Maya Stemmer (Photo Credit: Pablo Arkoshin)
Like most of the rest of humanity, chronically ill patients have taken to social media to discuss or chronicle their illnesses. However, there is a latent potential for medical insights in their posts just waiting to be unlocked. Now, Ben-Gurion University of the Negev engineers have developed a method to locate chronically ill patients and analyze their social media posts to generate medical insights.
Maya Stemmer, supervised by Prof. Gilad Ravid and Prof. Yisrael Parmet, developed the method for her PhD. They chose Twitter and Inflammatory Bowel Disease (IBD), such as Crohn’s disease and ulcerative colitis, to test their system. This is the first time in history that such medical information is publicly available. Instead of a physical letter to a friend describing symptoms, treatments and daily life, there are millions of posts to be analyzed using Big Data skillsets to yield real world insights.
Through their machine learning methods, they were able to discover chronically ill patients. They tended to post more often about their diseases. Furthermore, the researchers identified distinctive characteristics of Crohn’s disease that helped differentiate it from ulcerative colitis. They were also able to confirm existing information about foods that increased or reduced inflammation. When they classified the sentiments expressed, they found that patients tended to talk about their disease negatively and their drugs and treatments positively.
Will you offer us a hand? Every gift, regardless of size, fuels our future.
Your critical contribution enables us to maintain our independence from shareholders or wealthy owners, allowing us to keep up reporting without bias. It means we can continue to make Jewish Business News available to everyone.
You can support us for as little as $1 via PayPal at [email protected].
Thank you.
The system could be modified for other platforms and other diseases, according to Stemmer. Twitter makes it easier by offering an API for academic researchers.
“The work presented in this dissertation shows that it is feasible to derive health related insights from patients’ self-reported posts. Patients are not the only users who talk about health on social media, and the presented framework helps to eliminate tendentious posts by interested parties. The methods can be adapted to other diseases and enhance medical research on chronic conditions. Using the framework to identify more patients and collect more data can shed light on patients’ coping strategies with their disease and its influence on their quality of life,” explains Stemmer.
Stemmer, Prof. Ravid, and Prof. Parmet are members of the Department of Industrial Engineering and Management in the Faculty of Engineering Sciences and the Guilford Glazer Faculty of Business and Management.
