Comparison of Supervised and Unsupervised Deep Learning Methods for Medical Image Synthesis between Computed Tomography and Magnetic Resonance Images

<div>(a) MAE of sMR images for all the 6 patients in test set and their average value. (b) MAE of sCT images generated for all the 6 patients in test set and their average value. Both the blue columns denoted the U-Net method, and the red columns represented the CycleGAN method.</div>

BioMed Research International

fig5

Figure 5

Figure 5: Comparison of Supervised and Unsupervised Deep Learning Methods for Medical Image Synthesis between Computed Tomography and Magnetic Resonance Images 

Figure 5 | Comparison of Supervised and Unsupervised Deep Learning Methods for Medical Image Synthesis between Computed Tomography and Magnetic Resonance Images 