Tag Archives: glucose
For people who take insulin, self-measurement is a matter of life and death. No wonder, then, that people with diabetes who track their blood glucose have been so important in advancing techniques of visualization,and understanding data. At the Quantified Self Europe conference in Amsterdam this year, we were honored to host a panel discussion on Data Visualization and Meaning with Joel Goldsmith (Abbott Diabetes Care), Jana Beck (Tidepool), Doug Kanter (Databetes), and Stefanie Rondags (diabetes coach and blogger).
This discussion strikes me as widely important for self-trackers whether or not we have diabetes. Many of us will be tracking blood glucose in the near future. And the issues of data access, understanding, and clinical relevance that people with diabetes are working on resemble challenges commonly faced by anybody who is tracking for health.
For instance, Jana Beck was asked during the Q&A about her health care providers. How receptive are they to the important experiments she’s done to improve her health based on the data she’s collected? ”None of my endocrinologists have been very receptive to this approach,” she answered. “My A1C tends to fall within the range of what’s considered the gold range for people with Type 1. But I’m interested in optimizing that further. Often, I don’t even see them more than twice a year.”
Jana, Stefanie, and Doug all showed their own data in the context of discussing experiments and decisions that have had a major impact on their wellbeing. All were clear that the domain of these experiments and decisions is not healthcare as traditionally understood; but nor is it a matter of general fitness or lifestyle. The domain of these experiments is different and perhaps still unnamed. Self-collected data can and should essential health decisions, but the most advanced techniques of understanding this data are still being developed in an ad-hoc, grassroots way, by knowledgeable and open minded individuals who have a strong interest in learning for themselves.
At the end of the session I asked Joel Goldsmith, of Abbott Diabetes care, about the future prospects of the Freestyle Libre, a minimally invasive wearable blood glucose monitor that is not yet available in the US. (Disclosure: Abbott Diabetes Care was one of the sponsors of the QS Europe Conference.) The Freestyle Libre has a sensor in the form of a patch worn on the arm, and a touchscreen reader device that you lift close to the sensor to get a reading. There is no finger prick involved. While this and competing minimally invasive or non-invasive glucose monitors will almost certainly continue to be regulated as medical devices and understood as part of the health care system, many other people will also use them, and the flood of data and the questions that go with it will challenge our understanding of where this type of information should live.
The video above contains the full session, including the Q&A.
Bob Troia was interested in his blood glucose. While he’s not a diabetic and he’s not out of range, he wanted to see if he could lower his fasting glucose levels. He started a long-term tracking experiment where he tested his blood glucose and began to explore the effects of supplementation and lifestyle factors. In this talk, presented at our 2014 Quantified Self Europe Conference, Bob talks about his experiment and what he learned from analyzing his data. Make sure to read his take on what he did, how he did it, and what he learned below.
You can also view the slides here [pdf].
We also asked Bob to answer the three prime questions:
What did you do?
After learning via my 23andMe results that I had an elevated risk for Type 2 diabetes (and having an interest in the longevity benefits of maintaining low blood glucose levels), I began tracking my daily fasting glucose and the effects that diet, exercise, supplements, and stress have on glucose levels so I could take whatever steps I needed to proactively understand, control, and optimize it.
How did you do it?
Over the course of 7 months, each morning I would take a fasting glucose reading using a handheld glucose meter. After establishing a 30-day baseline of daily fasting glucose readings, I began to take supplement called oxaloacetate. It’s been shown to lower and more tightly regulate fasting glucose by mimicking the effects of caloric restriction. It’s a naturally occurring compound found in lots of foods, such as spinach, potatoes, or apples, and it’s as safe as Vitamin C. After several weeks, there was a noticeable improvement in my average values! I then started looking at day-of-week trends in addition to how exercise (in my case, playing soccer) and other things such as travel affected my glucose.
What did you learn?
I learned that I could indeed improve and better stabilize my fasting glucose levels using oxaloacetate – but only in conjunction with intense, interval-type exercise like soccer. My average fasting glucose levels are highest on Mondays (stress of a new work week) and lowest on the weekends. Long airplane travel can adversely effect my glucose levels for several days. Surprisingly, alcohol consumption did not have an effect.
As you may know, we get excited when someone in our community uses interesting data visualizations to help tell their self-tracking story. Jana Beck is no exception. As a woman living with Type 1 diabetes she’s constantly learning how to better understand what her Dexcom data is telling her. In this talk, Jana follows up on her previous show&tell presentation with some new visualization techniques she’s using. If you’re interested in Jana’s methods be sure to check out her Github repository and her work with Tidepool.org.
(Editor’s Note: I very interested in Jana’s use of Chernoff faces for multivariate data visualization. If you’re using this type of visualization for your own data I would love to see it. Get in touch.)
If you have diabetes, or know someone who does, you’ve probably encountered a blood glucose monitor. Like many medical devices, design and data visualization are usually an afterthought. While there are many new exciting products coming to market like the iBGStar designed by Agamatrix, there are individuals who want to learn more than just their current blood glucose values. Diabetes care is also moving towards an automated and coordinated process driven by continuous blood glucose monitoring and implantable insulin pumps. These devices live on data, huge amounts of data, but what do their users know? More specifically, what do their users understand about their data, their condition, and themselves?
Doug Kanter is a designer, photographer and a student in the Interactive Telecommunications Program (ITP) at NYU. He’s also a Type-1 diabetic who has a keen interest in applying actionable design and interaction schemes to the data he gathers from his monitoring systems.
It is time to re-imagine the entire user experience of being a patient with diabetes. There is tremendous potential in applying information technology, creative design and research into behavior change into a comprehensive product for patients. Technology-based solutions are increasingly important resources in these times of skyrocketing treatment costs and lmited doctor availability.
Doug has been using his skills to better visualize and understand his own data, particularly his continuous blood glucose monitor. His first project, 7729, explored one month of his continuous blood glucose monitoring – the 7729 readings to be exact.
His second project expanded on the 7729 project to include not only his blood glucose monitoring, but also the insulin he was receiving. Insulin on Board, is based on 100 days of data collection and includes 820 insulin pump reading and 25,012 blood glucose reading. By coordinating these two data sets he was able to look for patterns and identify the efficacy of his insulin dosing.
The goal of Insulin on Board was to better understand the relationship between the insulin I take and the resulting blood sugar readings. It visualizes not simply when I take a dose of insulin, but when that insulin “kicks in.” Because insulin has a latency, it is helpful to see it actually has an effect on blood sugar. Often times I’ll take two or more doses of insulin within a few hours. Insulin on Board calculates the sum overlapping effect of these dosages.
I think patients like me could benefit massively from having improved visualizations that give you both a solid overview of how you are doing but also allow you to dial down into the details if you want.
Being a student and designer, Doug has done a great job explaining the process he takes for developing these visualizations. If you’re interesting in learning more about how he created these visualizations, what he learned, and future work you can follow along at Databetic and his blog.
Every few weeks be on the lookout for new posts profiling interesting individuals and their data. If you have an interesting story or link to share leave a comment or contact the author here.