IMagine a disease that affects 4% of the population, kills 20% if untreated, has an effective and cheap treatment, yet kills 500,000 by suicide each year: you have imagined bipolar disorder (BPD), the single most costly psychiatric disease.
BACKGROUND ON BIPOLAR DISORDER
Notoriously difficult to diagnose, BPD has an average diagnosis delay of over a decade while the majority of suicides occur in the first five years of symptoms.
Treatment is cheap and effective, but BPD is often unrecognized and mis-diagnosed as Unipolar depression.
Early diagnosis and intervention directly impacts the treatment response and long-term prognosis.
The major symptoms of BPD, mania and depression, have distinct and persistent physiological biomarkers, such as physical activity and sleep duration/quality.
Using wearable devices, the Stanford Bipolar Clinic is developing physical activity and sleep biomarkers for mania and depression to dramatically cut the diagnosis delay, from a decade to a few months.
Students working on this project will have the opportunity to design and implement a full stack psychiatric research system. Work will include designing surveys and psychometric tests to run directly on smartwatches, building an information pipeline from the smartwatch to the server, and creating an elegant and reliable web interface for treating psychiatrists and researchers. Completed work will be applied immediately in an ongoing clinical trial at the Stanford Bipolar Clinic.
We are looking for students with experience in at least one of the following areas:
Web development: UI /UX
Server-side architecture; 3rd party web APIs
Database management; SQL and NoSQL
Wearable device programming
Currently, the project uses these languages and frameworks:
Pebble Smartwatch Platform
Node.js and Express on AWS Elastic Beanstalk
R and Matlab