Figure 2 | Scientific Reports

Figure 2

From: Distinct tumor signatures using deep learning-based characterization of the peritumoral microenvironment in glioblastomas and brain metastases

Figure 2

The pipeline of our classifier: input to the classifier were patches (boxes) extracted from the free water volume fraction map in peritumoral area from both glioblastoma (red) and metastases (blue) which were used to train the CNN. In test phase, the results of CNN on patches were combined by majority voting to get the final label of metastasis or glioblastoma for each patient.

Back to article page