AI Talk: Testing and tennis

May 15, 2020 / By V. “Juggy” Jagannathan, PhD

This week’s AI Talk…

COVID-19 diagnostic testing approaches

One of the continuing issues related to the current pandemic crisis is the availability of testing and confusing media coverage about the types of tests. I found this excellent article explaining testing methods and availability on the GoodRx site. Essentially, there are three main types of tests—molecular, serological and antigen. Molecular tests scan for the genetic material from either nasal swabs or saliva that indicate the presence of the virus. According to the authors, “The specific technique that’s used is called reverse transcription polymerase chain reaction, or RT-PCR, where genetic material from a sample is copied and then compared to the genetic sequence of the virus you’re trying to detect.” Serological tests look for antibodies using a finger-prick blood sample. The presence of antibodies indicates the person potentially already had the virus and recovered from it and may be immune from reinfection.

The latest entrant to the testing field is the antigen test. Last Friday, the FDA approved Quidel Corporation to develop these tests. Instead of looking for the genetic footprint of the virus, antigen tests look for viral proteins associated with the virus. The tests are inexpensive and fast, delivering results in 30 minutes, but their efficacy is yet to be proven. If they work, it might ameliorate the current acute shortage of tests.

All tests have false positives (meaning the tests come out positive incorrectly) or false negatives (test show negative but should have been positive). The FDA is actively monitoring and approving companies to administer these tests. Literally hundreds of companies are involved in  this effort. X-Rays, CT-scans, and biomarkers can also be used to aid diagnosis, but it turns out they are used more as secondary support tools, the primary tool being RT-PCR (molecular) test. We will need comprehensive, accurate tests and contact tracing before we can safely emerge out of the current lockdowns.

Live tennis

I’m a tennis nut and it was gratifying to see that they actually managed to get a live tennis match going again this past weekend. So, how did they do it? Well, they managed to get a private tennis court In Florida on a 5-acre lot and recruited four professional players to do round-robin matches. It was fun to watch live sports once again—some of the first after the virus shutdown! They had a lot of social distancing rules: Each player played with his own set of balls; there were no ball kids and players called their own lines! They had some weird scoring to make the match short: 4-game set and no add scoring and I was pleasantly surprised to note that some players were using wearables to collect data from their activity.

That got me curious: What wearables are they using these days and what is the state of tech in tennis? Turns out that there is a lot of activity on this front! Anybody who has watched a professional tennis match has seen the Hawkeye, which basically calls the lines. There is also a version for amateur players to use. It also turns out you can have sensors in your racket and a number of tennis racket suppliers incorporate these sensors. Sensors allow you to monitor the speed of the ball off the racket, how much spin you put on the ball, etc. Then there are wearables that you can actually attach to your clothes which track how fast you move, your heart rate and other parameters. One company is  even embedding the entire tennis court surface with sensors to collect movement data. You can also subscribe to a service which will analyze your match if you send them a video. It seems like a fascinating new world of tech for tennis! I haven’t tried any of these yet (maybe a bit out of my price range).

I am always looking for feedback and if you would like me to cover a story, please let me know. “See something, say something!” Leave me a comment below or ask a question on my blogger profile page.

V. “Juggy” Jagannathan, PhD, is Director of Research for 3M M*Modal and is an AI Evangelist with four decades of experience in AI and Computer Science research.