Monday, October 05, 2015

#Health2con – Data Sharing without Trust – Ran Gilad-Bachrach

Health 2.0 Conference

Realtime notes from the ninth Annual Health 2.0 Fall Conference at the Santa Clara Convention Center.
Here is the online Agenda:
http://2.healthca.mp/1WIDOXD

Health Data Exploration Project-Personal Data for the Public Good

Moderated by Bryan SivakU.S. Dept of Health and Human Services

The Health Data Exploration project, sponsored by the Robert Wood Johnson Foundation, is a network of academic, public sector, and corporate partners working together to conduct research in how personal health data can be used to answer public health questions (aka personal data for the public good). Join us for this interactive dialogue to learn out about the HDE project as we accelerate the development of new collaborative commercial-academic opportunities enabling precision medicine. You’ll hear about insights from industry leaders about how their companies are collaborating with researchers; learn about opportunities for public and private sector partnerships (and why it’s good for business); and have a chance to participate in a dialogue with leading figures in creating novel industry-academic partnerships.

Please complete this short form if you’re interested in attending this session.

Speakers

Kevin Patrick
Calit2
Carlos Rodarte
PatientsLikeMe (USA)
Jerry Sheehan
Calit2
Aaron Coleman
Fitabase
Stephen Friend
Sage BioNetworks
Ran Gilad-Bachrach
Microsoft Research

Sharing with understanding of risk

Name, Date of Birth and Zip Code allows 87% of US Citizens to be identified.

Its all about Trust – Can we take Trust out of the equation in data sharing

How do we make predictions on the cloud accurate yet private?

Send Encrypted data to the cloud – Without the keys.

Cloud can make an accurate prediction but can’t read the data.

How do you do that? Fully Homomorphic Encryption.

This allows addition and multiplication while encrypted with the result being encrypted.

Combine this with Neural Networks

Differential Privacy

Add a Privacy Guard between database and Querying Analyst

The Privacy Guard adds “noise” so that an individual can’t be identified.

This doesn’t help for individual predictions. It does work for populations.

The Privacy Guard can alert when the number of queries from a specific point creates a risk.

This is all at a research and laboratory phase.

Steve Friend – “EMR data is not worth much”

Rich data over time in context with other data is Far more valuable.

Ran: Just knowing measurements doesn’t change behavior.

Think about the interventions that are necessary to change behavior.

[category News, Health]
[tag health cloud, bluebutton]

Mark Scrimshire
IT and Health Data Ninja

Mark is available for challenging assignments at the intersection of Health and Technology using Big Data, Mobile and Cloud Technologies. If you need help to move, or create, your health applications in the cloud let’s talk.
Blog: http://2.healthca.mp/1b61Q7M
email: mark@ekivemark.com
Stay up-to-date: Twitter @ekivemark

Disclosure:
I am currently HHS Entrepreneur-in-Residence working at CMS on an assignment to update BlueButton for Medicare Beneficiaries. This involves creating a Data API. Watch out for more about BlueButton on FHIR.

The views expressed on this blog are my own.

I am also a Patient Engagement Advisor, CTO and Co-Founder to Medyear.com. Medyear is a powerful free tool that helps you collect, organize and securely share health information, however you want. Manage your own health records today.

Medyear: Less Hassle, Better Care.



via WordPress http://2.healthca.mp/1RrcV7H