Radiomics strategies for risk assessment of tumour failure in head-and-neck cancer.

On the other hand, the best radiomic model for distant metastases was found to possess highly significant predictive properties in the testing set (Table 1 : AUC of 0.86), and is composed of three metrics extracted from CT images ( LRHGE GLRLM : long run high gray level emphasis GLRLM , ZSV GLSZM : zone size variance GLSZM and ZSN GLSZM : zone size nonuifomity GLSZM ).

also in 132,142 other papers