Facebook CEO Mark Zuckerberg and his wife, Dr. Priscilla Chan, will contribute $10 million to UCSF to help fund an effort to merge data on 15 million patients across five UC medical campuses into one database.

The investment highlights the interest from investors and researchers in applying artificial intelligence to health data. The goal is to detect patterns in disease development and to allow doctors to better develop treatment plans for patients.

In oncology, for instance, computers could mine patient data to try to predict whether women diagnosed with ovarian cancer who stop responding to one type of drug may be more likely to respond to another type of treatment, based on previous cases.

The $10 million contribution is separate from the commitment by the couples limited liability company, the Chan Zuckerberg Initiative, to invest $3 billion over the next decade to cure disease.

It will go toward UCSFs Institute of Computational Health Sciences. In addition to merging data from health records, it will be used to hire faculty members for the institute over the next five years, said Dr. Atul Butte, the institutes director.

Big data and machine learning is hot in medicine right now, Butte said. If you want machine learning to work, you need to see many, many cases before you can learn the patterns.

The soon-to-be-merged data is from electronic health records that are housed separately at UCSF, UCLA, UC Irvine, UC San Diego and UC Davis, dating back between five and nine years, Butte said.

While UCSF would have access to identifiable patient information, such as names, patient privacy laws require researchers to get authorization from patients, or approval from UCSFs Institutional Review Board, before accessing any identifiable data.

Artificial intelligence in health and drug development is a booming area. Emerging companies like Londons Benevolent AI, San Brunos Numerate and Menlo Parks NuMedii co-founded by Butte have attracted hundreds of millions of dollars from investors over the last several years.

Alphabets health subsidiary Verily, formerly Google Life Sciences, recently launched a study to track health information from 10,000 people. IBMs Watson Health uses algorithms to sift through patient records and research papers from medical journals to help doctors diagnose and treat diseases. Amazon has assembled a team to build tools for electronic health records data, CNBC reported this week.

The major tech titans are moving into this space at full tilt, said Dr. Eric Topol, a professor of molecular medicine at the Scripps Research Institute. They realize this has unparalleled growth potential.

Artificial intelligence in medicine is taking off because until recently there wasnt enough data to draw meaningful conclusions, experts said. But improvements in genomic sequencing and medical monitoring technology are quickly changing that. Every persons genomic sequence alone generates billions of data points. Add that to the data collected by wearable devices such as monitoring tools that track glucose levels, blood pressure, heart rhythm and other measures and researchers have a rich pool of health information to parse.

Thats why this is a particularly exciting era in medicine, Topol said. Its really about having enormous data sets, not just a one-off, but on a continuous basis.

The challenge, though, is cutting through the noise in ways that will enable physicians to zero in on individualized screening, treatment and prevention plans for patients.

Just having all this data is not so important, Topol said. Its processing it, working with it to change the future of medicine. Artificial intelligence could result in this promise of true prevention or far better treatments.

Catherine Ho is a San Francisco Chronicle staff writer. Email: cho@sfchronicle.com Twitter: @Cat_Ho

See the rest here:

Zuckerberg, Chan give UCSF $10 million for health data research - San Francisco Chronicle