs à ce sujet
MEDLINE
...
LILACS
LIS