Stroke is devastating when left untreated. Early treatment significantly improves outcomes, and can be reliably detected using the FAST exam, which specifically assesses facial asymmetry, arm weakness, and speech deficits for signs of stroke. Our project aims to use smartphone technology to build a stroke detection algorithm based on this exam. We have collected video data of hospitalized stroke patients performing aspects of the standard neurology exam, which includes the factors previously mentioned. The next step is to build an algorithm that can detect facial asymmetry using this data.

The video data collected from hospitalized stroke patients resembles the data that would be available if a stroke patient were to use a mobile application on their own. While this data has high fidelity to a real-world application, it is noisy and messy. This project will therefore consist of two parts:

  1. Data preparation and cleaning to isolate the facial features vs. background for each video frame
  2. Computer vision modeling to identify key facial landmarks that neurologists look for to identify facial asymmetry.

Students are encouraged to leverage existing work done on face landmark detection, but usage of third-party APIs is limited because data cannot be sent to third-party servers.

This is an UNPAID research project.

Faculty Advisor

  • Professor: James Noble
  • Center/Lab:
  • This project is run under the joint mentorship of Drs. Noble and Williams, both professors in the Department of Neurology. Their past research together has investigated other novel approaches for early detection and intervention of stroke, including successful efforts such as Hip-Hop Stroke.

Project Timeline

  • Earliest starting date: 9/7/21
  • End date: 5/2/22
  • Number of hours per week of research expected during Fall 2021: ~10
  • Number of hours per week of research expected during Summer 2021: ~10

Candidate requirements

  • Skill sets: Computer vision, machine learning
  • Student eligibility: freshman, sophomore, junior, senior, master’s
  • International students on F1 or J1 visa: eligible
  • Academic Credit Possible: Yes
  • Additional comments: Interest and willingness to learn new models to analyze the various types of data involved in our project