Sharing Skin-Selfies for Science: melanoma study releases participant generated smartphone data to research community worldwide

Mole photos, measurements, and melanoma risk factor data contributed by over 2,500 participants are made available by Sage Bionetworks and OHSU to accelerate skin cancer research.

February 14, 2017, SEATTLE–Sage Bionetworks and Oregon Health & Science University (OHSU) today publicly released data contributed by 2,798 participants in the Mole Mapper melanoma study. The app-based research study uses Apple’s ResearchKit to enroll participants who use the phone camera to map and measure their moles over time. Abnormal or changing moles can be an indicator of the skin cancer melanoma, so remote monitoring with the possibility of early detection holds great promise for cancer prevention.

Whereas most research data are generated in a clinical or laboratory setting, Mole Mapper is crowd-sourced by individuals contributing data to the study from their own phones. Curated Mole Mapper data, consisting of mole photos and measurements together with melanoma risk factors, have been made available to qualified researchers on Sage Bionetworks’ collaborative science platform Synapse and accompanied by a publication in Nature Scientific Data. This is the second such mobile health study that has been made broadly available to qualified researchers around the world.

“In designing the study, we first wanted to know if research run remotely and entirely through an app could find the same melanoma risks as years of rigorous epidemiology and genetics research,” said lead author Dan Webster, Research Fellow at the National Cancer Institute. “We show, for instance, that Mole Mapper participants with red hair were significantly more likely to be diagnosed with melanoma. This is in alignment with previously published data showing that people with red hair caused by mutations in the MC1R gene have a higher risk for melanoma.”

The study data also touches on a frequently asked question about moles: “Is this normal?” Stanford University researchers recently demonstrated that algorithms can accurately diagnose skin conditions by training on a large database of high-quality medical skin images. The Mole Mapper team aims to create a similarly foundational database from participant-contributed data. While clinical resources will undoubtedly be important in answering this question, most moles that are measured and monitored in a clinical setting are already suspect and may already be abnormal.

“With Mole Mapper, we have a unique ability to collect thousands of measurements from ‘pre-clinical’ moles that people measure themselves at home,” said Webster. “Over time, this can provide a basis for mole size and shape distributions to serve as a new benchmark for future studies.”

The release of the Mole Mapper study data is a part of the larger mobile health ecosystem that Sage Bionetworks is cultivating. Developing open-source modules for integration into mobile applications and enabling the broad sharing of the resulting data are cornerstones of this effort.

“In the promising space of mobile health, too often data is controlled by private interests,” said study coauthor Brian Bot, Principal Scientist, Sage Bionetworks. “Shared data resources such as these will help enable the scientific community to more quickly determine what can and cannot be gleaned from these types of remote measurements.”

Additional information:

Contact: Brian Bot

About Sage Bionetworks Sage Bionetworks is a 501(c) (3) nonprofit biomedical research organization, founded in 2009, with a vision to promote innovations in personalized medicine by enabling a community-based approach to scientific inquiries and discoveries. Sage Bionetworks strives to activate patients and to incentivize scientists, funders and researchers to work in fundamentally new ways in order to shape research, accelerate access to knowledge and transform human health. It is located on the campus of the Fred Hutchinson Cancer Research Center in Seattle, Washington and is supported through a portfolio of philanthropic donations, competitive research grants, and commercial partnerships. More information is available at

Sage Bionetworks Advocates for Open Systems in Health Research

Nature Comment: Stop the privatization of health data

SEATTLE WA (July 20, 2016)

Sage Bionetworks, a nonprofit biomedical research organization, continues its work to redefine the way in which health data is gathered, shared and used through the use of open systems, incentives and norms. In a Nature commentary published today, a set of governing principles for digital health data analysis that are designed to maximize the contribution of large-scale digital data to advancing medical care are described.  This commentary was co-authored by John Wilbanks, Chief Commons Officer at Sage Bionetworks and Eric Topol, MD, Director of the Scripps Translational Science Institute, and Chief Academic Officer of Scripps Health.   The two work together on the NIH-funded Precision Medicine Initiative that was announced earlier this month.

“Emerging technologies for sensor-based monitoring of health, including wearable trackers and smartphone apps, provide an unprecedented resource for the collection of longitudinal data from individuals,” said Lara Mangravite, PhD, President of Sage Bionetworks. “These data may enable the tailoring of disease modulators to maximize benefit on an individual basis.” The opportunities arising from the availability of large-scale digital health data have caught the attention of many technology companies including those with extensive expertise in big data management and analysis across other domains. However, these companies tend to operate in closed manners that provide little transparency into methods and little opportunity to test reliability of outcomes.  “Closed systems disable the checks and balances necessary to ensure that observations are accurate and reliable,“ said Dr. Mangravite. “This is essential when building algorithms designed to influence decisions about health.”

