Connecting Type 1 Diabetes Researchers: The Sugar Science

This content originally appeared on Beyond Type 1. Republished with permission.

By Monica Westley

As a scientist and a parent of a child with type 1 diabetes, I was compelled to fully understand the etiology of the disease. I created a group called “The Sugar Mamas” to connect parents to live, interactive interviews with researchers. I reached out to scientists and scheduled regular “Lunch and Learns.” After each Skype call, parents went away feeling hopeful and inspired. It was a powerful connection for parents to understand how hard scientists were working on this disease. Last fall, I shared my process and helped Beyond Type 1 implement the connection of their community to researchers as well. I am a true believer in the adage, “the more information, the better!”

Creating The Sugar Science

Through countless interviews with researchers in the type 1 diabetes (T1D) field, I began to understand recurring pain points in the scientific community that was hindering more rapid progress. With the data in hand and a personal call to action, I began to build a digital platform in March 2020. We currently are a devoted and dedicated team of 23 talented volunteers, the majority with a close connection to T1D. Together we created The Sugar Science (TSS) to serve the wishes of scientists and catalyze a cure.

Our platform has already received endorsements from top researchers in the T1D world, including Dr. Douglas Melton (Harvard) and Dr. Alice Long (Benaroya), who act as our advisors. The Diabetes Research Connection (DRC), as well as Unanimous A.I., have partnered with us. Gaining validity, we were semi-finalists for the Women Who Tech grant, and we won a Google grant.

TSS revitalizes scientific communication in the same way that our social communications have transformed by digital tools over the last decade. Social networking and AI tools on the TSS platform are poised to bring together a field that has been silo-ed for decades, not due to the considerable effort of scientists, but due to the multi-factorial nature of the disease.

Providing Tools

Online meeting

Image source: Beyond Type 1

The Sugar Science provides tools that scientists working in T1D have requested. Current tools include The Collaborator, Thought Experiment, and KG.

The Collaborator is “speed dating” for ideas. Scientists post just three slides with short descriptions of their idea. The community gives feedback as to whether this is a “good idea,” and other scientists can connect to collaborate. A “match” can submit a fast-tracked grant to the DRC for funding.

Thought Experiments (TE) is a tool where scientists can post controversial (or not) ideas and the community can weigh in. Scientists whose answers gain “likes” from the community will gain a reputation. These scientists will be invited to participate in a SWARM AI event, tackling the toughest questions in type 1 diabetes along with experts in the field.

KG is the Knowledge Graph. We are building a knowledge graph to reflect all historical papers against a backdrop of negative data. This will give scientists studying T1D a new perspective on work that has already been done in the field as well as show places where work needs to be done.

Moving Forward

Overall, we remain true to our mission: to help T1D scientists connect, collaborate, and fund their best ideas. TSS features podcasts and interviews with scientists. We are scheduling “off the record” private brainstorms. We want to elevate young scientists interested in T1D and support them. In this spirit, we are hosting a PITCH COMP for post-docs and graduate students who study T1D on September 25, 2020. It will be a chance to shine for labs looking to connect, and the best pitches will be awarded funding. This event is particularly meaningful with the COVID-19 pressures that the scientific community is experiencing.

Please feel free to support our mission. The general public can donate (we are a tax-deductible foundation) on our website, via our socials or using Amazon Smile. All donors will receive our monthly digital newsletter.

As a parting comment, I would say for myself and my team, for us it is all about a cure. We know first-hand what this disease is like, what it does to those who have it and their families. As an all-volunteer organization, no one is paid. And yet, we are getting things done, moving forward. Our team at The Sugar Science is all about the end game: a cure.

Source: diabetesdaily.com

New App Uses Artificial Intelligence to Predict Blood Glucose Levels Without a CGM

It’s hard to deny that technology advances are improving the lives of people with diabetes worldwide. From smart insulin pumps that integrate with continuous glucose monitors (CGMs) to various applications with predictive features and alarms, the diabetes tech world continues to evolve rapidly.

One company, January.ai, has recently announced its new artificial intelligence (AI) platform can accurately predict blood glucose responses to various meals. The company was founded in 2017 by Silicon Valley veteran and CEO Noosheen Hashemi and  Mike Snyder, the Director of Genomics and Personalized Medicine at Stanford, with the vision of improving lives by providing comprehensive health data. The concept was recently validated, and the company has developed a user-friendly app to help people with diabetes learn more about what affects their blood glucose levels and improve outcomes.

