ResearchKit: A software framework designed to benefit medical research | Hacker News Reader

ResearchKit: A software framework designed to benefit medical research

apple.com

348 pointsbrbcoding123 comments

Great job transitioning from buying Coke using Apple Pay to managing diabetes in ResearchKit. Vertical integration at its finest.

Coke also sells water. Pretty funny though regardless.

Of course it does. Carbonated and flavored.

Well, they also bottle municipal water and sell it straight up (dasani[1] i think?). I certainly agree that the flavored water, being still sold as "water", is a pretty fucking awful trend though.

[1]: http://en.wikipedia.org/wiki/Dasani#United_States

Based on my rough calculations, at a 1,333,333% markup.

http://thetyee.ca/Opinion/2015/02/24/Nestle-Pays-Nothing-to-...

$2.25 for a million liters of water? Wow, that is crazy. This really exemplifies the age of big-corp we currently live in.

This is actually an improvement, in that before this legislation it was free to bottle your own water in BC. This price isn't nearly high enough, but it's better[] than before.

The last estimate I saw was that Nestle was going to pay about $550/yr to bottle and sell our water back to us.

Intel is marking up sand at a couple orders of magnitude greater than this.

I wonder how much Intel pays for its sand.

Silicon ingots run to about $70-$200/kg according to Alibaba. I think I read that Intel buys their silicon from a specialist supplier rather than grow their own ingots.

Look at it another way: the incremental unit cost of flavours, colours and carbonation probably isn't much. So really, every bottled beverage is primarily driven by logistics, marketing and consumer willingness. I doubt water is any more profitable than cola.

The difference with plain water is it's pretty much a true commodity product; assuming free-market economics, if it really were feasible to lower prices it would have already happened.

And don't forget the "premium milk"

http://www.usatoday.com/story/money/2015/02/03/coca-cola-cok...

Parks & Recreation made fun of "luxury milk" in a recent episode. It pains me to learn that they were parodying a real thing, not just making a joke.

(clip: https://www.youtube.com/watch?v=46iPFTCZ9LA)

I think I just premium vomited in my mouth a little. :(

It sounds to me like they're building an internal Truevault for managing patient data.

If you were in charge at Apple, would you have made Apple Pay unable to buy a Coke?

That's a bit of a straw man argument. I would have chosen another product to highlight Apple Pay though. It's not just the irony in following it with a diabetes app, but that Tim Cook's predecessor had a well-know quote about selling "sugared water".

ah I didn't even realize you were referring to the video.

The link has since been updated, it originally pointed to the live stream. I can see how this thread (and probably a few others) doesn't make very much sense now.

Well, someday if you work really hard, you can make the leap from pundit on the Internet to making that call. In the meantime, can we make HN better and avoid making silly comments top voted? "Does anyone get the irony..."

This is probably the coolest announcement from the event, at least from the perspective of societal benefits.

Medical research faces a real challenge of selection bias. Survey respondents and study participants largely come from two groups: 1) undergrads responding to a campus flyer, and 2) local university-town residents looking to make extra money. Obviously this means respondents are geographically homogenous, and in many cases socioeconomically homogenous as well. Enabling a wider reach for researchers will not only increase sample size, but also eliminate selection bias. This is a very, very good thing.

Also, now research isn't limited to only university researchers.

I wonder if the plan is to allow participants to earn credits on apple pay.

Considering the underwhelming nature of healthkit and the rather rudimentary substantial use-case, I am rather reserved.

We've had similar stuff like ResearchKit in existence for quite long. One of the most popular of such sites is patientslikeme.com .It and others did a great job around this, but usage is relatively rare.

The hard thing about such sites is:how do you incentivize people to contribute really personal information and do so on a regular basis(even if said action takes a relatively long time) ?

So i wonder what can Apple do to make this activity common, that other companies couldn't do ? How will they achieve this ?

> So i wonder what can Apple do to make this activity common, that other companies couldn't do ? How will they achieve this ?

A large marketing budget helps, e.g. see the video they launched with. Making it easy to use and frictionless would also help (no need to log into the website, no need to figure out how to sign a PDF or download a consent form or whatever the normal user flow is).

What concerns me is people gaming the apps; downloading them and filling them with fake data just to be malicious. You fix one problem and introduce another...

There's a huge opening here if pervasive sensors ever become consistent, calibrated, and easy-to-use-properly enough that they can supplement in-clinic data streams with at-home ones. This would open up the research contribution opportunity to people who cannot participate in clinical trials today due to distance from trial sites---this is something like 99% of the people who might feasibly be interesting to a clinical trial.

The big step in clinical research is the PIII trial and these are almost universally held up in enrollment. If this stuff was pervasive and trusty enough to be used in even a fraction of trial therapeutic areas it would revolutionize research and medicine in those indications.

