Skip to main content
. Author manuscript; available in PMC: 2020 Oct 30.
Published in final edited form as: Neuroimage. 2020 Jun 27;220:117081. doi: 10.1016/j.neuroimage.2020.117081

Fig. 13.

Fig. 13.

Quantitative (average Sensitivity) evaluation of various DL network architectures. From top to bottom rows we see results on the data from (a) UPenn, (b) TJU, and (c) MDA. The evaluated models in this figure include training on individual modalities and their ensemble using majority voting, as well as multi-modality training.