How It Works

The new algorithm relies on machine learning approaches to predict individual blood glucose responses to different meals and activities. To achieve this, the algorithm considers the users’ heart rate, and logs of their food and medication data, developing a personalized model for each patient to predict glycemic outcomes. The initial “training” period takes four days, and does incorporate data from a CGM; however, no CGM data is needed to make the predictions past the initial training period.

As per the recent press release,

“The company developed a series of underlying technologies including derived nutritional values, glycemic index and glycemic load, which estimates how a person’s blood sugar will rise based on the food they eat, for 16 million foods. January.ai built its own mobile application to capture and unify various data points into one AI platform, collecting nearly 25 million data points for the study.”

At the American Diabetes Association (ADA) 80th Scientific Sessions, the research team presented the outcomes of this algorithm in predicting the glycemic responses of over 1,000 participants. Some were diagnosed with pre-diabetes or type 2 diabetes, while others represented healthy participants.

Participants wore a CGM as well as a heart rate monitor for ten days. They also tracked their activity levels, specific food and water intake, as well as their medication doses. Following the four-day learning period., the algorithm developed an “individualized model” for each participant. Next, the system’s ability to accurately predict blood sugar responses without using any CGM data was put to the test. Excitingly, the predictive values were in close accord with the actual CGM readings, which were used to validate the accuracy of the predictions.

Rahili S. et al. 2020 (Presented at the ADA 80th Scientific Sessions) The above slide shows the model’s glucose prediction for a 33-hour period based on a participant’s heart rate, food, and medication data, compared to their actual glucose levels.

The App

The team has applied their state-of-the-art algorithm to develop an app that enables users to track their heart rate and blood sugar levels, as well as get a comprehensive picture of how factors like specific foods and exercise patterns affect them, personally. Moreover, due to the machine learning features, patients can also be alerted to potential pitfalls before they even consume a particular meal. The app also features various data displays, related explanations, suggestions, and offers rewards for making improvements.

Image source: January.ai

Summary

The ability to accurately predict changes in blood sugar levels using just heart rate data, and food and medication logs, can offer a more affordable and non-invasive way for those with diabetes to learn about how different foods affect their blood sugar levels.

Noosheen Hashemi, Founder and CEO of January.ai had this to say about what their product could do for those living with diabetes:

“Despite extensive efforts, the healthcare community has not been able to slow the rapid rise of diabetes, nor develop effective treatments. We believe that by applying AI to a mix of biological and behavioral data, we can empower people with the personalized insights and specific recommendations they need to enjoy better health.”

What are your thoughts on this technology? Please comment below, we love hearing from our readers!

Source: diabetesdaily.com

One Year into DIY Looping

One year ago, I built a DIY hybrid-looping insulin pump, using my Dexcom G6 and Omnipod. For those who aren’t in the know, DIY “looping” is basically “hacking” your insulin pump with a single-board computer, such a Raspberry Pi or Riley Link, to make it communicate with an existing continuous glucose monitor (CGM) to make basal adjustments accordingly.

It’s important to note that this is NOT FDA approved, but the #WeAreNotWaiting community has been sharing information on how to build your own DIY looping insulin pump for years now, and I took the plunge in 2019.

In July I celebrated one full year on my looping system, and wanted to share my thoughts on 365 days of looping.

I Still Have Diabetes

I remember when I first set up my Riley Link and switched on “auto-mode.” I had this magical vision of never counting carbohydrates again, limitless runs without lows, and forgetting what the thirst of a high blood sugar felt like. Then I realized, just as quickly, that I still have diabetes.

Even though my Dexcom continuous glucose monitor (CGM) readings now communicate with my insulin pump and make basal adjustments accordingly, the “hybrid” part means that it doesn’t anticipate, nor account for, any carbohydrates eaten. I also need to tell my pump when I’m about to exercise, and for how long. Since the insulin pump does not operate on artificial intelligence (AI), it cannot anticipate what I’ll do next.

So yes, I still have lows on runs and I still have highs when I eat something that isn’t appropriately accounted for. I still have to count carbohydrates and no, I haven’t forgotten what the Death Valley-like thirst of a 350 mg/dL feels like, although it happens less frequently.

My HbA1c Isn’t That Much Lower

I have always been maniacal about tight diabetes control. My A1cs have hovered in the low 6s for the last 10 or so years. With Loop, I immediately thought that my control would be *perfect* and I would ride out the 4s and 5s into an eternal sunset. NOPE. My latest A1c was 5.9%, which I am rightly ecstatic about, but it’s less than 1% point lower than I was on MDI and a CGM.

