These are chat archives for nightscout/intend-to-bolus

18th
May 2015
diabeticgonewild
@diabeticgonewild
May 18 2015 03:28
My electrical engineer cousin who works on medical devices said he had already heard about the stuff @bewest is doing and he wanted to see it. He said the FDA was very unhappy about what we are doing and that happened in his sector that the FDA would shut them down. He was telling about what they were doing to keep their devices secure and he said that Medtronic clearly messed up based on how I could just read the pump and stuff and how the pump handled it...not my problem.
He used to work at Medtronic but not on insulin pumps...
• if it happened in his sector
He was pretty shocked. LOL
diabeticgonewild
@diabeticgonewild
May 18 2015 03:38
But it's really not my problem or ours.
Jason Calabrese
@jasoncalabrese
May 18 2015 03:41
from what I've heard that isn't the case
diabeticgonewild
@diabeticgonewild
May 18 2015 03:42
Hey he was told that and "not to mess up" with the security of whatever device he is making...
It was a "dinner conversation", just FYI
diabeticgonewild
@diabeticgonewild
May 18 2015 04:48
Just watched a sunset on the Periscope app on my iPhone from Kona on the big island of Hawaii. Pretty cool use of Periscope. http://m.imgur.com/wWoAAVG
Made me happy.
diabeticgonewild
@diabeticgonewild
May 18 2015 05:13
And my joints hurt like hell. FML.
Even shoulders hurt. I hate being this way.
Ben West
@bewest
May 18 2015 06:29
I've talked with FDA firsthand and know they are not unhappy about this
since I've talked directly, first hand, with upper level of management at FDA, I'm hesitant to take other people's word for how FDA feels about this
they actually said informally they want more vendors to be like us
Oliver Schumacher
@oschumac
May 18 2015 06:50
But big question would be what we have to do to get their blessing. To roll out diy med devices.
Scott Leibrand
@scottleibrand
May 18 2015 07:01
If it's truly DIY it doesn't need their blessing. It's when we're distributing a medical device that we do.
Ben West
@bewest
May 18 2015 07:23
:+1: distributions is the issue, WSJ outlined this nicely
it's clear there is a "medical device" that is "being distributed", but it's hard for anyone to tell at what point a "device" is "distributed"... this grey zone leads to chilling effects
suppression of free speech, etc
there's no warehouse where parts come in on one end and products leave on the other
so distribution doesn't happen from a facility per se
the grey zone also allows us to exploit the ambiguities at the same time
we can distribute isolated, independent bits of code
we can describe what it might do
but we can't ship units
and even distributing runnable binaries is questionable
Oliver Schumacher
@oschumac
May 18 2015 08:46
yes the question would be when it starting to be considert as an medical device. Bevor it is only bits and bytes in the cloud.
Holger Sanft
@holgi13
May 18 2015 12:04
hello everyone, does anyone know how it hurt the behavior of international patent law? could we get there problems?
Matthias Granberry
@mgranberry
May 18 2015 12:09
Not as individuals
Technically maybe
diabeticgonewild
@diabeticgonewild
May 18 2015 13:52
I was just relaying the message...I don't really care what he thinks. I mean I have already done my part.
With OpenAPS woek
Scott Leibrand
@scottleibrand
May 18 2015 16:17
At some point patent lawyers might get interested, but not unless they consider us a commercial threat, which we don't want to be.
diabeticgonewild
@diabeticgonewild
May 18 2015 16:43
But yeah my cousin was shocked...
Holger Sanft
@holgi13
May 18 2015 16:45
@scottleibrand - I think that too.
Matthias Granberry
@mgranberry
May 18 2015 16:47
yep. That's what I meant too. Patents don't mean a lot unless someone is making money. As for the international part, international doesn't matter in patent law. Patents are all local protections by design.
Holger Sanft
@holgi13
May 18 2015 16:48
It would be interesting than really, what are the conditions for a medical device? How could we use the gray zone at best?
diabeticgonewild
@diabeticgonewild
May 18 2015 20:07
Will be out of Seattle in an hour.
Matthias Granberry
@mgranberry
May 18 2015 20:41
I think the plan is to distribute the pieces necessary to implement an APS, but no instructions for assembling them. A bunch of pieces isn't an APS, only tools to manipulate data, so it should fall outside of the domain of regulation and inside the domain of research tools
diabeticgonewild
@diabeticgonewild
May 18 2015 20:48
@mgranberry I am going to provide an image of Ubuntu with OpenAPS installed and SSH and WiFi configured for people with vision impairments so they can access it via the respective OS they are using via a screen reader, but for mainstream users, no, as this is an accessibility issue due to no screen reader installed default while installing Ubuntu on the Pi. It's a complicated issue.
Oliver Schumacher
@oschumac
May 18 2015 20:49
yes, i think the same. The question is how small the pieces should be. I like the idea to provide some kind of image of Raspb or any other system only containing working interfaces. Maybe on another repo somebody can find. Somekind auf algorithm which works on the image running the interface. In this case somebody put the pieces together and do whathever they want to do with.
diabeticgonewild
@diabeticgonewild
May 18 2015 20:49
Cabin door just closed on the airplane. Talk to you in 4:30.
It's for people who use screen readers only. They can't access it.
Everyone else can figure it out on their own.
The blind T1Ds need OpenAPS as its really the only interface to access the screen/data on the pump.
diabeticgonewild
@diabeticgonewild
May 18 2015 20:54
It's a different situation for them
Matthias Granberry
@mgranberry
May 18 2015 20:55
@diabeticgonewild that's a lower-risk category. Very unlikely for anyone at the FDA to get upset. Absolutely no reason not to go and post it.
