Share on Pinterest
Image courtesy of Lane Desborough

One of North America’s leading diabetes innovators wants to take automated insulin delivery (AID) technology to the next level. He’s developing an algorithm called Nudge BG that would make these systems even easier to use and more effective, with little effort by the user.

All the magic would happen just like “Intel Inside” once powered home computers, making that company’s processing technology a household name.

California-based diabetes dad and seasoned engineer Lane Desborough has been at the forefront of diabetes technology for at least a decade, holding a senior position at Medtronic and later helping to launch the do-it-yourself (DIY) #WeAreNotWaiting movement pushing grassroots innovation.

While today’s current closed loop systems require people with diabetes (PWDs) to monitor data, input information, and take many actions daily, Desborough envisions a fully automated system that would not require the user to do any manual tasks like entering carbohydrate or glucose info.

In early 2020, he founded the startup Nudge BG with the simple idea of developing a smart algorithm to achieve just that.

“Simpler is better,” Desborough tells DiabetesMine. “In diabetes tech, we need to make something anyone can use, not just people who are willing and able to maintain high therapy engagement.”

Desborough is an engineer, statistician, competitive runner and sailor, scholar, inventor, and devoted dad. He has more than a dozen patents in his name.

Lane Desborough

Desborough’s influence has likely affected anyone familiar with the Medtronic Minimed insulin pump or integrated continuous glucose monitor (CGM), diabetes data-sharing, DIY D-tech, or people who’ve heard about future devices like Bigfoot Unity.

Before all that, he had a notable career at General Electric and Honeywell, working in remote monitoring and automation at oil refineries and power plants.

But after his son Hayden was diagnosed with type 1 diabetes (T1D) in 2009, Desborough took his engineering skills to the diabetes tech space. He started at Medtronic Diabetes and worked his way up to become chief engineer there, developing insulin delivery tech as well as the software and automation systems that are part of those devices.

In fact, it was Desborough who first uttered the words “we are not waiting” to describe what he saw at the inaugural DiabetesMine D-Data Exchange in November 2013, where our team gathered grassroots innovators tinkering with diabetes technology for the first time ever at Stanford University.

“We are not waiting” went on to become a popular hashtag and rallying cry for the DIY diabetes movement. It would spark further efforts on the Nightscout Project for remote data-sharing, homemade automated insulin delivery systems, and a host of phone apps and digital platforms now used by thousands across the world.

Desborough co-founded the startup Bigfoot Biomedical, along with others that included D-Dad Jeffrey Brewer, who was CEO of JDRF for several years, and Bryan Mazlish, who built a homemade “artificial pancreas” for his wife and son with T1D. He stayed there until the end of 2019.

Now, out on his own, he’s devoting his considerable brainpower to Nudge BG, currently a young startup with big dreams.

As Desborough describes it, he won’t be launching a physical device. Hardware is hard, he quips. Rather, Nudge BG will be an algorithm to use alongside or inside the hardware to manage one’s diabetes — “an adjunctive tool to make existing diabetes therapies better,” he says.

This algorithm “nudges” insulin to respond to CGM data without the need to manually bolus insulin for meals or corrections (of course, you can if you want, he adds). This may sound simple, but it would actually be a huge step forward for people dependent on insulin.

If you want to chase or be chased by the number, you’d still be able to call up the CGM app on your phone. If you want to bolus insulin for food, call up the pump app on your phone or press a couple of buttons on your pump.

“My goal is that it would work on its own so you can just get on with life, confident that Nudge has got your back,” Desborough says.

Nudge BG inherits settings from the insulin pump, and adapts over time to what’s being “learned” by the system from physiology, behavior, and data changes. He sees it being as easy to prescribe as any diabetes pill, without any settings or configurations needed and no user interface.

The form factor could be a smartphone app, but even that might be too much for some individuals. So, he’s also mulling the idea of just licensing the algorithm to pump-CGM-AID companies to offer as part of their package.