The key difference is that my time in range has increased from around 30% to 75%, and the number of lows that I experience has gone down from around 3 per day to 3 per week. It’s easy to have a low HbA1c when you have highs and lots of lows to average it out- it’s much harder (and healthier!) to have a lower HbA1c with few lows. And plus, I just feel healthier. And that has made all the difference.

Dexcom graph by Christine Fallabel

It’s a Mental Vacation

Being a human pancreas 24/7/365 is not easy (why didn’t anyone tell us this at diagnosis?!). In addition to running a household and having a full-time job (and texting everyone back, and maintaining some semblance of a fitness routine, and trying to eat something green at every meal), being an organ all of the time is hard work.

More than anything, a year into looping has given me the mental break I didn’t know I needed. Sure, I still have to count carbohydrates, adjust for exercise, and dose for meals, but hours can go by where I don’t think about diabetes at all, and that never used to happen. My mental distress has gone way down, and I don’t experience diabetes burnout at nearly the frequency I used to. This also helps maintain my motivation to continue to take care of myself and my diabetes.

Dosing Is More Discreet

As I make my way through my 30s, this is less of an issue (if you have a problem with me dosing in public, the problem is you, not my diabetes), but looping has definitely made checking my blood sugar (read: checking my phone) and dosing (also read: checking my phone) way more discreet in public than manually testing my blood sugar and dosing used to be. It’s also more hygienic (I change my insulin pump with plenty of alcohol swabs every 3 days from the comfort of my home), and more convenient. This is perfect when I’m out at a crowded concert, or squeezed into a small table at a restaurant.

It can also cause issues. For instance, if I’m in a public place where cell phones aren’t allowed, sometimes it’s difficult to explain that my iPhone is actually durable medical equipment (DME) that I need to survive. Let’s just say there have been some teachable moments.

Loop app screenshot by Christine Fallabel

I Am Happier

When I was diagnosed with type 1 diabetes in June of 2000, my doctor told me that the cure was just 5 years away. I thought the cure was just around the corner, we all did. And learning that the “cure” is still out of sight, 15 years hence, has been a hard pill to swallow.

I’ve dealt with anxiety and the impending depression of only someone who has a chronic disease with no cause and no cure can experience, but having something like Loop feels like someone is finally on my side, looking out for me, and making things just a little bit easier when the load becomes too heavy of a burden to carry. I can go to sleep and know that my basal will immediately shut off if I start to go low overnight. I can relax if I’m digging into dinner at a friend’s house and I don’t know the exact carb count for a meal, knowing my basal will tick up to cover the difference.

Having a Loop feels a little bit like you have a certified diabetes educator (CDE) and best friend just sitting on your shoulder, making constant adjustments, never judging, and ensuring that you have a better go of it, a little bit of help when you need it. And that help has been life-changing. The cure may never have been 5 years out, but with Loop, I finally feel okay waiting just a little bit longer.

Do you DIY Loop? How has your experience been? Share this post and comment below; we would love to hear from you. Follow the #WeAreNotWaiting hashtag on Twitter to learn more about the DIY movement.

Source: diabetesdaily.com

Automated Insulin Delivery: Six Universal Observations and Understandings

This content originally appeared on diaTribe. Republished with permission.

By Laurel Messer

Six universal facts about automated insulin delivery systems, and the things you should keep in mind about this revolutionary technology

Automated insulin delivery (AID) systems are moving towards the forefront of diabetes management. AID systems combine continuous glucose monitors (CGM) with smart algorithms to automatically adjust insulin delivery.

The Tandem Control-IQ system was recently cleared by the FDA, and the Insulet Horizon and Medtronic Advanced Hybrid Closed Loop systems are beginning pivotal trials. These are encouraging developments. As more systems move through the pipeline and eventually into the commercial market, important patterns are emerging in user expectations and user experience. As a diabetes nurse, certified diabetes educator and research investigator, I, along with my team at the Barbara Davis Center, have worked with nearly every AID system in the pipeline, and other systems that will never make it to market. Here are six insights we have gleaned, which seem to be universal (thus far) to all AID systems:

