SGPNet: A Three-Dimensional Multitask Residual Framework for Segmentation and IDH Genotype Prediction of Gliomas

<div>The ROC curves of the SGPNet across different groups of learning targets, which indicate that the SGPNet performs best when the ground-truth segmentation labels are provided as the learning target.</div>

Computational Intelligence and Neuroscience

fig4

Figure 4

Figure 4: SGPNet: A Three-Dimensional Multitask Residual Framework for Segmentation and IDH Genotype Prediction of Gliomas 