Harnessing these data will require large-scale efforts for data management and interoperability  – a problem that has been solved across many in other fields but remains an issue within the health space. “Many data providers offer a mechanism to return data to consumers, but not in a way that they can easily combine it with data collected across providers or shared with others,” said Mr. Wilbanks. “The digital health revolution only works if we place the patient in control of how and where their data is shared.”  Sage Bionetworks has conducted a series of clinical research studies in which more than 75% of those enrolling choose to share their data broadly, in contrast to “closed loop” systems where the data disappear into a silo for private analysis. “Over the long term, open systems are necessary to maximize the benefit derived from health data,” said Dr. Mangravite. “They promote multiple avenues of scientific progress in parallel and force a degree of scientific evaluation that can not be otherwise performed.”

A case study in Open Systems: mPower

In March 2015, Sage Bionetworks launched mPower, a smartphone application-based study to pilot the feasibility of remotely collecting frequent measurements of symptom severity and treatment sensitivity in Parkinson’s disease.  The mPower app collects data on capacities affected by Parkinson’s disease, including dexterity, balance and gait, memory, and certain vocal characteristics, through tasks that make use of iPhone sensors.  mPower launched through Apple’s ResearchKit, which helps doctors and scientists quickly gather data for medical research on an ongoing basis using iPhone apps. Unlike traditional studies, mPower participants are able to choose who to share their data with. Sharing options include only those researchers associated with mPower, or qualified researchers worldwide. So far, over 75 percent of the more than 12,000 mPower participants chose to share their data broadly with researchers.

In February 2016, Sage Bionetworks released the first six months of data donated by those mPower participants that chose to share broadly. Researchers can become qualified and access the Parkinson’s database via Synapse, a data and analysis sharing platform. From a preliminary analysis of the data, Sage has found enormous variation of symptoms within individuals, which could help researchers better pinpoint windows of intervention. Sage has also noted distinct patterns between medication intake and symptoms, insights that could eventually inform care and treatment regimes. By opening up the data and making it available to all researchers, Sage hopes to accelerate new insights that can be implemented in the clinic and lead to improvements in the lives of people with Parkinson’s disease – insights that can only be realized with a large, open community of data scientists analyzing and re-analyzing the data.

About Sage Bionetworks

Sage Bionetworks is a nonprofit biomedical research organization, founded in 2009, with a vision to promote innovations in personalized medicine by enabling a community-based approach to scientific inquiries and discoveries. Sage Bionetworks strives to activate patients and to incentivize scientists, funders and researchers to work in fundamentally new ways in order to shape research, accelerate access to knowledge and transform human health. It is located on the campus of the Fred Hutchinson Cancer Research Center in Seattle, Washington and is supported through a portfolio of philanthropic donations, competitive research grants, and commercial partnerships. More information is available at

About mPower

mPower (Mobile Parkinson’s Observatory for Worldwide, Evidence-based Research) is an iPhone app-based study notable as one of the first observational assessments of human health to rapidly achieve scale as a result of its design and execution purely through a smartphone interface. Learn more by visiting the mPower Website or download the mPower app on iTunes.

For more information on the Precision Medicine Initiative (PMI), see

Contact: Diane Gary

“Open Science Champion of Change” Sage Bionetworks’ Stephen Friend: Honored at White House Event

SEATTLE–(BUSINESS WIRE)–The White House hosted an Open Science Champions of Change event on June 20 and honored Dr. Stephen Friend of Sage Bionetworks ( for Sage’s vanguard work building open biomedical data systems and organizing Big Data Health projects that spur citizens and researchers to share and develop ideas in real time, leading to spectacular examples of how to harness the power of many minds to accelerate scientific progress.

United States Chief Technology Officer, Todd Park, presided over the event and applauded Friend’s contributions to Open Science as well as the two Big Data Challenge projects that Friend announced to crowdsource important questions for arthritis and Alzheimer’s disease research. “These Challenges that Friend and Sage Bionetworks announced here today are truly awesome examples of what we at the White House want to highlight. It’s these types of open efforts that will help grow our economy and improve our world.”

