바로가기  >

Abstract
Radiation dose estimations performed by automated counting of micronuclei (MN) have been studied for their utility for triage following large-scale radiological incidents; although speed is essential, it also is essential to estimate radiation doses as accurately as possible for long-term epidemiological follow-up. Our goal in this study was to evaluate and improve the performance of automated MN counting for biodosimetry using the cytokinesis-block micronucleus (CBMN) assay. We measured false detection rates and used them to improve the accuracy of dosimetry. The average false-positive rate for binucleated cells was 1.14%; average false-positive and -negative MN rates were 1.03% and 3.50%, respectively. Detection errors seemed to be correlated with radiation dose. Correction of errors by visual inspection of images used for automated counting, called the semi-automated and manual scoring method, increased accuracy of dose estimation. Our findings suggest that dose assessment of the automated MN scoring system can be improved by subsequent error correction, which could be useful for performing biodosimetry on large numbers of people rapidly, accurately, and efficiently.

Affiliations

Younghyun Lee 1 2, Young Woo Jin 3, Ki Moon Seong 1, Ruth C Wilkins 4, Seongjae Jang 5
1Laboratory of Biological Dosimetry, Korea Institute of Radiological and Medical Sciences, National Radiation Emergency Medical Center, Seoul, Republic of Korea.
2Department of Biomedical Laboratory Science, College of Medical Sciences, Soonchunhyang University, Asan, Republic of Korea.
3Korea Institute of Radiological and Medical Sciences, National Radiation Emergency Medical Center, Seoul, Republic of Korea.
4Consumer and Clinical Radiation Protection Bureau, Health Canada, Ottawa, ON, Canada.
5Laboratory of Biological Dosimetry, Korea Institute of Radiological and Medical Sciences, National Radiation Emergency Medical Center, Seoul, Republic of Korea. sjsjj@kirams.re.kr.