Skip to main content
. 2021 Jun 8;13(12):2866. doi: 10.3390/cancers13122866

Figure 1.

Figure 1

Deep learning strategy: DenseNet 161 architecture for tumor grading in MRI [44]. The network receives the 2D transversal slice from the VOI and outputs the probability of the image for the two classes. In the lower part of the figure, each component of the DenseNet is described. Abbreviations: Avg: Average, T1FSGd: contrast-enhanced and fat-saturated T1-weighted sequence, T2-weighted fat-saturated (T2FS) sequence.