Establishment and verification of the dose-effect curve of dicentric chromosomes by semi-automatic analysis of human peripheral blood irradiated by X-ray in vitro

To establish and validate a dose-response curve for dicentric chromosomes (DC) induced by X-rays in human peripheral blood in vitro using semi-automated scoring. Peripheral blood samples were collected from three healthy individuals and exposed to X-ray doses of 0, 0.25, 0.5, 0.75, 1, 2, 3, 4, and 5 Gy at a dose rate of 1.158 Gy/min. Dicentric chromosomes in metaphase were scored both full- and semi-automatically, and a dose-response curve was generated with CABAS software based on dicentric yields. Dose estimations were then performed for 12 biodosimetry standard samples and one sample from a patient with X-ray-induced skin injury. Dicentric yields increased with X-ray doses from 0 to 5 Gy (r = 0.943, P < 0.01). The dose-response association followed a linear-quadratic model: Y = 0.0002 (± 0.0001) + 0.0379 (± 0.0032) × D + 0.0253 (± 0.0014) × D2 (R2 = 0.998, P < 0.01), where Y represents dicentric yields and D is the dose. The estimated dose for 12 validation samples aligned closely with actual doses. The estimated whole-body average absorbed dose for the patient was 0.73 Gy, with observed DC over-dispersion suggesting partial body exposure. By using the Dolphin model, this dose was refined to 2.22 Gy and estimated irradiated body volume was 35.94%, consistent with clinical diagnosis. The dose-response curve developed using semi-automated scoring offer a reliable and efficient approach for dose estimation and clinical diagnosis in nuclear radiation emergencies. It could also support retrospective biodosimetry of partial-body, non-uniform radiation exposure.