CNN-based Local Vision Transformer for COVID-19 Diagnosis
This work addresses the need for accurate COVID-19 diagnosis tools for medical professionals, but it is incremental as it builds on existing Vision Transformer and CNN methods.
The paper tackled the problem of limited feature richness and training difficulty in Vision Transformers for COVID-19 diagnosis by proposing a CNN-based local Vision Transformer structure, achieving improved performance on small COVID-19 datasets and ImageNet.
Deep learning technology can be used as an assistive technology to help doctors quickly and accurately identify COVID-19 infections. Recently, Vision Transformer (ViT) has shown great potential towards image classification due to its global receptive field. However, due to the lack of inductive biases inherent to CNNs, the ViT-based structure leads to limited feature richness and difficulty in model training. In this paper, we propose a new structure called Transformer for COVID-19 (COVT) to improve the performance of ViT-based architectures on small COVID-19 datasets. It uses CNN as a feature extractor to effectively extract local structural information, and introduces average pooling to ViT's Multilayer Perception(MLP) module for global information. Experiments show the effectiveness of our method on the two COVID-19 datasets and the ImageNet dataset.