“Nudge BG succeeds when there’s nothing to see,” he says. “Its benefit is from its absence,” says Desborough.

Desborough is already talking with some insulin pump, CGM, and AID companies about future opportunities. He sees a world where a people have an opportunity to choose not only their pump and CGM that could work together, but different algorithms that could offer varying ways to manage diabetes.

“My sense is that at different points and times in their lives, PWDs want to choose their level of engagement,” he says. “And wouldn’t it be nice if there was a wingman who says ‘I got this’ instead of forcing people to enter carbs and bolus for every meal or snack? I’m trying to create something that meets people where they are for that transformation in care.”

“The majority of people dependent on insulin are not actively engaged in building their own homemade diabetes tech or participating in clinical research studies,” Desborough notes.

He worries that most companies are overly focused on the super-engaged PWDs, without considering the needs of the broader community.

That’s why he’s designing Nudge BG for users who prefer to be less engaged than more, and maybe even aren’t currently monitoring their glucose trends regularly. “They’re not interested in constant data notifications about what they’re ‘doing wrong,'” he says.

“Everyone is caught up in this feature-function, more-is-better battle… where one (product) has an exercise mode, so someone else needs an exercise-plus mode. I often felt that… marketing folks who have no understanding of diabetes believed the only way to improve was by adding more features wanted by lead users. By doing so, they were further distancing their offerings from people like my son,” says Desborough.

Now in his 20s, Desborough’s son Hayden uses the Tandem Control-IQ system but chooses to sidestep most of the advanced functions it offers. Instead, he largely keeps it in a sort of “sleeping beauty mode” all day — meaning it adjusts basal insulin rates but won’t deliver extra boluses automatically, and he has all pump and CGM notifications silenced except for the mandatory 55 mg/dL Low Alert that can’t be turned off.

“Having all those built-in features and notifications can create stigma, or emotional distress that interferes with someone’s own care,” Desborough says.

He likens it to a teenager who needs a car ride somewhere, but doesn’t want to drive or take the needed actions of owning or maintaining a car.

“Just be in the car, and I’ll put the gas in and do oil changes and maintenance,” Desborough says. “That’s what this is with Nudge BG. I’m not trying to build a new Tesla or a new car part, but trying to make the existing gas pedal or speedometer work better in the car you already have,” he says.

For inspiration, he looks to the Abbott FreeStyle Libre as a product that’s successfully appealed to a broad swath of PWDs who want to eliminate most fingerstick tests but don’t want a full CGM, out of fear of being “chased by numbers and alerts.”

Libre users get to choose their level of engagement, rather than having a full set of features thrust on them without any choice, including some mandatory alerts that can’t turn off. The Libre 2 model offers optional alerts. Also, the next-generation Libre 3, expected soon in the United States, will allow a choice for more traditional CGM functions if the user so chooses.

“With Nudge BG, you can still get notifications and alerts from your devices if you choose to activate them,” Desborough explains.

Wait, haven’t we heard this before? Nudge BG isn’t the only company developing a stand-alone algorithm. Several others have had their hands in this area for a while:

Desborough says that “Nudge BG is different,” because those other companies are developing what he views as “hybrid closed loop” algorithms that still require user engagement. Those systems automatically adjust insulin doses to keep PWDs more in range but still ask too much of people.

“If we design for early adopters and engaged users, we’re out of reach for most PWDs,” Desborough says. “I can see what’s coming, but we’re not there yet. We’re on the cusp.”

Since forming the startup, he’s collected thousands of data points to analyze results that help build Nudge BG with a base data store.

He won’t pinpoint a timeline for launch, preferring to underpromise and overdeliver.

“The landscape is changing so quickly that many things are unknown and unknowable,” he says. “But I do think a huge advantage of a small, software-intensive company like Nudge BG is that we can adapt quickly to unfolding events. Like a mammal in a dinosaur world.”