Skip to main content
Fig. 1 | BMC Pediatrics

Fig. 1

From: Identification and validation of radiomic features from computed tomography for preoperative classification of neuroblastic tumors in children

Fig. 1

Workflow in this study. Tumors in portal venous–phase contrast-enhanced computed tomography images were manually contoured by an experienced radiologist, and radiomic features were analyzed within the tumor region. Variance analysis and the maximum relevance–minimum redundancy (MRMR) algorithm were used to selected the top 10 diagnostic features. These features were then used to develop a classifier based on least absolute shrinkage and selection operator (LASSO) regression

Back to article page

BMC Pediatrics

ISSN: 1471-2431

Contact us