Settings
  • Light
  • Dark
  • Auto
Select Page

The aim of this study was to determine the value of MR-based preoperative nomograms in predicting DNA copy number (CN) subtype in patients with lower grade glioma (LGG). The authors achieved this by retrospectively analyzing the overall survival (OS) data of 170 subjects. They found that the shortest OS was observed in patients with CN2 subtype. This preliminary radiogenomics analysis showed that the MR-based preoperative nomograms provide individualized prediction of DNA copy number subtype it LGG patients.

Key points

  • This preliminary radiogenomics analysis of LGG revealed that the MR-based preoperative nomograms provide individualized prediction of DNA copy number subtype in LGG patients.
  • The AUC for the ROC curve was 0.781 for CN1 subtype, 0.854 for CN2 subtype, and 0.808 for CN3 subtype. Decision curve analysis supported that all nomograms were clinically useful.
  • The sensitivity was 0.779 (CN1), 0.731 (CN2), and 0.851 (CN3), respectively. The specificity was 0.664 (CN1), 0.872 (CN2), and 0.625 (CN3), respectively. And the accuracy was 0.717 (CN1), 0.849 (CN2), and 0.692 (CN3), respectively.

Article: Development of MR-based preoperative nomograms predicting DNA copy number subtype in lower grade gliomas with prognostic implication

Authors: Siwei Zhang, Shanshan Wu, Yun Wan, Yongsong Ye, Ying Zhang, Zelan Ma, Quanlan Guo, Hongdan Zhang & Li Xu

Latest posts

Welcome to the blog on Artificial Intelligence of the European Society of Radiology

This blog aims at bringing educational and critical perspectives on AI to readers. It should help imaging professionals to learn and keep up to date with the technologies being developed in this rapidly evolving field.