I need to implement accessibility API support for xDrip and the android-uploader if it hasn't already been done. There's no excuse not to in this community.
Ben West
@bewest
May 18 2015 20:56
:+1:
Oliver Schumacher
@oschumac
May 18 2015 20:56
@diabeticgonewild What do you need ?
Ben West
@bewest
May 18 2015 20:57
actually, I like idea of integrating braille display in with e-ink as part of the housing/case
diabeticgonewild
@diabeticgonewild
May 18 2015 20:57
Can't talk. Supposed to be in airplane mode and plane is taxiing.
Matthias Granberry
@mgranberry
May 18 2015 20:57
@diabeticgonewild It is when there are automatic dosing decisions etc that it gets riskier.
Ben West
@bewest
May 18 2015 20:57
using tactile feedback of some form will reward the expert users
require less attention/looking
diabeticgonewild
@diabeticgonewild
May 18 2015 20:58
@bewest they also need Bayer contour link integration with OpenAPS. Got a request for that for the blind but really got to go. It's important as they need access to a meter on OpenAPS
Matthias Granberry
@mgranberry
May 18 2015 20:58
sounds expensive re the braille display. Both in terms of battery and device cost.
diabeticgonewild
@diabeticgonewild
May 18 2015 20:58
Integration is important
Oliver Schumacher
@oschumac
May 18 2015 20:58
Yes a bought a vibration motor for Edison. But did not start to implement.
diabeticgonewild
@diabeticgonewild
May 18 2015 20:58
They already have Braille displays
It's a driver issue
Ttyl. Especially @bewest regarding the contour meter integration
Ben West
@bewest
May 18 2015 20:59
that's the kind of problem I eat for brunch though
I can add contour to openaps easily
anyone can, actually
driver is already in python :-)
was thinking, I could touch the contours of the device, eg in my pocket, and the touch would tell me something about status
but maybe more info than fits on pebble
Dana Lewis
@danamlewis
May 18 2015 21:01
THinking about what people have been talking about re: haptics & apple watch
Matthias Granberry
@mgranberry
May 18 2015 21:01
@bewest that's an interesting idea, because you would begin to internalize it over time
Ben West
@bewest
May 18 2015 21:01
the apple thing seems wrong direction, re "pinching" with motors
everyone seems to devolve to talk about wanting to "pinch" their teenager
that is really unsettling to me
Dana Lewis
@danamlewis
May 18 2015 21:02
I read a really good article from someone with visual impairment about her experience with it
Ben West
@bewest
May 18 2015 21:02
ooo, would be interesting
but if I can touch it in mypocket, and the texture of the contours or patterns/braille on it could tell me something, that's cool
Dana Lewis
@danamlewis
May 18 2015 21:03
True
Matthias Granberry
@mgranberry
May 18 2015 21:04
I have a friend who made a compass belt out of vibrator motors and said that, after a few days, she no longer felt the vibration and instead just had a killer sense of direction right up until she got near an electromagnetic lock and got disorientated
Ben West
@bewest
May 18 2015 21:04
oh, that's fascinating
ok, that's getting into augmenting mind/sense
Matthias Granberry
@mgranberry
May 18 2015 21:05
I suspect a BG indication from a similar device would be an interesting biofeedback tool
Ben West
@bewest
May 18 2015 21:05
there's been some great TED talksk ont hat
Matthias Granberry
@mgranberry
May 18 2015 21:05
or maybe a rate indicator
Ben West
@bewest
May 18 2015 21:05
rate
yes
rate/direction
but raw values, not some bs "trend"
my bayesian brain will figure it out
that's interesting
bio machine learning
maybe that's what @danamlewis was thinking too, I had not considered that fully
Dana Lewis
@danamlewis
May 18 2015 21:08
Yes
Matthias Granberry
@mgranberry
May 18 2015 21:14
@bewest to really train the brain as a new sense, the input needs to be fairly frequent unless there is some internal sense that you're cuing the brain to pay attention to. I'm not sure how much of a sense there is of BG trends, although I'm fairly certain that there is some of it.
Ben West
@bewest
May 18 2015 21:15
just getting fed the deltas, bg, and raw as haptic feedback every 5 min?
combine with the belt idea maybe
Matthias Granberry
@mgranberry
May 18 2015 21:18
it sounds like too much information to me. Without a good way to encode the information, it will just be irritating
it sounds like something a cognitive science researcher or some random person in the MIT media lab would love to take up, though, and there might be some interesting results that come out of it. It wouldn't be too hard to make a belt and encode BG as location and rate as a pulse rate
Matthias Granberry
@mgranberry
May 18 2015 21:23
it's too bad making heat requires so much energy. It would be really interesting to present BG data as a warm link on a Pebble smart strap. It might be possible to do it with a linear actuator in a link, though, to push a (flat) pin up to provide some feedback
I wonder how small electropermanent magnets could be made
Matthias Granberry
@mgranberry
May 18 2015 21:34
Some of the electro-permanent magnets they use here would make a bracelet possible at low static power.
Ben West
@bewest
May 18 2015 22:48
my intel edison just arrived
but no blocks came with it, have to wait for those to do anything. this thing is tiny
Matthias Granberry
@mgranberry
May 18 2015 23:04
I just looked at idle current figures for the Edisons. Those might actually be useful as mobile devices.
Ben West
@bewest
May 18 2015 23:05
looking forward to playing with the cpu governor
maybe put it in cycle where it's off most of the time
Matthias Granberry
@mgranberry
May 18 2015 23:07
USB devices are probably going to be the big problem. Carelink sticks are pigs.
Ben West
@bewest
May 18 2015 23:10
there's a constant pull of 5v I believe, whether or not stick is in use
I haven't tried the usb power control yet
setting that sys property