“At the core of Sage Bionetworks is the idea of building a precompetitive commons where citizens and researchers can come, interact, give and take basic research and build on one another’s insights,” said Dr. Friend. “It’s a novel combination of two existing concepts. The first is the Commons, a community where information is shared in order to achieve greater social benefits. The second is a precompetitive space – the idea that there is a basic research foundation that informs all effort to improve the health of people and families.”

An Information Commons to accelerate medical progress

Dr. Friend co-founded Sage Bionetworks in 2009 out of the belief that biomedical research will be more successful and affordable if conducted from a layer of data and models within an open “information commons” that individuals and teams can access to make better, faster and more relevant discoveries.

Synapse ( is Sage Bionetworks’ open compute platform and represents the type of open data system that can fuel a biomedical information commons where transformational new community efforts can run. Friend refers to Synapse as a “sandbox for geeks:” it is built to meet the needs of data scientists by providing an open, well-governed repository of analysis-ready data that scientific teams can work on in an open, online form that is accessible to all through a collaborative web portal equipped with engaging social media tools.

With the Synapse platform in place, Sage Bionetworks has initiated partnerships with innovative groups such as DREAM ( to apply the tools of open science to solve big problems in health.

What Participation in an Information Commons Might Look Like

In 2012, Sage Bionetworks demonstrated the types of democratized projects that an open compute platform like Synapse enables. Partnering with DREAM (, a visionary distributed systems biology group that has run 24 successful open computational challenges over the last five years, Sage Bionetworks ran the Breast Cancer Prognosis Challenge (BCC). The BCC invited citizens and scientists alike to try their hand at building a computational model that would accurately predict breast cancer survival: the Challenge ended up attracting 350 participants from more than 35 countries who submitted a total of 1,700 models over a three-month period of time. Remarked DREAM co-founder Gustavo Stolovitzky, “By running the BCC off of Synapse, we were able to engage our Challenge players in exciting new ways that increased the level of participation. Synapse made it possible to share breast cancer patient data with anyone who wanted to compete and to run a real-time leaderboard where participants could track their own progress and access the model code of others to forge new models.”

Friend and Stolovitzky recognized that the most powerful incentive to gather a crowd of data scientists might be scientific publication instead of cash, and they worked with the editors of the journal Science Translational Medicine to position an invited publication on the winning model as the Challenge’s top prize.

Based on the success of the BCC, Sage Bionetworks and DREAM merged earlier this year to run open science computational Challenges that foster the broader collaboration of the research community and that provide a meaningful impact to both discovery and clinical research. DREAM’s 8th season of Challenges started earlier this month with the launch of three DREAM8 Challenges that are open for participation through September 15, 2013 (!Challenges:DREAM8). Quips Stolovitzky, “These are three great Challenges tackling important questions in breast cancer, environmental toxicology and basic biology. We really hope that Challenge participants will use Synapse’s leaderboards, code-sharing and provenance tools to start a real-time dialog that fosters rapid learning and better predictive models: Sage Bionetworks and DREAM will help get the ball rolling by sprinkling in a few fun rewards.”

Sage Bionetworks Announces Grand Challenges In Rheumatoid Arthritis and Alzheimer’s Disease

Dr. Friend also announced two further open science Challenges that Sage Bionetworks and DREAM have planned to improve scientists’ understanding of two debilitating diseases affecting our society: rheumatoid arthritis and Alzheimer’s disease. Accompanied by key leaders from the arthritis and Alzheimer’s communities, Friend referred to these as Grand Challenges “…that are already engaging diverse groups eager to leverage DREAM Challenges to advance biomedical research and solve problems that matter most for patients. We fully expect that these Challenges will generate winning models that then inform the design of new clinical trials or that spell out the patient data we most need to help improve treatment options for arthritis and Alzheimer’s disease.”

The Rheumatoid Arthritis Responder Challenge

“This Challenge represents a unique interaction among an international team of researchers from academics and industry who share a common goal: improve the lives of patients with a debilitating form of arthritis,” said Director of Genetics & Genomics at Brigham & Women’s Hospital Dr. Robert Plenge, who is also a leading organizer of the Rheumatoid Arthritis Responder Challenge. “For people with RA, strong immunosuppressive medications are administered in order to treat pain and inflammation; unfortunately the medication only works for 30% of patients and we don’t know why. We want to run this Challenge to make complex genetic and genomic patient data available and ask the crowd of Challenge participants to generate predictive models that will help doctors know which patients are most likely to respond to treatment.”

