Research Paper Volume 13, Issue 9 pp 12833—12848

Predicting intraventricular hemorrhage growth with a machine learning-based, radiomics-clinical model

(A) Regions of interest were manually segmented. (B) A total of 396 features were extracted. (C) Features were selected using LASSO method. (D) Rad-score was calculated. (E) Predicting model was developed using support vector machine.

Figure 6. (A) Regions of interest were manually segmented. (B) A total of 396 features were extracted. (C) Features were selected using LASSO method. (D) Rad-score was calculated. (E) Predicting model was developed using support vector machine.