Abstract:The ongoing coronavirus disease 2019 (COVID-19) is a pandemic causing millions of deaths, devastating social and economic disruptions. Testing individuals for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the pathogen of COVID-19, is critical for mitigating and containing COVID-19. Many countries are implementing group testing strategies against COVID-19 to improve testing capacity and efficiency while saving required workloads and consumables. A group of individuals’ nasopharyngeal/oropharyngeal (NP/OP) swab samples is mixed to conduct one test. However, existing group testing methods neglect the fact that mixing samples usually leads to substantial dilution of viral ribonucleic acid (RNA) of SARS-CoV-2, which seriously impacts the sensitivity of tests. In this paper, we aim to screen individuals infected with COVID-19 with as few tests as possible, under the premise that the sensitivity of tests is high enough. To achieve this goal, we propose an Adaptive Group Testing (AdaGT) method. By collecting information on the number of positive and negative samples that have been identified during the screening process, the AdaGT method can estimate the ratio of positive samples in real-time. Based on this ratio, the AdaGT algorithm adjusts its testing strategy adaptively between an individual testing strategy and a group testing strategy. The group size of the group testing strategy is carefully selected to guarantee that the sensitivity of each test is higher than a predetermined threshold and that this group contains at most one positive sample on average. Theoretical performance analysis on the AdaGT algorithm is provided and then validated in experiments. Experimental results also show that the AdaGT algorithm outperforms existing methods in terms of efficiency and sensitivity. Note to Practitioners—Real-time reverse transcription-polymerase chain reaction (rRT-PCR) tests provide scope for automation and are one of the most widely used laboratory methods for detecting the SARS-CoV-2 virus. This paper is motivated by the following challenges: (1) Many countries are experiencing an acute shortage of professionals and consumables for conducting rRT-PCR tests; (2) Group sizes of existing group testing methods against COVID-19 may not be optimal, which adversely impacts the efficiency of the screening of the SARS-CoV-2 virus; (3) Existing group testing methods do not consider the fact that the sensitivity of rRT-PCR tests usually decreases with the group size. The objective of this paper is to improve the efficiency and sensitivity of large-scale screening against COVID-19. For achieving this goal, we propose an Adaptive Group Testing (AdaGT) algorithm, which has the following advantages: (1) It can improve the efficiency for screening the SARS-CoV-2 virus, mainly by adaptively adjusting its testing strategy between an individual testing strategy and a group testing strategy based upon an estimated ratio of positive samples during the screening process; (2) It can guarantee a high sensitivity of the rRT-PCR tests by determining the group sizes of the group testing strategy based upon some constraints; (3) We derive an appropriate threshold for the estimated ratio of positive samples such that the AdaGT algorithm can achieve a minimum average number of rRT-PCR tests and can be directly employed in practical applications.

Diagnostic accuracy of autoverification and guidance system for COVID-19 RT-PCR results

A prediction model based on machine learning for diagnosing the early COVID-19 patients

Analysis of factors associated early diagnosis in coronavirus disease 2019 (COVID-19)

Effectiveness of a standardized quality control management procedure for COVID-19 RT-PCR testing: a large-scale diagnostic accuracy study in China

Laboratory diagnosis of COVID-19 in China: A review of challenging cases and analysis

Laboratory Verification of an RT-PCR Assay for SARS-CoV-2

Analysis of the Causes and Countermeasures for Positive Recurrence of SARS CoV-2 RNA in Discharged Patients with COVID-19

Application and Optimization of RT-PCR in Diagnosis of SARS-CoV-2 Infection

The application of a testing platform in high-throughput nucleic acid detection of SARS-CoV-2

AdaGT: An Adaptive Group Testing Method for Improving Efficiency and Sensitivity of Large-Scale Screening Against COVID-19

Sensitivity evaluation of 2019 novel coronavirus (SARS-CoV-2) RT-PCR detection kits and strategy to reduce false negative

A review of current effective COVID-19 testing methods and quality control

Quality Management for Point-Of-Care Testing of Pathogen Nucleic Acids: Chinese Expert Consensus

Automatic system for high-throughput and high-sensitivity diagnosis of SARS-CoV-2

Chinese Expert Consensus on the Nucleic Acid Detection of SARS-CoV-2

Machine Learning-Based Method for Interpreting the Guidelines of the Diagnosis and Treatment of COVID-19

Diagnostics and Analysis of SARS-CoV-2: Current Status, Recent Advances, Challenges and Perspectives

Recent Progress on Rapid Diagnosis of COVID-19 by Point-of-care Testing Platforms.

Exploration of turn-positive RT-PCR results and factors related to treatment outcome in COVID-19: A retrospective cohort study

Highly accurate and sensitive diagnostic detection of SARS-CoV-2 by digital PCR

Routine Laboratory Blood Tests Predict SARS-CoV-2 Infection Using Machine Learning