20 minutes into her conversation with a patient with a diagnosis of irritable bowel syndrome (IBS), Dr. Zurcher realizes that she and her patient aren’t at all on the same page. With her own concept of “IBS” in mind, she approaches the encounter focused on uncovering and characterizing gastrointestinal symptoms. Her patient, on the other hand, is less interested in discussing his constipation or medication for IBS than he is in bringing to his doctor’s attention his crippling social anxiety, which disrupts his life much worse than any of his gastrointestinal complaints. Dr. Zurcher’s grasp of IBS as a diagnosis established according to the Rome III criteria, while medically sound, has little to do with her patient’s conceptualization of his illness, and unless she appreciates this, the encounter is unlikely to be productive.

To better understand how my patients conceptualize health and illness, I trained gensim’s word2vec implementation on 2 million disease-specific tweets. The beauty of this method is its capacity to uncover both obvious and less obvious semantic relationships among words. I challenge healthcare professionals to take this website for a spin and contrast their understanding of disease with their patients’ conceptualizations of illness. How the results should be interpreted isn't always obvious, but that's the fun of applying this method.

Try searching for “heart failure”, “obesity”, “alcohol”, "menstrual cramps", or “IBS”, for example. Each query returns the 10 semantically and/or lexically nearest neighbors in 100-dimensional space, along with their cosine similarity to the query term. The closer to 1.0, the closer they are in hyperspace.

Note: None of the results returned represent my own views of the world. Any obscene or politically incorrect search results reflect the reality of online discourse in this network.

irritable bowel0.655028522015
social anxiety0.654120802879
menstrual cramps0.614413142204
acid reflux0.603365182877

Please cite this website, Gensim's word2vec implementation, as well as Mikolov et al's original word2vec paper, if you use these results in your own work:
Metwally O. Conceptualizing health and illness through word embeddings. https://smmc.pythonanywhere.com/wordvec. 4 March 2016.

About Me

I'm a software developer, health technologist, and an M.D. I enjoy building solutions to healthcare's many problems.

I serve as a health technology consultant and an advisor to startups. I also review Business Track applications to Stanford's Medicine X conference.

Previously, I founded a tech-enabled home healthcare company called PulseBeat (a Blueprint Health portfolio company). My team (Vidrio) won MIT's 2014 Hacking Medicine competition.

Send me a line at omar.metwally (at) gmail dot com. I'm happy to connect with like-minded people.

Find me on LinkedIn or Quora. Check out my blog.

"Let the beauty you love be what you do. There are hundreds of ways to kneel and kiss the ground."

--Jalal al-din Rumi