All that to suggest the opportunity. I'm not exactly bullish on Research Kit tbh though. The "if" in my first sentence still appears to be a ways out and the kind of "throw the kitchen sink at it and average everything out in post" analytics championed by big data/machine learning of today doesn't fly well with the FDA.

At least not yet.

I don't see the similarity with patientslikeme.com

Apple has created a UI toolkit to enable researchers to directly get data from participants via an app. All control is in the hands of the researcher creating the app.

And the app will be used for traditional research studies, the only difference is that participants enter data via their app.

You would move selection bias towards users who can afford the Apple Watch. This will not be inherently worthless data, but will be biased nonetheless.

Apple watch is not required.

I think having millions of participants in most cases is very worth having the bias that it will introduce. It'll be better than the middle class white kid bias that already exists in a lot of this research.

Selection bias will now be those who have Apple products.

APPL said research kit will be open source, so if that is indeed true, then it will depend on those non profits if they will support android. if health kit gets ported to android.

I wonder how easy it will be to bias the samples for profit. Make some bots to submit bogus data that shows that people that consume product XYZ are healthier. Or, just make the health apps bias the data. Like say company XYZ gives kickbacks to "RecordYourActivity" app in exchange for reporting something that suggests MySnakeOil is good for you.

Of course I guess that kind of thing happens already. Just wondering will this make it worse, better or about the same.

Responses will likely be correlated to iPhones and AppleIDs in some way, so the spammers won't be able to do much.

On the flip-side this would be great advertisement information.

spammers maybe not but what about just subtly reporting false/bias results in otherwise legit apps?

The Parkinsons app, for example, correlates symptoms and physical activity. I don't know who would have an incentive to bankroll data skewing.

The current crop of apps have nothing to do with rating products.

The page is up for ResearchKit:

http://www.apple.com/researchkit/

The Apple Watch would likely be an ideal tool to improve prescription compliance in older patients. Perfect for issuing reminders about time and dosage (and for getting a confirmation that the task has been completed).

They just have to remember to recharge it daily

Yeah, this wouldn't work for my dad who has Parkinsons. The watch would be dead and he'd still need reminders for pills...

Thanks. We changed the url to that from http://www.apple.com/live/2015-mar-event/f407b6b3-5575-4069-....

We've actually already solved this problem pre Apple Watch. It's called Blister Packs, which basically is an arrangement of the medications of when a patient is supposed to take them.

Blister packs help, but they won't buzz on your wrist at the right time to take the meds (with a text reminder indicating what all of the buzzing is about).

Blister packs are great for figuring out which doses you forgot.

And that's with nominally functioning adults.

> Perfect for issuing reminders about time and dosage

It wouldn't surprise me if that needs FDA approval.

There's already apps that do that, some targeted to specific conditions.

The FDA has, as far as I'm aware, reserved judgement as to whether they will regulate apps like this.

People rich enough to own iPhones is an interesting confounding variable for health studies.

Not sure how bad that is, though. After all a lot of studies just employ university students

Though I'd imagine if a study requires the use of these sensors, an Apple Watch may be cheaper than something from a medical device manufacturer.

Android compatibility looks feasible since the kit will be open source [1]. That should open up the ResearchKit demographic quite a bit!

[1]: https://developer.apple.com/researchkit/

Open? Like Facetime was supposed to be?

Why does this one keep getting trotted out, lets assume unless they get sued we're fine.

Blame the patent trolls, not Apple for that.

Not quite so feasible if the kit is written in Swift.

As the spec for Swift solidifies, we'll see open source implementations. I wouldn't be surprised if Apple ends up releasing Swift as open source eventually as well.

If a person lives in the States and is willing to subsidize, iPhones are about as cheap as a lot of other offerings.

True, though you can always loan equipment to participants.

This study uses iPads for cognitive training in patients with Schizophrenia:

http://woolleylab.ucsf.edu/forecast

Personal use of the iPad is one of the perks of participating.

Regardless of any new constraints, this new method of acquiring necessary data is an improvement on the current state of affairs.

I agree. You would have to have Android support to have a more representative selection. Some Android phones are cheap enough they could be purchased for participants as well.

iTunes links to the first five Apple ResearchKit apps:

Asthma Health by Mount Sinai

https://itunes.apple.com/us/app/asthma-health-by-mount-sinai...

Share the Journey by Dana-Farber Cancer Institute, Penn Medicine, UCLA’s Jonsson Comprehensive Cancer Center, and Sage Bionetworks

https://itunes.apple.com/us/app/share-the-journey/id97218060...

