The efficiency of coronavirus testing can be improved significantly by using a mathematical technique, according to a new study.
Researchers from Ben-Gurion University of the Negev in Israel adopted a new approach to a technique known as pool-sampling, which tests a group of people rather than individuals to identify suspected carriers quickly at a relatively low cost.
An algorithm programmed to use combination, a mathematical technique that deals with the selection of items from a collection, was able to identify cases with a 100 per cent accuracy rate, according to the research published in Science Advances last week.
Get the latest insights and analysis from our Global Impact newsletter on the big stories originating in China.
The team divided 384 samples into 48 pools, and the content of each sample was mixed to ensure it appeared in six different pools.
Four positive cases were found among the samples and scientists were able to use the technique to pinpoint where it had come from.
About 10 to 30 per cent of Covid-19 patients had no symptoms but could spread the virus significantly, according to the lead researcher Angel Porgador.
“Until there is a vaccine, there will be an urgent need to increase diagnostic testing capabilities to allow for screening of asymptomatic and pre-symptomatic populations,” he said in a statement.
Pool sampling is already being used as a weapon in the global fight against the pandemic. Beijing, for instance, screened 10 million residents in about a week using this method.
Researchers in the capital grouped one sample from each person in a pool and all people in the pool – usually three or five people – were tested again if the group result was positive.
In the Israeli study, each person was only tested once and fewer tests were needed.
Porgador’s team warned that their method would only work in areas where the disease was prevalent in 1 per cent or less of the population because too many positive cases meant the algorithm could no longer produce reliable results.
But the researchers remained positive about the technology’s effectiveness, writing in the study that with the implementation of lockdown and other measures, “we anticipate that in the near future carrier rates are likely to drop below this threshold in many countries”.
Another common problem with pool testing is the dilution of samples.
If oral or nasal swipes from different people are mixed, viral genes become more difficult to detect with commercial test kits.
The researchers said the optimal sample-to-pool ratio was around eight. As long as the number of pools stayed under this threshold, false reports rarely occurred, if at all.
The new pooling method, called P-BEST, has been approved by the Israeli Ministry of Health for use in clinical laboratories earlier this month.
More than 1,000 health care workers have received rapid screening using the method.
But it is unclear whether other countries such as China will embrace the technology.
An official working at one of Beijing’s coronavirus test centres told state media in June that the handling of a large number of samples could be very challenging, and their biggest nightmare was messing up the labelling.
Although the new approach can reduce the number of tests, it will increase the burden on technicians if the division and combination of samples has to be handled manually.
But a liquid-handling robot could do almost all the work, according to the Israeli scientists, and would take about an hour to pool the 384 samples without human intervention.
This article Coronavirus tests ‘could be more efficient using’ this mathematical technique first appeared on South China Morning Post