Skip to main content
. 2022 May 26;11(11):3013. doi: 10.3390/jcm11113013

Figure 1.

Figure 1

Illustration of our proposed ViT model for COVID-19 detection using CXR images. The input image is split into equal-sized patches and embedded using linear projection. Position embeddings are added, and the resulting sequence is fed to a Transformer encoder.