Evaluating Lens Dose Reduction in Pediatric Neuroradiology Examinations Using Automated Kilovoltage Selection Software.


Journal Article

OBJECTIVE: The purpose of this study is to evaluate the potential of an automated kilo-voltage selection software for the reduction of lens dose in pediatric CT scans. MATERIALS AND METHODS: Two metal oxide semiconductor field effect transistor (MOSFET) detectors measured the lens dose in two anthropomorphic 1- and 5-year-old phantoms. These phantoms were scanned using a clinical pediatric brain protocol at 120 kVp as a control with the MDCT scanner. Scans were then repeated using automated kilovoltage software. The automated kilovoltage was set to operate at tube potentials of 120, 110, and 100 kVp. Dose savings were compared with the average lens dose of both eyes between automated kilovoltage and the control setting. Image quality was studied by contrast-to-noise ratios (CNRs) for each setting. RESULTS: The mean (± SD) lens dose from the routine brain scan without automated kilovoltage was 0.92 ± 0.03 cGy and 0.81 ± 0.03 cGy for the 1- and 5-year-old phantoms, respectively. Use of the automated kilovoltage software at 120 kVp, 110 kVp, and 100 kVp resulted in dose reductions of 9.8%, 17.4%, and 19.6%, respectively, for the 1-year-old phantom and 1.2%, 8.6%, and 17.3%, respectively, for the 5-year-old phantom. The CNR for all automated kilovoltage scans was within 11% of the control scans for the 1-year-old and within 6% for the 5-year-old phantom. CONCLUSION: Our results show that automated kilovoltage software is effective for reducing the radiation dose to the lens of the eye in pediatric patients. Furthermore, the image quality by CNR remained acceptable within 11% of the baseline for all kilovoltage settings used.

Full Text

Duke Authors

Cited Authors

  • Raudabaugh, J; Smith, AK; Moore, B; Ramirez-Giraldo, JC; Januzis, N; Yoshizumi, T

Published Date

  • September 2018

Published In

Volume / Issue

  • 211 / 3

Start / End Page

  • 635 - 640

PubMed ID

  • 29949420

Pubmed Central ID

  • 29949420

Electronic International Standard Serial Number (EISSN)

  • 1546-3141

Digital Object Identifier (DOI)

  • 10.2214/AJR.17.19089


  • eng

Conference Location

  • United States