During the last year the world has been impacted by the worst pandemic in the last 100 years. A key factor in controlling the spread of the virus has been the early diagnose of patients. In this talk we present a new molecular diagnose test for SARS-CoV-2, called esPCR, and validate its use. The new method is portable, fast, and inexpensive. We evaluate the accuracy of the esPCR diagnosis method in the presence of the gold-standard test, known as PCR, by means of the receiver operating characteristics curve modeled under a Bayesian nonparametric approach. The method is illustrated in a real data set obtained from a Mexican hospital during the COVID-19 pandemic as well as in simulated data sets.