Caveat lector: This blog is where I try out new ideas. I will often be wrong, but that's the point.

Home | Personal | Entertainment | Professional | Publications | Blog

Search Archive


brainSCANr: Paper rejected and Health2.0!

(This is cross-posted from the brainSCANr blog)

Sorry for the silence these last few weeks! With Jess back in school and Brad back from break, we've had to refocus on our "real world" work again.

To give a brief few updates, the brainSCANr manuscript was not even sent out for review (hooray!) so we decided to do a few presubmission inquiries at journals to see if it's even of interest to anyone. We're sure the paper will find a home, we're just not sure where.

Also, we've set up a Google Group for discussion. You can join it here:

Bradley Voytek Jessica Voytek brainSCANr careMAPr Health2.0

This past Saturday, Brad and Jess spent all day at the Health2.0 Developer Challenge (aka the "code-a-thon") at the Googleplex. This event was intended to pair researchers, developers, designers, and folks from the health industry to provide rapid prototypes to address social and person issues in health and medicine.

Neither of us really knew what to expect when we got there, but it ended up being a really cool experience. We started with Brad giving a brief presentation to the crowd about brainSCANr and why we built it. This piqued the interest of a few people in the crowd, and soon we were off in a group discussing some other possibilities with data-mining PubMed.

After about an hour, we were met by Alex from MEDgle (another great heath data mining service), who recognized Brad from Twitter.

We quickly decided on a tool to map where the primary research on health topics was being performed, and one of the members of the group (Sean) dubbed this "careMAPr".

You can see the rapid demo prototype here:

Bradley Voytek Jessica Voytek brainSCANr careMAPr Health2.0

Right now this only works for ADHD for the USA, but the idea would be to allow anyone to enter any search term, and we would query PubMed for that topic. From the results we can identify where the primary research is being conducted. The user could even search by year to highlight only researchers who have recently published on a topic, for example.

The problem that we are trying to address with this site is how to connect the lay community with research specialists. The use case scenario we presented was this: imagine you are a parent whose child was recently diagnosed with ADHD. Maybe your kid was prescribed a drug as part of their treatment. But maybe your child also has another disease, and you want to know about research looking at the relationship between that disease and ADHD.

Right now, the best most people have is to do some Google searching (which can result in very dubious or even misleading information). You can talk to a psychiatrist, but maybe that doctor doesn't really know much about ADHD. How can you find a clinician in your area who does? We hoped that our site might serve as such a resource.

Anyway, the project was very fun, and we were all pretty surprised with how much we managed to put together in just a few hours. We're not sure if we're going to see this one through the the finish, but we hope to.

What do people think about the idea? Any suggestions?


  1. Dan H09:30

    The careMAPr concept actually sounds like it would benefit more from the ClinicalTrials.gov database than pubmed. Pubmed is full of research that was completely 1-many years ago, analyzed, and published. ClinicalTrials is a database of all active and closed studies that receive funding from any branch of the US government along with a lot of privately funded studies and some international studies. I'm not completely sure who is required to submit, but I suspect any trial that is part of the FDA approval process is in the data base.

    It already allows searches to be narrowed studies actively recruiting volunteers in a specific geographic location.

  2. We were initially choosing not to focus on *just* clinical trials, actually, because "clinical trials" are a very specific type of research that would limit the results. That said, if we could also incorporate that information, that would be nice.

  3. Dan H10:31

    I think it includes things beyond what you're calling clinical trials. For example, there is a whole category of observational studies with no required interventions. Looking into it a bit more, the big minus is that I don't think it includes everything. I think it includes EVERY NIH study involving human volunteers where someone in the NIH intramural program is a collaborator or leader, but not every NIH funded extramural study. Therefore it's a bit sparse, but it's still useful information.

    The other potentially useful database is: http://projectreporter.nih.gov/ which is everyone funded by NIH. Reliably finding grants with human volunteers might be a bit more challenging, but it looks like they've started a beta test of linking it to clinicaltrials.gov.

  4. Thanks for the post. I like the way Health2.0 Developer Challenge brings together developers and sponsors to create practical/tactical apps. I'm in Food Find challenge myself as a team leader for Food Prints.

    I like careMAPr; it reminds me BiomedExperts. Having a similar tool for patients would certainly be useful. I agree with Dan H that it would benefit more from the ClinicalTrials.gov database than pubmed. I have been working on http://clismap.appspot.com for the last couple of weeks. There was an interesting post published in Pixels & Pills that I recommend: e-Tools: Pharma Study Patient Recruitment Is Going Digital

  5. Wow, thanks for all the links, Sarkis!