The National Institutes of Health, under the National Library of Medicine's program on data science research, has awarded $1.55 million to an interdisciplinary team lead by Luis Rocha, a professor of informatics and the director of the NSF-NRT complex networks and systems program at the Indiana University School of Informatics, Computing and Engineering.
The four-year project, a collaboration between the IU School of Informatics, Computing and Engineering and the IU School of Nursing, will employ innovative data- and network-science methods to produce myAURA, an easy-to-use web service for epilepsy patients.
This service will integrate practical, location- and patient-specific health-care information with electronic health records, targeted scientific literature, biomedical databases, social media, and epilepsy-related websites with information about specialists, clinical trials, drugs, community resources and more. The project builds on previous research conducted by Rocha's team and funded by Persistent Systems, which extracted big data about chronic conditions and drug interactions from Twitter and Instagram.
myAURA will be based on a large-scale epilepsy knowledge graph built by integrating data from social media, electronic health records, patient discussion boards, scientific literature databases, advocacy websites and mobile app data. The knowledge graph will, in turn, be used to fuel recommendation and visualization algorithms based on the automatic inference of relevant associations.
The inference will follow algorithms developed by Rocha's team to remove redundancy and extract factual information from large knowledge graphs as well as parsimonious network visualizations developed by Katy Börner, an IU Distinguished Professor in the IU School of Informatics, Computing and Engineering.
The project will also focus on user-centered development and pilot testing of myAURA to validate if and how it improves patient activation through a series of studies lead by Wendy Miller, an assistant professor at the IU School of Nursing.
"Our previous work on depression showed that it's feasible to look at social media for health-related issues," said Rocha. "The collaboration with Dr. Miller, who is an expert in epilepsy management, will allow us to translate our data and network science research, and develop personalized inferences useful to patients of this chronic disease."
Updated information on epilepsy treatment can be difficult for patients to obtain, and the team hopes to create a novel resource for patients.
"One of the problems in the United States is that most small towns--and even not-so-small towns--don't have epileptologists," said Rocha. "At best, patients might see a neurologist, who is not specialized in epilepsy, but most often, people just see a general practitioner. It often happens that they get prescribed older medications not ideally targeted to the patient's case, and follow-up often does not work very well.
"We think data science can help this situation. We came up with the idea to integrate data from social media, from electronic health records, from the published literature, from doctor databases, and from clinical trials into the large-scale knowledge graphs that enable myAURA, which is a personalized portal where people with epilepsy can manage their disease and find personalized information."
Rocha and his team will enjoy the cooperation of the Epilepsy Foundation of America, which manages epilepsy.com, the premiere site for sufferers of epilepsy.
Rocha hopes to expand the techniques used to create myAURA to develop web-based applications for other chronic diseases.
"Projects such as these are the perfect example of the ways big data, data science, network science and informatics can come together to solve important real-world issues," said Raj Acharya, dean of the IU School of Informatics and Computing. "This interdisciplinary work showcases the potential SICE has to impact lives."