Scholarlink

<p>When internal doses are calculated for the purpose of epidemiology studies, it is important that an assessment of the uncertainty in these doses is performed. This assessment should include both uncertainties in measured quantities and those in the various parameter values used in the calculation. In the Mayak Worker dosimetry system this is achieved by assigning a probability distribution to each parameter value. Multiple realisations of the dose for each worker can then be produced by randomly sampling from each distribution during each realisation of the dose. This process is repeated for each worker. It is important to distinguish between parameter values which are shared between workers and those which are unshared. Monte Carlo techniques are used to implement this scheme. When running such a scheme in reverse, from measured data to intake and dose, it is necessary to take account of the varying goodness of fit to the data for different selections of parameter values. This requires the use of Bayesian statistics. A novel method called WelMoS was developed to apply the Bayesian method to these calculations.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Methods to assess uncertainties in doses arising from internal contamination with plutonium: Bayesian statistics and hyper-realisations

  • 期刊论文

  • 发表日期：

    2025年12月3日

查看全文
  • Richard Bull,
  • Matthew Puncher,
  • Alan Birchall,
  • Vadim Vostrotin

摘要

When internal doses are calculated for the purpose of epidemiology studies, it is important that an assessment of the uncertainty in these doses is performed. This assessment should include both uncertainties in measured quantities and those in the various parameter values used in the calculation. In the Mayak Worker dosimetry system this is achieved by assigning a probability distribution to each parameter value. Multiple realisations of the dose for each worker can then be produced by randomly sampling from each distribution during each realisation of the dose. This process is repeated for each worker. It is important to distinguish between parameter values which are shared between workers and those which are unshared. Monte Carlo techniques are used to implement this scheme. When running such a scheme in reverse, from measured data to intake and dose, it is necessary to take account of the varying goodness of fit to the data for different selections of parameter values. This requires the use of Bayesian statistics. A novel method called WelMoS was developed to apply the Bayesian method to these calculations.