Optimizing Diagnosis: Unmasking the Effectiveness of Dual Viral Testing

Konstantin Pluzhnikov
Data And Beyond
Published in
4 min readJun 7, 2023

--

Image by the author. It is just a descriptive picture. The accuracy of the tests below is separate from Flowflex ones.

I have taken the problem statement from a mathematician, Boris Trushin’s youtube channel https://youtu.be/8TBib_RSE5g (the video is in Russian, subtitles n/a, but the language of numbers is international)

This article is behind a paywall. To bypass it, please open this pinned article with friend links inside.

Intro

I got the idea for the article a month ago by visiting the first Dutch lesson at the school for adults. I had a cold then but could not miss the class because of the social contract with the Belgian government. After an explanation of the situation to the school’s staff, I received three COVID tests and took my way back home.

  • Why did they give me three COVID tests? Was it for my wife and me?
  • Or was it to test myself for several days?
  • Or were they good at probability theory?

I do not know for sure.
I took the test twice (but why?) and received negative results in both cases. Ultimately, it was just the flu, and I fully recovered in four days.
What was the probability I would be ill if the test were positive? And how did the probability change if the second test also were positive? Let’s try to study it.

--

--

Konstantin Pluzhnikov
Data And Beyond

an ML enthusiast, CQF alumnus, and auditor. I talk about ways to study real-world scenarios with machine models and statistics. Friend links on Twitter!