Hierarchical group testing for multiple infections.

摘要: Summary Group testing, where individuals are tested initially in pools, is widely used to screen a large number of for rare diseases. Triggered by the recent development assays that detect multiple infections at once, screening programs now involve testing pools simultaneously. Tebbs, McMahan, and Bilder (2013, Biometrics) recently evaluated performance two-stage hierarchical algorithm chlamydia gonorrhea as part Infertility Prevention Project United States. In this article, we generalize work accommodate larger stages. To derive operating characteristics higher-stage algorithms with more than one infection, view pool decoding process time-inhomogeneous, finite-state Markov chain. Taking conceptualization enables us closed-form expressions expected tests classification accuracy rates terms transition probability matrices. When applied data from four states (Region X States Department Health Human Services), provide, on average, an estimated 11% reduction when compared algorithms. For applications rarer infections, show theoretically percentage can be much larger.