Error Correction Codes for Testing Chemical/Biological Agents: Using Pooled Testing to Increase Test Reliability

Jirong Yi, Myung Cho, Jingchao Gao, Xiaodong Wu, Weiyu Xu, Raghu Mudumbai

Abstract: In this paper, we propose a novel method of error correction code that increases the reliability of testing for chemical or biological agents, as well as general substances, through pooled testing. When gross errors, or outliers, occur in testing results, instead of conducting multiple tests on each individual sample, our method performs testing on well-designed mixtures of samples from multiple subjects. Unlike group testing, which aims to reduce the number of tests, our method aims to increase the reliability and accuracy of testing, even in the presence of gross test errors. Through our theoretical results and extensive simulations, we demonstrate that our method can significantly improve testing accuracy, even under gross testing errors. Furthermore, our method is proven to be more effective than repeated testing of individual samples.