Our R&D team has developed a sophisticated machine learning model to predict whether a patient is SARS-CoV-2 positive or negative using 20 routine laboratory tests collected within a 2-day period. Our model predicted the results of the SARS-CoV-2 test with a specificity of 86.8%, a sensitivity of 82.4%, and an overall accuracy of 86.4% (with a 95% confidence interval of [86.0%, 86.9%]), which we published in the  journal “Clinical Infectious Diseases” (link:

Since developing this system, we have significantly improved the system’s accuracy using new patient data and refinements to the model timing and criteria, and have been developing it for broader applications such as predicting other diseases, infectious and non-infectious. We are also exploring methods of identifying treatment regimens that are highly beneficial to patients using a host of statistical and machine learning methods. In the long-term we hope to use a patient’s data to predict the conditions they have or will develop.  

Please email or use the contact form below if you are interested in learning more about these projects and solutions and how they can help your organization.