Parkinson mPower study app by Sage Bionetworks, University of Rochester, Beijing Institute of Geriatrics, and The Michael J. Fox Foundation for Parkinson’s Research

https://itunes.apple.com/WebObjects/MZStore.woa/wa/viewSoftw...

GlucoSuccess by Mass General

https://itunes.apple.com/us/app/glucosuccess/id972143976?mt=...

My Heart Counts by Stanford Medicine

https://itunes.apple.com/us/app/myheart-counts/id972189947?m...

As a Type 1 diabetic, I can't wait to see what happens next. Well done Tim!

Open Source, wow that was unexpected from Apple

It's calculated. Reception would be distinctly negative if something as crucial as health research were closed source—here, the utility is in the transparency as well as apple's traditional offers of usability and reliability.

Plus, these are all academic partners - that code usually needs to be open source to publish.

It's not "usually". More like "rarely". Papers without code are very common.

> Reception would be distinctly negative if something as crucial as health research were closed source

I don't know about that. HealthKit is essentially closed and no one seems to mind, and it's all about the data ultimately anyway. Maybe it's a regulatory thing? Regardless, it's nice to see.

HealthKit is a glorified database. Would definitely like to see it open sourced, though.

Can you name an app that is not "a glorified database"? Seems to apply for almost anything. Hacker News, Facebook, Twitter and so on... It's all about putting in and getting data from a database

> Can you name an app that is not "a glorified database"?

Apps, sure. Tools? Hell no. File processors, browsers, libraries.... But HealthKit adds virtually no value aside from the schema of health data. There's not much to be gained from opening it. Honestly, I'd argue the same about hacker news and reddit: it's great that reddit is open source, but it's hardly necessary for me to perform research on it because the semantics of loading/storing/existing are so well defined.

HealthKit is not an app.

What's the point about speculating about motivation? Whether motivated by altruism or calculated strategy, who cares?

These kinds of discussions tend to devolve into platform bickering rather than substantive commentary.

> These kinds of discussions tend to devolve into platform bickering rather than substantive commentary.

True. However, occasionally (I at least) learn something new from these conversations.

Why would that be unexpected? Apple does plenty of things open source, just not GPL.

Apple usually only does things open source when it's required of them. I can't think of one piece of useful Apple open source software that's open.

I'd consider Webkit pretty useful.

clang too...

WebKit has been open sourced 10 years ago. CUPS is even older than that. What has Apple been up to recently?

---

You know, you could name projects instead of clicking the upside-down arrow. Clang comes to mind (not LLVM). Google and others contribute heavily to clang, but Apple started it. They deserve credit for that. Anything else?

Vikram Adve and Chris Lattner started LLVM at the University of Illinois in 2000. Only 5 years later, Apple got involved and hired Lattner.

Webkit was derived from KHTML, which is LGPL.

However, the bulk of modern Webkit is BSD 2.0, using LGPL libraries. Apple (and the other contributors) were under no obligation to contribute that bit.

Don't say forget CUPS

Clang/llvm

CUPS

Keep in mind when responding that Webkit, Clang/LLVM, and Cups are all not created by Apple, they're all preexisting projects or forks that essentially necessitated doing their work in the open.

Noted, but I'd argue that contributing to existing open source projects is better for the general populace than reinventing wheels.

Clang/LLVM is BSD, what do you mean by "necessitated doing their work in the open"?

He probably means that making it open source would benefit Apple moreso than if they kept it closed, not out of the goodness of their hearts

> they're all preexisting projects or forks that essentially necessitated doing their work in the open.

That's true of Cups and part of WebKit (most of the actual browser engine is BSD-licensed, but there are LGPL libraries used); however, Apple was certainly under no obligation to contribute its changes to LLVM back, or to contribute clang back _at all_.

Apple owns CUPS though, I highly suggest that you read the license terms[1] for it now:

License Exceptions

1. (a) Software that is developed by any person or entity for an Apple Operating System ("Apple OS-Developed Software"), including but not limited to Apple and third party printer drivers, filters, and backends for an Apple Operating System, that is linked to the CUPS imaging library or based on any sample filters or backends provided with CUPS shall not be considered to be a derivative work or collective work based on the CUPS program and is exempt from the mandatory source code release clauses of the GNU GPL.

[1] http://www.cups.org/documentation.php/doc-1.5/license.html

AFAIK Clang (not LLVM) was entirely created by Apple and could have been proprietary

Also the x86 Darwin kernel was made open source with no benefit to Apple, and winocm managed to port it to ARM too (although it's pretty debatable if it actually provides anything useful)

Why does that matter?

>Apple does plenty of things open source

most of the time, it's broken or need closed source dependencies.

Kind of a no-brainer if universities are going to be involved in building apps for it.

Wonder if they announced anything about external addon biosensors?

