Radiomic Features in IOERT of Unilateral Breast Cancer as a Biological Dosimetry

Document Type : Original Paper

Authors

1 Department of Medical Physics, School of Medicine, Iran University of Medical Sciences, Tehran, Iran

2 Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran

3 Department of Radiation Physics, Division of Radiation Oncology, University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

4 Pharmaceutical Research Center, Mashhad University of Medical Sciences, Mashhad, Iran

5 Department of General Surgery, School of Medicine, Iran University of Medical Sciences, Tehran, Iran

Abstract

Introduction: In the current study, we used CT texture analysis of irradiated breast compared to the contralateral breast after 12 Gy boost radiation dose in IOERT to obtain radiation-sensitive indicators with achieving parameters utilized as the biological markers or the biological dosimeters.

Methods: Chest CT scan series of 35 patients (with unilateral ductal carcinoma in situ (DCIS) who had undergone boost IOERT) were used in this study. A total of 259 CT radiomic features (including first-order features, textural features, gradient features, and autoregressive model-based features) were extracted using MaZda software. The features that were significantly different in the two breasts were selected. A score was assigned to each of the features and the highest scores were characterized (according to the level of significant differences). The feature selection process was performed using the hybrid feature selection method.

Results: CT Texture analysis indicated that radiation dose causes significant changes in some radiomic features of the breast tissue.

Conclusion: With more research in the future, we can fit the Delta-radiomics values with the received radiation dose and achieve a biological dosimeter to detect low-dose radiation.

Keywords

Main Subjects



Articles in Press, Accepted Manuscript
Available Online from 06 April 2022
  • Receive Date: 19 November 2021
  • Revise Date: 03 April 2022
  • Accept Date: 06 April 2022
  • First Publish Date: 06 April 2022