Stephen TranovichOkay, let's get started! A big HELLO to @Curt White for joining us today!
Hi everyone!
Curt, why don't you get us started by telling us a little about yourself!
OK
I build and research wearable devices related to mental health at CMI's MATTER Lab
link here: https://matter.childmind.org/
I'm also a long time mental health activist. I've been heavily involved with the Icarus Project for many years: https://matter.childmind.org/
should the 2nd link be different?
That's really awesome. Super important work.
http://nycicarus.org/ here you go
The Icarus Project
The Icarus Project is a support network and education project by and for people who experience the world in ways that are often diagnosed as mental illness. We advance social justice by fostering mutual aid practices that reconnect healing and collective liberation. We transform ourselves through transforming the world around us.
Read this on The Icarus Project
sorry
Yup
cheers
The "Tingle" gesture recognition device at CMI is what I've spent the vast majority of my time on
Could you tell us a little more about that project?
yeah, and that is a good segway into a more general conversation about devices and mental health
Perfect!
Mental health is extremely difficult to engage with at a device level - at least in terms of interventions
one of the more insidious things about mental illness is its ability to undermine problem solving
Yeah - I wasn't too surprised when ginger.io had to pivot - but I was very excited about their earlier model..
in other words, if I build some kind of device to combat depression, a person experiencing depression may be unable to use it
so before even building anything, I looked long and hard for problems that I could physically engage with
Heh. :-)
Quick Q, do you have a presence in the UK?
shouldn't it be the other way around? looking for problems that your device solves seems backwards
Trichotillomania is a compulsive disorder in which individuals pull their hair out
crucially, the initiation of this behavior is unconscious
ꝺeshipu - not quite - it's usually a good idea to look for problems that a "class" of devices could solve, before you start actually customizing them for it.
so that leaves us with an information deficit problem, if we can make individual aware that they are initiating that behavior (pulling out their hair) we can help them stop doing it
so we have a physical gesture and lack of information
this is a really promising situation from a device perspective - much more approachable than depression and the like
What the "Tingle" device that I'm working does is detect a gesture (raising your hand to your head) associated with a problem behavior (pulling out hair) and provide feedback (vibration motor) to alert them and make them more conscious of their actions (fill information deficit)
Finding the right problem is probably the most challenging step when it comes to devices and mental health
Anyway, I came up with all this on my own and then got hired to pursue it at CMI. This started as a straight-up hardware hacking project and graduated to a research project and now a medical device.
Wow, that's awesome. Really smart to start with issues that have very measurable physical characteristics.
This general class of solution also works for physical problems like balance and falls, tremors and vertigo, etc. The devices provide feedback to the user (or observer) to help tailor behavior.
the main rule of marketing is: if you can't find a good problem to solve, create it
That segway's into @Lutetium 's question:
How did you get your hacking projects to actually be put to use in medical contexts?
Hi Kris, I actually built something along those lines a few years ago!
Thanks Stephen. FYI this is something Stephen and I bounced a bunch of emails about.
Hi Curt White, do you have working examples for ID107HR device
I have been building medical device-related projects for over a decade. Mostly just for fun but sometimes on things that had a lot of promise.
I also see tons of very earnest medical device related projects on Hackaday.io
Doesn't that preclude our ability to deal with the most prevalent (and thus impactful) problems? We can probably go after the "low hanging fruit" - and it drives interest in the field, but at the same time, are there any usable indicators and potentially corrective measures for the more prevalent ones?
A medical device project (not lifestyle) can't go anywhere without published research
For good reason, people's lives are on the line
For a long time, this seemed insurmountable
Not necessarily. A device that targets patient mobility does not need FDA approval, for example. And there are plenty of affliction-based fora out there with potential users that don;t require a doctor's permission to self treat.
I really only got to where I am now with medical research though dumb luck
give me an example of a product on the market
if it isn't available to the people who need it it isn't real (although I love doing projects for fun)
That said, FDA did start a program for certifying "apps" (currently in pilot)
were talking about devices
I got a ID107HR working individually with HR sensor, BLE detail and OLED. But when I try to work with OLED and Ble Serial, my device stops working after 3-5 hours.
Well - somewhere in the middle. The device can be sold as general purpose (smartwatches, etc) - and the medical component added as software.
