Skip to main content
Fig. 5 | BMC Medical Imaging

Fig. 5

From: MedFusionGAN: multimodal medical image fusion using an unsupervised deep generative adversarial network

Fig. 5

Qualitative comparison of the proposed MedFusionGAN with eight state-of-the-art DL fusion methods are illustrated for three different image slices. Green arrows show the spatial distortion similar to the MRI motion artifacts. Blue arrows within the white boxes indicate the distinction between skull and scalp. Red boxes are served to zoom in to the cancerous regions to show the image contrasts of different traditional fusion methods. White arrows and yellow arrow indicate the regions that the CT bone information was not and with low spatial contrast delivered to the fusion images, respectively. Red horizontal lines indicate the location of the image profile shown in Fig. 8b

Back to article page