Opt-In Signup

Enables privacy controls for the user, with a minimum signup process. Meanwhile, both 'opt-in' and 'opt-out' options are supported.

Facebook (Default)

Realtime monitoring of Facebook profile content. Specifically, posts and status updates. (Other information remains private, without special permissions.)

Profile Control

Users accept terms via an easy to read Consent Form. Users can also change their participation status.

Additional Networks

Users can add their Twitter and Google+ (coming soon) profiles to observed content.

Twitter Posts

Tweets are cross-referenced with similarly timestamped Facebook content to build a profile.

Networks Status

Users are informed of the social networks that are enabled.

Medical Database

Residing behind the medical firewall, the participant data is stored for analysis by medical professionals according to Human Subject Study/HIPAA privacy rules.

CPHS #23781

Clinician Dashboard

Here clinicians log in to a HIPPA compliant online system to view patient risk.

Patient View

Patients are presented with an overall risk rating, with statistics, and clinical notes.

Disclaimer: Poulin and Thompson are the study investigators.

Realtime Patient Monitoring

Here the patient is flagged for a variety of risks from Green (Low), to Yellow (Nominal), to Red (High) risk. This profile is monitored at 1 minute intervals.

Suicide Ideation Rating

Here we examine the patient's risk rating in detail. Including: his/her overall numeric rating in the system, the probability (or confidence) in this rating, keywords that are specific to his/her profile, and the specific date that the risk assessment was made.

Time Series/Trends

We view the collection of predictions for the patient, to look for trends. Much like a physician of internal medicine looks for a consistent heartbeat, we provide a 'mental health ticker' to the mental health professional.

By Cohort

Here we visualize the patient's risk in relation to the clinician's other patients.

Clinician Notes

Clinicians add notes, specific to the patient. The combined analysis of the patient's self-reported text, as well as the clinician's observations provides a broader picture of text driven risk.

Source Content

If a patient’s risk suddenly changes, or the clinician wants to drill down on specific phrases, the clinician can view the original source content from the various networks.