18 June 2021 - - Seattle, US-based technology startup Pattern Computer, Inc. has developed a novel test for near real-time (15 sec) diagnosis of SARS-CoV-2 infections, the company said.

In contrast with widely used antibody and PCR tests, which take at least 15 minutes to produce results, and require scarce reagents to process, the newly developed PCI method enables high-volume throughput of tests without requiring chemical reagents at difficult to service locations, such as company entrances, event venues, court houses, state and federal facilities, TSA screening areas and ports of entry.

Developed using the Pattern Discovery Engine comprised of proprietary software methods and specially configured hardware, the initial results suggest this low-cost, high-volume COVID-19 test could create new opportunities for reducing transmission risk in high-headcount, high-density entertainment venues, large-scale events, office buildings, trains, busses, aircraft, and subways.

Pattern Computer's new test has been used to analyze data from symptomatic and asymptomatic patients between 10 and 80 years old.

The reported balanced accuracy of the new test is >96%. The aspects of the testing that are worth noting: the Pattern team has been working on the problem for a few months, and is working with a dataset that numbers in the hundreds of data points.

Competing organizations and teams have been working on the same problem for over a year with millions of data points, to only deliver tests that are 97% accurate.

The results from the Pattern tests are competitive with the performance of antibody tests currently in the market, which under ideal conditions (3 weeks after onset of symptoms) reports a balanced accuracy of 95%.

In contrast to antibody tests, which are highly sensitive to timing--only 30% of people infected with SARS-CoV-2 tests positive 1 week after the onset of symptoms--the performance of PCI diagnosis is agnostic regarding how long the patient has been symptomatic.

PCI has built an initial model that predicts COVID-19 status (as determined by multi-target RT-PCR) of a patient based on its proprietary hardware measurements. Sensitivity and specificity are computed with 5-fold cross-validation.

PCI expects that as more data points are added to the PCI data base, particularly data from non-infected individuals, its technology will be able to match or exceed RT-PCR for accurate SARS-CoV-2 infection detection, with the reduced result time of 20 seconds rather than the current standard wait time of 30 minutes or more.

Pattern Computer, Inc., a Seattle-area startup, uses its proprietary Pattern Discovery Engine to solve the most important and most intractable problems in business and medicine. Its proprietary mathematical techniques can find complex patterns in very-high-order data that have eluded detection by much larger systems.
While the company is currently applying its computational platform to the challenging field of drug discovery, it is also making pattern discoveries for partners in several other sectors, including additional biomedical research, materials science, aerospace manufacturing, veterinary medicine, air traffic operations, and finance.