Diabetes Technology Video Perspectives

August 14, 2023
4 min watch
Save

VIDEO: Exploring new devices, options to measure glucose for diabetes

Transcript

Editor’s note: This is an automatically generated transcript. Please notify editor@healio.com if there are concerns regarding accuracy of the transcription.

I am excited about new methods for measuring glucose. The devices are getting more accurate all the time, but there's room for improvement. We're seeing better continuous glucose monitors. They're smaller, they last longer, and they're more accurate. And I think as these products improve, it'll be easier for people to give themselves the exact correct treatment, and therefore I expect better outcomes. Another area I'm particularly interested in is metrics. That is, when you measure somebody's glucose every five minutes for weeks at a time, you have a lot of glucose levels, and it would be helpful to be able to assemble these and make some sense out of them. Otherwise it's just overwhelming to look at all these numbers. And recently, I led a international project to create a new metric that is a description of the quality of glycemia. It would be similar to saying you go to a restaurant and you want to have a reviewer tell you how good is the restaurant. So they can look at the food, the service, the cleanliness, the parking, the menu, the reservations. There's a lot of things to look at. Clearly the food is the most important, but the others are also important.

And how do you put all that together to come up with a score if the doctor wants to know just how well is a person doing? Well, what a lot of these services do is essentially they have five or six different categories. Each one gets a fifth or a sixth of the rating. With diabetes, we do know that we get seven main types of information. They're not all equal. And we did a type of a survey of 330 international experienced diabetes doctors. We took their comments and they looked at actual tracings, 225 tracings, 22 per doctor, and we were able to figure out using artificial intelligence which of the different features of a continuous glucose monitor are most important, and the ones that are more or less important, how much more important is one than the other, so we could have a weighting, and then we created a formula. And using that formula, we can pretty much say this is what an experienced diabetes doctor would come up with. So if you have a patient, maybe you're not an experienced diabetes doctor, you put in the simple formula and it gives you a score between 0 and 100. 0 is the best, 100 is the worst. So you can see what percentile, these are all percentiles, what percentile your patient is in. So if your patient is, say, doing about average, they'll probably get a 50. If they're doing above average, maybe they'll be a 25 or 30. And if the person's doing poorly, below average, maybe they'll be 70 or 80. But it's a single number that puts together all these hundreds and thousands of readings over a 14-day period into a single number. So that'll make it easier for doctors to deal with the data that comes from a continuous glucose monitor. And I would say the last area that we're working on here is devices.

At Mills-Peninsula Medical Center, we're doing a study of a duodenal mucosal resurfacing product that, as I mentioned, works on the cells of the small intestine. We're also working with a spinal cord stimulator, and we're even working with a new way of measuring glucose that is so new there's no product like it on the market, and I'm just not allowed to say what it is at this point. But if things work out, this is going to be totally different and it's going to be a very popular way to measure glucose. I'll just say it's non-invasive, there's no needles, and it's a very clever idea.