Press "Enter" to skip to content

JOB: RehaBot project seeks a developer

I’m helping a friend with research related to the development of RehaBot—a conversational agent for promoting rewarding activities in users with traumatic brain injury. The team is looking to hire a developer to join the project. Tech includes Microsoft Azure and Python/Django.
If you are interested (or have a recommendation/referral), could you please contact me at daniel.shiffman@nyu.edu?

RehaBot Project

Making, remembering, and initiating plans can be very challenging for people living with chronic traumatic brain injury (TBI). RehaBot is a chatbot designed to help people with TBI engage in goal-directed behavior. RehaBot was initially designed as part of a treatment program for depression based on Behavioral Activation– the idea that doing enjoyable and values-based activities increases an individual’s exposure to environmental reward and counteracts feelings of depression. A working prototype of RehaBot was tested in a small pilot study (Rabinowitz et al., 2022). My lab recently received funding to initiate a clinical trial testing RehaBot as a mobile Health tool to promote exercise and physical activity in people with chronic TBI, scheduled to begin recruiting participants by the end of this year. 

 

RehaBot is designed to augment one-on-one treatment with a therapist. The therapist and client work collaboratively to come up with personalized goals, plans, and schedules that the therapist inputs into a web-based clinician portal. The client then can communicate with RehaBot over SMS to get reminders of their planned activities, track progress, and receive reinforcing messages for completing planned activities. Users’ interactions with RehaBot are based on personalized information about their goals, plans, schedules, etc. as entered into the clinician portal. 

 

Before initiating the clinical trial we have the following development needs:

 

  • Fix identified bugs in natural language interactions
  • Add RehaBot initiated conversation (right now all interaction must be initiated by the user)
    • Scheduled activity reminders 
    • Prompts to report on their enjoyment/satisfaction upon completing an activity
    • Prompts to report on their satisfaction with progress towards physical activity goals
  • Add “coping plan” reminders– personalized messages to encourage users to carry out activities when they don’t feel like it (available on demand and/or when they decline to do a scheduled activity)

 

Technical Specifications

 

  • Django Web Application, Python
  • Microsoft Azure Language Understanding (LUIS) and HealthBot framework 
  • Twilio SMS messaging

 

Rabinowitz, A. R., Collier, G., Vaccaro, M., & Wingfield, R. (2022). Development of RehaBot—a conversational agent for promoting rewarding activities in users with traumatic brain injury. The Journal of Head Trauma Rehabilitation, 37(3), 144-151.