Fig. 1 | Scientific Reports

Fig. 1

From: Machine learning method based on radiomics help differentiate posterior pituitary tumors from pituitary neuroendocrine tumors and craniopharyngioma

Fig. 1

Workflow diagram. First, manual segmentation was performed on each patient’s contrast-enhanced (CE) T1-weighted images. Then, the radiomics features was extracted. Finally, the data of training and testing cohort were including in a nested cross-validation to build machine learning models which was used to validated in the validation cohort.

Back to article page