Group Testing With Nested Pools

被引:0
|
作者
Armendariz, Ines [1 ,2 ]
Ferrari, Pablo A. [1 ,2 ]
Fraiman, Daniel [3 ,4 ]
Mario Martinez, Jose [5 ]
Ponce Dawson, Silvina [6 ,7 ]
机构
[1] Univ Buenos Aires UBA, Dept Math, C1428EGA, Buenos Aires, DF, Argentina
[2] UBA CONICET, Luis A Santalo Math Res Inst IMAS, C1428EGA, Buenos Aires, DF, Argentina
[3] Univ San Andres, Dept Matemat & Ciencias, B1644, Victoria, Argentina
[4] Univ San Andres, CONICET, B1644, Victoria, Argentina
[5] Univ Estadual Campinas, Dept Appl Math, BR-13083859 Campinas, Brazil
[6] FCEN UBA, Dept Phys, C1428EGA, Buenos Aires, DF, Argentina
[7] UBA CONICET, Buenos Aires Phys Inst IFIBA, C1428EGA, Buenos Aires, DF, Argentina
关键词
Dorfman's retesting; group testing; nested pool testing; adaptive pool testing; DEFECTIVE MEMBERS;
D O I
10.1109/TIT.2021.3123929
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In order to identify the infected individuals of a population, their samples are divided in equally sized groups called pools and a single laboratory test is applied to each pool. Individuals whose samples belong to pools that test negative are declared healthy, while each pool that tests positive is divided into smaller, equally sized pools which are tested in the next stage. In the (k + 1)-th stage all remaining samples are tested. If p < 1 - 3(-1/3), we minimize the expected number of tests per individual as a function of the number k + 1 of stages, and of the pool sizes in the first k stages. We show that for each p is an element of (0, 1 - 3(-1/3)) the optimal choice is one of four possible schemes, which are explicitly described. We conjecture that for each p, the optimal choice is one of the two sequences of pool sizes (3(k) or 3(k-1) 4, 3(k-1), ..., 3(2), 3), with a precise description of the range of p's where each is optimal. The conjecture is supported by overwhelming numerical evidence for p > 2(-51). We also show that the cost of the best among the schemes (3(k), ..., 3) is of order O (p log(1/p)), comparable to the information theoretical lower bound p log(2)(1/p) + (1 - p) log(2) (1/(1 - p)), the entropy of a Bernoulli(p) random variable.
引用
收藏
页码:1119 / 1132
页数:14
相关论文
共 50 条