Infectious Diseases

Biography

Biography: Mark R Baker

Abstract

Statement of the Problem:. Existing tests for SARS-CoV-2 like PCR slow, expensive and and the case of lateral flow insensitive. There is a need for better for flash testing and variants.

Findings: Our method detects the early immune response by looking at variation in the leucocyte ratios, a methods that we publisehed in the peer reviewed press in 2020.

Using training data from the COVID-19 Risk and Treatments (CORIST) collaboration and the Moli-Sani Project, both  from southern Italy, comprising WBC differential data for 72 positive (RT-PCR confirmed) and 4742 negative  COVID patients, respectively, two variations of the test were applied. Algorithmic ensemble CLDC 21-04- SE optimized for sensitivity  has a sensitivity of 97·29% and specificity of 67·95%. Algorithmic ensemble 21-04-SP has a sensitivity of 81·82%  with a specificity of 96·50%.

Using data from the Regional Service for the Epidemiology and Surveillance of  Infectious Diseases database, a longitudinal cohort study of 379 patients with laboratory-confirmed COVID-19  admitted at the Italian National Institute for Infectious Diseases “Lazzaro Spallanzani” (INMI), in Rome (Italy). The  21-04-SE algorithm showed a sensitivity of 98·64% during the first seven days during symptom onset and 96·73%  sensitivity for data taken over 21 days after initial symptoms. The 21-04-SP algorithm showed a sensitivity of 79.0%.

Conclusion & Significance:

We have a robust repeatable test for SARS-CoV2 with a turnaround time of 30 seconds for a self-test from patient presenting to getting a positive or negative result. The ability to turn around tests very rapidly and de-skill the data collection promises to have a massive impact on COVID worldwide. The fact that no test laboratory is needed, just a smartphone, and the tests can be carried out by an unskilled individual gives a huge improvement.