A general regression framework for group testing data, which incorporates pool dilution effects.

摘要: Group testing, through the use of pooling, has been widely implemented as a more efficient means to screen individuals for infectious diseases. Typically, in these settings, practitioners are tasked with complimentary goals both case identification and estimation. For purposes, many group testing strategies have proposed, which address issues such preserving anonymity estimation studies, quality control, classification. In general, require that significant number be retested, either pools or individually. order provide general methodology can used accurately precisely analyze data this form, herein, we propose binary regression framework incorporate arising from any strategy. Further, relax previously made assumptions regarding error rates by relating diagnostic results latent biological marker levels being tested. We investigate finite sample performance our proposed simulation applying techniques hepatitis B collected part study involving Irish prisoners.