1. You can always beat an AID system with compulsive diabetes management

Many people with diabetes compulsively attend to diabetes care in order to achieve ultra-tight glucose ranges – and are the first to ask about automated systems. What ends up happening is that these “super-users” are invariably frustrated that the system is not yielding the same results that they were able to achieve with their own calculations and management. An important point is that many automated systems are excellent at reducing mental burden for taking care of diabetes, excellent at reducing hypoglycemia, and adequate at improving glucose levels. Humans can beat automated systems if they attend to diabetes care near-constantly. The individuals who will likely be satisfied with AID are those who are comfortable with an A1C in the 7s or above, but they want to reduce the mental load of adjusting settings and micromanaging high glucose levels. The most important question to ask is, “Why do I want to start using an automated system?” If it is to achieve near-perfect glucose levels, the system will likely disappoint. If it is to reduce the burden of “thinking like a pancreas” all the time, it may be a good option. AID will excel at the marathon of diabetes care but may disappoint in the hour-to-hour sprint.

2. Systems work best when you let them work

Using both research and commercial systems, we have seen all the ways to “trick” AID systems—entering phantom carbohydrates, changing set points, performing manual corrections, overriding recommended doses. More often than not, these behaviors lead to glucose instability – reactionary highs and lows from the system destabilizing. All systems will perform best if they are used according to user instructions. This is difficult for the individual who would prefer to micro-adjust settings or desire control over all insulin delivery. Most systems work best when users learn to trust them.

3. Give the system a chance – 2-4 weeks before deciding long term potential

It may benefit us to think about AID like a new significant relationship – it can take some time to “settle.” I mean this both on an interaction level (learning how to respond to alerts, when to intervene, when to let it ride) and on an algorithm level (allowing the system to adjust internal algorithm parameters based on usage). In addition, programmable user settings may need some adjustment in the first few weeks of use, so working with diabetes educators can be helpful for initial set-up and early follow-up.

4. Bolusing is still king

If I could go back in time, I would caution device manufacturers against any whisper of not needing to bolus with AID systems. Bolusing is the singular most important action a person with diabetes can do to optimize insulin delivery on current and near-future automated systems. This will be true until insulin action time gets exponentially faster or artificial intelligence gets better at predicting human behavior, neither of which is on the immediate horizon. In order for people with diabetes to see the best performance on any system (automated or manual), they need to bolus before carbohydrates are consumed. Specific to AID, the timing of the bolus (prior to carb intake) is especially important, as the system will automatically increase insulin delivery after an initial rise of glucose levels, so a late bolus (e.g., after the meal) could lead to insulin stacking and hypoglycemia.

5. Rethinking low treatments

Low glucose levels (hypoglycemia) still happen when using automated systems. What is different with AID is that the system has been trying to prevent the low by reducing/suspending insulin, possibly hours before the low occurs. This means that an individual may need to consume significantly fewer carbs to bring glucose levels back into range – perhaps 5-10 grams of carb at first, reassessing 15-20 minutes later. This can be difficult when wanting to eat everything in sight; however, it can reduce the chance of rebounding into the 200s after over-treating.

6. Infusion sets are still infusion sets

While AID algorithms are revolutionary, the infusion set is not. It is the same plastic or steel cannula that occludes, kinks, or inflames. This hardware limits automated systems and can very quickly lead to hyperglycemia or diabetic ketoacidosis (DKA). It is important for people using AID to recognize signs of infusion set failure – persistent hyperglycemia, boluses that do not bring glucose levels down, ketones, vomiting, etc. Knowing how to treat ketones (via syringe injection of insulin and set change) can prevent a hospital admission or worse.

I love that the diabetes community learns from its members and experiences. Check out our Barbara Davis Center PANTHER (Practical Advanced THERapies for diabetes) website for our team’s latest insights on automated insulin delivery, and tools for people with diabetes, clinicians, and engineers.

Are you considering AID? Feel free to share this article with your healthcare team. For more information about AID systems that are currently available or in the pipeline, click here.

About Laurel

Laurel H. Messer is a nurse scientist and certified diabetes educator at the Barbara Davis Center for Diabetes, University of Colorado School of Medicine, Aurora, CO. She has spent the last 15 years studying how to best utilize new diabetes technologies, and remembers fondly teaching families to wrap up their corded CGM system in a plastic shower bag for bathing. Ok, not that fondly, but look how far we have come! Dr. Messer works with the Barbara Davis Center PANTHER team (Practical Advanced Therapies for diabetes), conducting clinical research trials on promising technologies to make life better for children, adolescents, and adults living with type 1 diabetes. Get in touch at Laurel.Messer@cuanschutz.edu

Source: diabetesdaily.com

Search

+