RENEB Inter-Laboratory Comparison 2021: Inter-Assay Comparison of Eight Dosimetry Assays
Bibliographic record
Abstract
Tools for radiation exposure reconstruction are required to support the medical management of radiation victims in radiological or nuclear incidents. Different biological and physical dosimetry assays can be used for various exposure scenarios to estimate the dose of ionizing radiation a person has absorbed. Regular validation of the techniques through inter-laboratory comparisons (ILC) is essential to guarantee high quality results. In the current RENEB inter-laboratory comparison, the performance quality of established cytogenetic assays [dicentric chromosome assay (DCA), cytokinesis-block micronucleus assay (CBMN), stable chromosomal translocation assay (FISH) and premature chromosome condensation assay (PCC)] was tested in comparison to molecular biological assays [gamma-H2AX foci (gH2AX), gene expression (GE)] and physical dosimetry-based assays [electron paramagnetic resonance (EPR), optically or thermally stimulated luminescence (LUM)]. Three blinded coded samples (e.g., blood, enamel or mobiles) were exposed to 0, 1.2 or 3.5 Gy X-ray reference doses (240 kVp, 1 Gy/min). These doses roughly correspond to clinically relevant groups of unexposed to low exposed (0-1 Gy), moderately exposed (1-2 Gy, no severe acute health effects expected) and highly exposed individuals (>2 Gy, requiring early intensive medical care). In the frame of the current RENEB inter-laboratory comparison, samples were sent to 86 specialized teams in 46 organizations from 27 nations for dose estimation and identification of three clinically relevant groups. The time for sending early crude reports and more precise reports was documented for each laboratory and assay where possible. The quality of dose estimates was analyzed with three different levels of granularity, 1. by calculating the frequency of correctly reported clinically relevant dose categories, 2. by determining the number of dose estimates within the uncertainty intervals recommended for triage dosimetry (±0.5 Gy or ±1.0 Gy for doses <2.5 Gy or >2.5 Gy), and 3. by calculating the absolute difference (AD) of estimated doses relative to the reference doses. In total, 554 dose estimates were submitted within the 6-week period given before the exercise was closed. For samples processed with the highest priority, earliest dose estimates/categories were reported within 5-10 h of receipt for GE, gH2AX, LUM, EPR, 2-3 days for DCA, CBMN and within 6-7 days for the FISH assay. For the unirradiated control sample, the categorization in the correct clinically relevant group (0-1 Gy) as well as the allocation to the triage uncertainty interval was, with the exception of a few outliers, successfully performed for all assays. For the 3.5 Gy sample the percentage of correct classifications to the clinically relevant group (≥2 Gy) was between 89-100% for all assays, with the exception of gH2AX. For the 1.2 Gy sample, an exact allocation to the clinically relevant group was more difficult and 0-50% or 0-48% of the estimates were wrongly classified into the lowest or highest dose categories, respectively. For the irradiated samples, the correct allocation to the triage uncertainty intervals varied considerably between assays for the 1.2 Gy (29-76%) and 3.5 Gy (17-100%) samples. While a systematic shift towards higher doses was observed for the cytogenetic-based assays, extreme outliers exceeding the reference doses 2-6 fold were observed for EPR, FISH and GE assays. These outliers were related to a particular material examined (tooth enamel for EPR assay, reported as kerma in enamel, but when converted into the proper quantity, i.e. to kerma in air, expected dose estimates could be recalculated in most cases), the level of experience of the teams (FISH) and methodological uncertainties (GE). This was the first RENEB ILC where everything, from blood sampling to irradiation and shipment of the samples, was organized and realized at the same institution, for several biological and physical retrospective dosimetry assays. Almost all assays appeared comparably applicable for the identification of unexposed and highly exposed individuals and the allocation of medical relevant groups, with the latter requiring medical support for the acute radiation scenario simulated in this exercise. However, extreme outliers or a systematic shift of dose estimates have been observed for some assays. Possible reasons will be discussed in the assay specific papers of this special issue. In summary, this ILC clearly demonstrates the need to conduct regular exercises to identify research needs, but also to identify technical problems and to optimize the design of future ILCs.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.135 | 0.113 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.003 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.005 | 0.002 |
| Open science | 0.004 | 0.006 |
| Research integrity | 0.003 | 0.002 |
| Insufficient payload (model declined to judge) | 0.012 | 0.007 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".