Skin Health Data Collected by App Offered for Research

Mole Mapper app screens (play.google.com)

14 February 2017. A smartphone app collecting images and data about moles on the skin of volunteers is now available to researchers for studies of cancer and other disorders. The Mole Mapper data set, collected by Sage Bionetworks in Seattle and Oregon Health and Science University in Portland, is described in a paper published today in the journal Scientific Data, and can be accessed by researchers from a portal at Synapse.org.

Mole Mapper was started by Dan Webster, a postdoctoral researcher at National Cancer Institute, part of National Institutes of Health in Bethesda, Maryland and the study’s lead author. The project, which continues to collect data, aims to show if crowd-sourced data from smartphones can reveal skin conditions for diagnostics and research. “In designing the study,” says Webster in a Sage Bionetworks statement, “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.”

Volunteers in the Mole Mapper project download an app written for both Apple and Android smartphones, which captures photos of moles growing on their skin. Participants provide the images and measurements of their moles over time that can indicate if the mole shows signs of becoming melanoma, an aggressive form of skin cancer. People with the app can also choose to share their non-identified data with medical researchers. One part of the app, was designed with Apple’s ResearchKit that provides modules for tracking activity, conducting surveys, and gaining user consent.

The current data set offers data from more than 2,000 participants providing 2,422 images, 3,274 mole measurements, and responses to 1,920 demographic surveys since October 2015. The data show the average mole size is nearly 4 millimeters, and individuals with red hair were more likely to be diagnosed with melanoma. Webster adds that, “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 accumulation of data from large numbers of participants offers opportunities for more intensive analysis by algorithms that process vast amounts of image and medical data with artificial intelligence making it possible for an automated diagnosis of melanoma. A paper published last month in the journal Nature shows the feasibility of this technique.

“They are close to having a computer with artificial intelligence that performs as well as board-certified dermatologists in its ability to discriminate melanomas from moles using digital images,” notes Sancy Leachman, professor and chair of dermatology at Oregon Health in a university statement. “If the technology lives up to its potential, users that need to be seen can get into their physician more quickly and those who have nothing of concern can avoid making an unnecessary trip to the doctor.” Leachman is one of the Scientific Data paper’s senior authors.

Sage Bionetworks says Mole Mapper is its second research project collecting data with smartphone apps designed with ResearchKit. The first initiative collects data on Parkinson’s disease, with the data set described in Scientific Data in March 2016.

Oregon Health tells more about the Mole Mapper app and study in the following video.

• Univ. Spin-Off Offering Pediatric Care Software
• Mobile App Designed to Study Rare Inflammatory Condition
• Wearable Health Devices Offer Early Disease Detection
• Many Health Apps Found Not Helping People in Need
• Heart Research Built Into Personalized Activity Tracker

*     *     *