Indeed, the Rheumatoid Arthritis Responder Challenge has already received scientific support and funding from the Arthritis Foundation, the Consortium of Rheumatology Researchers of North America, Inc. (CORRONA), Merck, Novo Nordisk and likely additional pharmaceutical companies.

Accompanied by leaders from arthritis research and the Arthritis Foundation, Friend described the Rheumatoid Arthritis Responder Challenge as starting later this year and announced that the Challenge’s “winner” will get to see their winning classifier potentially drive a prospective clinical trial with an already existing registry of RA patients (the Arthritis Internet Registry).

“Innovative research collaborations like this are crucial to help answer key biologic questions as we work toward a goal of personalized medicine for rheumatoid arthritis,” said Arthritis Foundation Vice President of Research, John Vernachio, Ph.D. “We are excited about gaining genetic insight from the Rheumatoid Arthritis Responder Challenge and applying the findings to citizen scientists enrolled in the Arthritis Internet Registry.”

The Alzheimer’s Disease Big Data Challenge

With 5 million Americans and almost 40 million people worldwide currently afflicted with Alzheimer’s disease, the Global CEO Initiative on Alzheimer’s Disease ( is joining with Sage Bionetworks/DREAM to co-lead this Challenge. The first partner for the Challenge is the national Alzheimer’s Association. The goal of this Challenge will be to respond to science and industry demands for accurate predictors of Alzheimer’s risk in pre-clinical populations. Dr. Robert C. Green of Brigham and Women’s Hospital and Harvard Medical School is co-chairing the scientific advisory board for this effort and remarked: “We hope that this first Big Data Challenge on Alzheimer’s disease will be part of a global approach to identify faster and more accurate predictive models to guide science and industry toward more effective treatment strategies.”

For the first Alzheimer’s disease Big Data Challenge, Sage Bionetworks/DREAM will utilize data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI,, already a leader in sharing data. The ADNI dataset consists of cognitive, imaging, biochemical, and whole genome sequencing data on cohorts of volunteers, who are cognitively normal, have mild cognitive impairment and have Alzheimer’s disease. The winning predictive model from the Challenge will be determined by evaluation against a similarly structured validation data set that is either newly generated for the Challenge or not yet released. The winning team will be featured as lead authors in a Challenge article in a prominent journal, to be announced.

Stated George Vradenburg, Convener of The Global CEO Initiative on Alzheimer’s Disease, “It’s time to disrupt ‘business-as-usual’ with innovative ‘big data’ techniques. And there is no more pressing a challenge on which to try this innovation than finding the best mix of predictors for Alzheimer’s. We are co-leading this Challenge to bring innovative approaches to the bedeviling challenges of Alzheimer’s.”

Excited by the impact of an Alzheimer’s Challenge, Friend reflected, “A DREAM Challenge for Alzheimer’s disease based on this ADNI cohort provides a unique opportunity to promote open science in the study of this terrible disease, and an opportunity for an even larger community to collectively develop an understanding in the integration of these complex data types. By engaging a diverse community of Challenge participants and letting them learn how to build predictive models driven by a variety of complex data, we set the stage for running Alzheimer’s Disease Challenges #2, #3 and so on.”

Dr. Michael Weiner, ADNI’s Principal Investigator, confirmed, “This innovative proposal by Stephen Friend at Sage Bionetworks to launch a Challenge using ADNI and other data from the Alzheimer’s field is expected to lead to new discoveries that will accelerate diagnosis and effective treatments and preventions of this terrible disorder which affects millions.”

Added Maria Carrillo, vice-president for medical and scientific relations of the national Alzheimer’s Association: “As a global leader in Alzheimer’s research, the Alzheimer’s Association applauds Sage Bionetworks’ recognition of the critical need for Alzheimer’s research and investigation of big data. We look forward to promoting the availability of this project to our broad and diverse international research community.”

Friend envisions a world where open science is flourishing and helping to afford individualized maps of disease and wellness to an engaged citizenry. In keeping with this vision, John Wilbanks, Chief Commons Officer at Sage Bionetworks and leading the organization’s groundbreaking work with online, patient-centered informed consent, was also chosen by the White House as one of only twelve open science leaders to present a scientific poster in affiliation with today’s Open Science Champions of Change event. Certainly Friend’s efforts at Sage Bionetworks, with DREAM and the early adopters of open science in arthritis, Alzheimer’s disease and other disease communities are leading us all in that direction and reflect what it means to be an Open Science Champion of Change.

Press Release on BusinessWire