Fit bit is an example of a device on the market that required no FDA approval.
lifestyle, not medical
(and Fitbit is a lifestyle device. It was never recommended as treatment and isn't covered by insurance, refunded etc.)
you are right, lifestyle devices can be developed just like any other consumer device
What I am *very* excited about is the software upgrades to "lifestyle" devices that cross into medical.
Mobility is at the junction of lifestyle and medical, it tracks patient movement, activity for the doctor's use as well as the patients.
Let's keep the focus on mental health hacks here, as that is the nature of the discussion
for the tingle device you mentioned you created, did you come up with the idea, and then run medical trials at CMI later?
lifestyle improvement devices are awesome, and they may be the best path for hackers, but.........
I came up with the idea
@Curt White so how did you get your devices into research?
I had been working with my current boss on some brain modeling software and we started chatting about devices
he was impressed by some of my past projects so I decided to build something and sell him (and the org) on it
they liked it and hire me to build + research it
dumb luck, but it didn't have to be
knowing what I know now, I could pull off the same thing in a variety of contexts
OK
I'm just going to assume we're talking about medical devices that definitely require a clinical study
think heart stent
there is a lot of opportunity for hackers to work with researchers
Tell us, Curt. Tell us your social engineering trick to getting your hacks into research!
How sensitive is tingle to different gestures? In other words, could it be taught to recognize individual patterns of OCD behaviors as well, or are those to general?
(And then we'll get back into community questions. We have a few ready to go!)
1) Researchers are under intense pressure to publish as much as possible. Your project is their published paper. You do a lot of work which save them time + effort, you get a peer-reviewed publication which is crucial to raising money and getting traction (and eventual FDA approval)
by the way, I'm assuming a startup company or convincing someone to hire you to do the project, I don't know how else you would make any of this real
2) Provisional patents. A provisional patent will protect you from anyone who might want to steal your idea. It also protects potential partners! there is no reason for this to be adversialial
I wrote a provisional patent for the "Tingle" before presenting it to my employer. Transfering the patent to my employer was part of the hiring process
A provisional patent costs less than $100 and gives you a year until you have to submit a full application
Over the past year I've written several for new projects at CMI
neat, that's cheaper than i thought it would be
Hello. I joined a little late. My organization is just coming off the heels of a Mental Health summit we hosted. During the event, I moderated a panel on complementary and alternative modalities. The attendees were from the VA Medical Center in San Francisco. We had an interesting discussion on how non-traditional methods are researched in partnership with non-federal agencies. In this case, UCSF. I was highly interested in this project for many reasons.
3) Grad students, medical residents and MDs at university or university-affiliated hospitals. MDs at university-affiliated hospitals (there are TONS) are also under pressure to publish research. They have even less time than professional scientists.
These are the people who are really easy to sell on a project. For example, my wife is an MD at a university-affiliated hospital and I'm doing a research project with her on apps for MD continuing education.
but it could just as easily be a device
when she has chatted with other MDs about DIY device research they are very enthusiastic
Its really about developing a relationship
Apparently it costs £250 in the UK to file a patent. I thought it was more expensive...
https://www.innovate-design.co.uk
4) Scientists and MDs are experts at the problem. If I didn't have the input of MDs and scientists, things that took me a week could easily have taken months.
The VA has come a long way with this, and one of our participants gave the example of Yoga. VA and the federal government always need to be convinced with evidence. So they are starting to synthesize data, and one way they do this is by measuring physiological responses in the body.
Scientists and MDs are also experts at identifying the problem. Talking them would be extremely useful before even brainstorming about building a device.
OK, my 2 cents
questions?
(Yeah. The alternate pipeline is the Feds/EU. Between SBIRs and related grants, there is a truckload of contacts..)
That's such an awesome model for getting hacks into research!
(@zxzxzx0101^)
Thanks for sharing that! We have a bunch of community questions I will feed you for the rest of the chat
cool
If anyone has quesitons, add them here: https://hackaday.io/event/160158-hacking-for-mental-health-hack-chat
With this wearable technology, how is it able to capture and synthesize data?
@ꝺeshipu asks: I've heard about many ideas that use various form of tracking to early diagnose things like depression or bipolar disorder, and I think that's great, however, what makes me very uncomfortable with them is that they all collect and store the data on their servers. Do you know of any similar software where the data never leaves my own computer? I feel very uncomfortable with the thought that data about my mental health might be accessed by anyone else.
What exactly is being tracked?
Text content of email/social media? physical activity? sleep patterns?
Discussions