9 Responses

  1. Bernard Farrell
    Bernard Farrell May 28, 2013 at 9:32 am | | Reply

    This is great, and long overdue. I hope they’ll post some pictures of the visualizations from blip soon.

    Thanks Amy for keeping us updated on developments to do with data representation and visualization.

  2. Type 1 Electrical Engineering Student
    Type 1 Electrical Engineering Student May 29, 2013 at 1:23 am | | Reply

    I feel bad right now after receiving plasmapheresis today for autoimmune autonomic neuropathy, a rare disease I have but…

    While this project seems ambitious, there are people out there already writing programs that accept meter, nutrition, insulin pump, and CGMS data from file exports from the respective data retrieval software.

    I essentially wrote a specific version of the artificial pancreas project in MatLab with the goal of running my all of my diabetes data simultaneously through the program. In fact, I carry my food scale, USB cables, and my netbook with me almost every place I go, so if I need to make adjustments, I can, while minimizing risk.

    There is also somebody out there, who has programmed their HP 50g calculator to dose using similar equations.

    Dexcom already has patents on a “lookup” function, where you can find “similar days” of glucose patterns in your BG data “history”. I already created such a program for myself, using statistics and numerical methods.

    I am just saying, there are a lot of useful ideas out there. But, as a community of people who deal with diabetes, we cannot settle for less than the ISO/IEEE 11073 universal healthcare data standard, which covers all healthcare data. Without this standard, none of our other non-diabetes related data can be synced, and in the long run, this will only hurt us.

    I have been disappointed too often with the open-source attempts of data management in general. The situation can and will get messy, even with commercial partnerships. Conform to a standard and stick to the plan.

    Without using the standard, ISO/IEEE 11073, all of these companies will be unable to make the data generated from healthcare related devices engaging.

  3. Simon Carter
    Simon Carter May 30, 2013 at 5:05 am | | Reply

    Guys – all this has been done, and much more – at

    An open platform, Android app, desktop app, browser access etc etc.

    Live sharing, PREDICTION of BGLs, live dose coaching, bolus wizard etc.

    What you guys have to do is stop reinventing the wheel – but use the one that is already there.

  4. Type 1 Electrical Engineering Student
    Type 1 Electrical Engineering Student May 30, 2013 at 9:32 am | | Reply

    Simon, while this seems cool, it doesn’t do everything I need it to do. I can program, why do I need somebody else’s code to run my data through some code somebody else wrote? I can also write as many programs with as many “features” I need for the data I use, because I can program.

    Also, the API is not streamlined, it does not work with the actual uploading devices (USB dongles, USB Cords). You have to upload the data to the corresponding diabetes management software, export the data, and go from there. That creates unnecessary steps.

    Amy and the Diabetesmine team need to learn the difference between, open-source, platforms, and standards. When the team communicates issues about diabetes data and advocates for any of the above, there is all sorts of misinformation spread, when referring to these technical terms. I don’t even think the team knows what these definitions entail from a professional standpoint.

    The individuals who work on this site have good, genuine, intentions. I also understand that they are journalists and that may present limitations.

    However, no individual advocating for such progress can use such technical terms in articles such as, open-source, platform, or standard (technical), without knowing exactly what the terms mean and what they entail from a technical standpoint. Otherwise, an individual just looks unprofessional to a bunch of engineers and scientists, who are responsible for bridging this gap.

    I suggest, at the very least, consulting Wikipedia for the differences between open-source, platform, and standard (technical).

    Open-source (far from ever being a or having anything to do with a standard, which would be needed for FDA approval of any devices or software that works directly with diabetes software)

    Computing Platform (Which can never be open source, because anyone could change the code, whether this is the user who willingly does it, or if it is malicious code. This could cause potentially fatal dosing errors. Also, without adopting a standard such as ISO/IEEE 11073, the computing platform would be prone to programing errors, lack optimization, and thus would be limited in terms of integration and what could be done with the data)

    Platform Technology (Basically “product development” and nothing else. Has nothing to do with standardization or technical standards, for that matter. Think about “features” a device has to offer, but without the standardization of how it works.This is useless without adopting a technical standard, especially when significant regulation is involved with the FDA.)

    Standard (technical) (A definition that every person writing about technology in general needs to know.)

    ISO Standard (International Standards Organization standard, which is the standard that has been adopted by both the ISO and IEEE in ISO/IEEE 11073. Without worldwide adoption and use, a standard will not be successful. Please read what an ISO standard complies with, which is interesting.)

    ISO/IEEE 11073 (The only technological proposal that seems fitting for all of the devices that we use for diabetes and does not seem problematic in general, because this is a standard. The standard applies worldwide, which is advantageous.)


    I hope that this clarifies some of the issues that surround this matter. I do not think that collaboration is the issue here. The matter here is a lack of education about technical information and definitions.

    1. Ben West
      Ben West July 23, 2013 at 2:57 pm | | Reply

      I agree, EE student, which is why all of Tidepool’s source will be open and accessible, pending our license.

      When that happens, we look forward to reviewing any forthcoming patches!

      We also hope to bring about reference hardware + software bench-top models and prototypes, all open source. Perhaps you can help do some bring up. Get in touch with me!

    2. Nathan West
      Nathan West July 26, 2013 at 12:19 am | | Reply

      @ EE Student — As a fellow EE student I
      a) agree with you
      b) would be up for collaborating on some projects. Feel free to e-mail me or checkout my github (I’ve been a little busy with other things this summer so you’ll have to excuse lack of activity relating to diabetes…

  5. Why Tidepool is Open and Non-profit
    Why Tidepool is Open and Non-profit January 6, 2014 at 9:26 pm |

    [...] Tendrich at “The making of a D-Tech Startup: Tidepool’s Universal Data Platform,” May 28, [...]

  6. Everet
    Everet October 21, 2014 at 10:44 pm | | Reply

    Great discussion.

Leave a Reply