Social Media & AI Fight Cancer

Last year Brandon Welch, PhD, a biomedical informatics researcher with the Medical University of South Carolina’s (MUSC) www.musc.edu Department of Public Health Sciences, received a grant for $867,000 from the National Cancer Institute www.cancer.gov to continue working on his project involving social networking and artificial intelligence to help patients fight cancer.

His goal is to create a simple tool that will allow patients to crowdsource their family health histories, since he hopes this information will give doctors insights into the risks they face when it comes to heritable diseases like cancer or heart disease. As Welch explains, “Although genetic information is very valuable, doctors can help patients identify risks and get the treatments they need, often before they experience any symptoms.”

However, according to Welch, “Very often, few doctors have the time to collect detailed family histories, so as a result, doctors only collect family histories from about half of their new patients and from less than a quarter of their returning patients.”

CMS reports that patients often report inaccurate family histories and in fact less than four percent of family histories are detailed enough to accurately assess a patient’s risk for heritable diseases.

So Welch created http://ItRunsinMyFamily.com which is now in beta testing. What sets the site apart is its user interface. Rather than filling out a form, patients have a conversation with “dokbot”, a sophisticated computer program that walks users through the family history collection process much like a doctor or genetic counselor would.

Welch hopes the system will be able to provide personalized recommendations for users with a strong family history of cancer to mitigate their risks. The intent is to also integrate family history information with EHRs and create interfaces for clinicians to more easily gather data for their own research. In addition, the plan is to not only collect family health information but also be able to combine the information with large scale genomic sequencing data.

Welch and his team have started recruiting beta users in February and right now are working to find and fix any bugs in the code. Right now, the system deals only with cancer history, but the research team expects to expand it to include heart and vascular disease, diabetes, and others in the future. Welch hopes to add the first patients by March 2017.

If interested in taking part and would like to be a beta user, email Brandon Welch at welchbm@musc.edu.

Share Button