We'll see how this stuff actually works in practice pretty soon, but as this is more than just a chunk of code to run on a device, what exactly is going to be "open" here?

A formal standard that can be implemented and used by other platforms, both on the device and server side, would be great. But Apple's last promise to make something they developed "open" (iMessage as a standard) evaporated and has since been completely proprietary.

No, it was facetime they were going to make an open standard. And patent trolls killed that. So..

iOS and Mac OS are closed systems with lots of open source components.

Open Source in this case might not make much of a difference. If this is written in Swift and only runs under iOS, or if it ties with a certain Apple service that requires certificates issued by Apple it won't make much of a difference if you can read the full source code, even if it's released under an MIT or similarly open license.

As long as the data this gathers can't be used (or contributed to) by a third party, it's still a 100% Apple property, regardless of code availability.

This to me if it's even close to the promise it has is the real news of that event. The Watch will of course help the data collection.

Summary for those of us who don't have access to a mac right now?

They are streaming it on twit, that's what I've been using: http://live.twit.tv/

It is an open-source framework allowing medical researchers and professionals to interact directly with patients via iOS applications. Patients will be able to perform tests, fill out questionnaires, etc., and have the results sent directly to researchers.

Sure, open-source is a good thing. I strongly support OSS and as much as possible run systems at work using FreeBSD, Linux, etc.

I've worked in medical practice (and research) for a long time. Considering the vital importance of transactional transparency in carrying out the mission of these fields, as useful as open source is generally, I think the impact in medicine is even greater.

But open source is only part of it, open platform is at least as important. I guess I'm not a very trusting soul. Frankly, I don't believe Apple is going to do anything primarily for our good. Rather to my senses the announcement has the flavor of an attempt to gain a toehold or increase market share in the big money research domain.

If Apple (or other player) truly wants to generously donate to the public cause, that's wonderful. Then why not support efforts to develop open source, cross-platform apps, assuring all medical research can benefit? Not directed only or in particular at Apple, it's an issue I've raised whenever I have the chance: open source/platform agnostic development can save money and produce more reliable and secure systems.

As someone about to start a medical residency in the Bay Area who is interested in developing projects to work on a research platform like this, could the HN braintrust recommend a resource for learning the basics of whatever developer tools or frameworks that would be required to work with ResearchKit?

Research works best when the subjects of long term studies can see and react to the data they are generating in real time. Anti-blind trials are the future. /s

Here's the developer overview:

https://developer.apple.com/researchkit/researchkit-technica...

It appears to mostly be an iOS UI toolkit of sorts for easing the creation of individual apps to collect research data. The hard work must still be done by each individual research project.

Asking to learn.

How is this better an an online polling software - like surveymonkey or polldaddy?

From reading the press release, it seems like it is a data collection software from patients. Is there more to it?

I think the benefit is that the iPhone has sensors that can collect lots of data. For example, the Perkinson's(?) app makes you tap on two targets quickly and say "aaaaa" for long and analyzes the ups and downs in vocal cords using the mic.

Patients can't track all this by themselves. Plus, it will automatically send it to researchers. Much easier than making someone enter information on surveymonkey or polldaddy.

There's certainly going to be a selection bias, as others have said. I think the most exciting opportunity though is for very fine grained and longitudinal studies.

Anyone here who has already experience with it? Apparently a handful of hospitals and universities got early access.

They would most likely be under strict NDA until the public release.

This is so awesome. If this takes off, medical advancements in the next 10 years are going to be astronomical.

The FDA has to get more efficient, too

Beyond being more efficient, they need to have a shift in thinking as to what constitutes a quality SW development process. Example: as of right now, they still think that a waterfall process is the only process that is "correct".

I'm not sure. There were several medical crowd sourcing sites, with patients like me quite successful, but only small incremental advances. Not sure if it's because real research needs more than patient symptoms - it also needs objective stuff like blood tests.

We'll see.

Like we are going to find that exercise helps prevent diabetes, heart disease and dementia?

Wasn't FaceTime supposed to be an open-sourced framework/protocol?

That was before VirnetX sued them for $708 million for patent violations, reduced to $368.2 by the judge. To avoid having to license the patents in question, Apple re-architected FaceTime so it's no longer peer-to-peer and instead routes everything through central servers.

This makes gets unreasonably excited. Love that it is open-sourced

What is it?

"Sorry, your browser doesn't support our live video stream" Firefox 36.0

"Sorry, your browser doesn't support our live video stream" Chrome 40.0.2214.115 (64-bit)

That's the reality-distortion field. It _should_ say, "Sorry, our live video stream doesn't support your browser."

The ultimate example that the cure of X also come from programmers, not solely doctors.

Reply on news.ycombinator.com