A sewer “tree network” can be accessed through many service holes (typically about 300 feet apart). A new algorithm can dynamically and adaptively select which maintenance holes in a community should be tested for indications of SARS-CoV-2 in order to lead to outbreak areas. In this case, a “hot spot neighborhood” of infections is found. Photo credit: Courtesy of the researchers.
Having an algorithm deciding which maintenance holes to test for signs of coronavirus can improve pandemic containment efforts.
Covid-19 is a respiratory disease that spreads when infected people catch the novel coronavirus (SARS-CoV-2) that causes it. While this appears to happen primarily through close contact and breath droplets, there is evidence that the disease can also spread through airborne transmission. Distancing, masks, and improved ventilation are important measures to stop this spread.
Many who suffer from Covid-19 also throw the virus in their stools. If properly installed, this is an unlikely source of virus transmission. However, with the right tools, this can also be an unlikely source for virus detection. Viral traces of the novel coronavirus SARS-CoV-2 can be detected in wastewater up to a week before physical symptoms appear. This means that sewage can serve as an early warning signal that Covid is present in a community. However, in larger communities, it can be difficult to further isolate the occurrence of infections.
A recent one from Richard Larson, professor at the Institute for Data, Systems and Society at WITH, describes two “tree search” algorithms that can be used to dynamically and adaptively select which maintenance holes in a community should be tested to find possible causes of outbreaks. “The algorithms depend heavily on the structure of the sewer network,” says Larson. “It is a ‘tree network’ in which the sewage from its source flows in one direction via a unique route to the sewage treatment plant.”
Larson and his co-authors – Oded Berman from the University of Toronto and Mehdi Nourinejad from York University – used this tree graph structure and developed two algorithms. The first is designed for a community that is initially free of infections and the second for a community that is known to have many infections.
Several sewage treatment plants around the world are testing for coronavirus to gauge the extent of the infection in the community. The first algorithm should react when the wastewater from a sewage treatment plant has just discovered traces of SARS-CoV-2, which indicates the presence of a new infection in the community. This algorithm usually identifies the city block or even part of a city block that the infected person is in.
In the case of a more widespread infection, the second algorithm is located in the most infected neighborhoods or “hot zones”, usually several city blocks.
MIT recently started testing wastewater to detect Covid-19 on campus, with sampling ports collecting wastewater from the exit pipes of several buildings. However, dormitories house dozens and sewage treatment plants serve thousands. For a dormitory, a positive result could mean targeted follow-up measures such as individual testing and quarantine. For sewage treatment plants, results can be a useful indicator of infection in the community, but are often too broad for localized responses.
Larson believes the next step could be sequentially testing wastewater from a fraction of the community’s many maintenance holes. “With hundreds of manholes, we could test about six to ten and find a swelling area of 100 people or less,” says Larson. “The group to be tested now consists of people who are resident in the immediate catchment area of the source shaft.”
Larson’s research could make up for shortcomings in widespread community testing, which continues to be challenging in many places. Testing thousands of people requires equipment, manpower, and other resources, not to mention buy-in from the affected communities. Finding newly infected people can be like looking for a needle in a haystack. “Successful implementation of this algorithm could significantly reduce the size of this haystack,” says Larson.
While the mathematics of the algorithm has been developed and tested with numerous data sets, the operational implementation of the method is waiting for the invention of a fast, accurate and inexpensive SARS-CoV-2 test to be carried out at the maintenance holes. Current research on the detection of viruses at MIT and elsewhere is close to developing such a test. At this point the method could be tested on site.
“On-site testing can identify other problems affecting the flow of infected sewage in pipeline systems,” warns Larson. “Problems that must be resolved before a reliable implementation can be achieved.”
Nevertheless, the algorithmic wastewater tests could give individual districts an early warning sign of a coronavirus infection and trigger targeted follow-up through tests and distancing. This could help minimize the spread of disease, take pressure off health systems – and even save lives.
Reference: Richard C. Larson, Oded Berman and Mehdi Nourinejad, October 5, 2020, “Sampling of manholes in SARS-CoV-2 infections”. PLUS ONE.
DOI: 10.1371 / journal.pone.0240007
