AI Talk: New normal, mining research, surveillance

March 27, 2020 / By V. “Juggy” Jagannathan, PhD

This week’s AI Talk…

New Normal – Clinician response to Covid-19

I saw this commentary by Dr. Thomas Lee on NEJM Catalyst. He gives his perspective on the virus as a practicing physician in pondering the post-pandemic future in his article: 1) Will we go back to our old, “normal” state? 2) Or will we have a new normal that incorporates what we have learned in dealing with the pandemic? The answer is hopefully, the latter. Personal hygiene and responsible social distancing are here to stay. But what about clinical practice patterns? Dr. Lee radically transformed his clinical practice by reaching out to a majority of his patients via phone calls and televisits. And that, he points out, is the right thing to do, not just in a crisis, but on an ongoing basis. We looked at how telehealth is becoming the new norm last week, and here is testimony from a practicing physician that it is the right thing to do for the majority of patient care interactions. There is the lingering worry as to how this is going to be squared away financially – but he offers a telling quote from Dr. Samuel Thier, former CEO of Partners Healthcare: “He said we should focus first on figuring out how to deliver the best possible care, which he pointed out was plenty hard to do. Only after that, he said, should we worry about how to get paid for it.” Amen to that!

New normal – What is it going to be?

Last week, MIT Technology Review featured an article titled: “We’re not going back to normal.” What is behind this viewpoint? A modeling study of the non-pharmaceutical interventions (NPI) used to reduce COVID-19 by Imperial College in London. What did this study find? Well, if you impose social distancing and shut down most businesses in cities you end up flattening the infection curve. But then what you do? Relax the social distancing rules which causes the infection rate to spike again, followed by more social distancing measures. In essence, one gets a saw-tooth like curve of infection, where the infection periodically increases and drops. But as the study notes, it is purely based on NPI—non-pharmaceutical interventions. Let’s hope the myriad of pharmaceutical interventions and vaccine efforts underway now bear fruit rapidly and bring this ordeal to an end.

Mining coronavirus research

This article in Wired is an eye-opener: The coronavirus has been around four months now and there are more than 2,000 research papers published about it! I guess we should not be surprised, as the whole world is focused on this pandemic, Now the question is how do you make sense of all this mind-boggling research which deals with every aspect of the pandemic? AI to the rescue? Microsoft Research, the National Library of Medicine and Allen Institute for AI are curating all papers related to Coronavirus—29,000 of them, not just the 2,000 focused on COVID-19—in an effort to comb through this corpus with AI algorithms to learn as much as possible. Hope the search results in a practical therapeutic solution.

Portable surveillance device

Saw this news item in AI in Healthcare. University of Massachusetts at Amherst computer scientists have developed a portable device, dubbed FluSense, that can monitor non-verbal sounds, such as coughs, and thermal imaging data to determine if there is possible flu present in an area. The data is processed through a machine learning algorithm and the device itself is portable and can be deployed anywhere. The idea is to use this as a surveillance mobile device for detecting flu. Perhaps another piece of the jigsaw puzzle on how to deal with the current crisis.

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.

During a pandemic, healthcare information is gathered, studied, and published rapidly by scientists, epidemiologists and public health experts without the usual processes of review. Our understanding is rapidly evolving and what we understand today will change over time. Definitive studies will be published long after the fact. 3M Inside Angle bloggers share our thoughts and expertise based on currently available information.