Discovering AI Applications for Traumatic Brain Injury Care (2024-2025)


Each year, five million Americans seek emergency medical care for traumatic brain injury (TBI), a major cause of death and disability overall and a leading cause in young adults. The harmful effects of TBI can manifest over the long-term as mood/attention disorders, cognitive impairment and suicidality. The best course of treatment is often difficult to determine due to the complexity of TBI presentations and the nonlinear relationship between patient data and phenotypes. One survey found that only 37% of healthcare providers believed they could make an accurate prognosis and 67% believed a better predictive model would change how they handle patients. Therefore, developing better TBI approaches will have a large impact on the health of Americans. 

Recent advances in computing and machine learning have made it possible to process and learn from large-scale healthcare datasets. For example, a clinical sepsis model has already made an impact at Duke Health. However, appropriate management and care of TBI presents complex challenges that other treatment processes do not. Therefore, to successfully employ a new model, it is critical to cultivate early engagement with stakeholders and leverage their input for a detailed assessment of the care pathway and the data it generates.

Project Description

This project team will conduct a holistic survey on TBI management and care at Duke that will inform the development of AI solutions and pave the way for future deployment. To achieve this goal, team members will use qualitative methods such as interviews and focus groups to engage with a broad range of healthcare team members involved in the TBI clinical pathway, including radiologists, radiology technicians, ICU and ER nurses, neurosurgeons, EMTs, neurologists, ER staff and rehab specialists. Team members will be guided by experts in qualitative studies to create interview materials and advise on conducting successful studies. Additionally, students will have the opportunity to shadow healthcare professionals and learn about their roles and responsibilities. 

The project team will also develop AI modeling solutions for application to different points of the TBI pathway. Team members will work with experts to develop prototypes of AI solutions that can be used in the clinical workflow. Students will learn approaches to model building and deployment in the clinical setting, as well as ensuring ethical and responsible AI methodologies. The team will work to combine the analysis of a large existing dataset of TBI-related encounters in the Duke Health system with insights extracted from the qualitative study team. 

Anticipated Outputs

Publications; prototypes of AI solutions; summary report on the state of TBI care at Duke Health; interactive website showcasing findings and student-curated blog posts

Student Opportunities

Ideally, this project team will include 4 graduate students and 12 undergraduate students from an array of backgrounds, including engineering, statistics, computer science, neuroscience, economics, policy and/or health and medicine. 

The team will break into two subteams, one focusing on qualitative data collection and the other focusing on quantitative research. Members of the qualitative subteam will build skills in literature review, healthcare comprehension and effective scientific communication. Team members will learn qualitative research techniques, conducting interviews and shadowing to bolster data interpretation and communication abilities. 

Those on the quantitative research subteam will handle intricate healthcare data, building fair models and collaborating with stakeholders, refining their data analysis proficiencies. Aspiring medical professionals will gain comprehensive healthcare system knowledge, bridging theoretical understanding with practical insights. 

In Fall 2024, the full team will meet on Mondays from 2-3:30 p.m. and subteams will meet on Wednesdays from 2-3 p.m.

All students will refine skills in collaborative writing and presenting. Regular updates and blog posts will sharpen their ability to articulate scientific concepts cogently, catering to diverse audiences. 


Fall 2024 – Spring 2025

  • Fall 2024: Begin team meetings; conduct literature surveys; attend expert lectures; begin clinical shadowing; design interview material
  • Spring 2025: Continue interviews and focus groups; transcribe data and extract insights; start developing prototype AI models; submit results for publication


Academic credit available for fall and spring semesters


Image: Gene Activity After TBI, by Douglas Arneson and Drs. Gomez-Pinilla and Yang, UCLA, NIH Image Gallery, licensed under CC BY-NC 2.0  

Image: Gene Activity After TBI, by Douglas Arneson and Drs. Gomez-Pinilla and Yang, UCLA, NIH Image Gallery.

Team Leaders

  • Samuel Berchuck, Arts & Sciences-Statistical Science
  • Bradley Kolls, School of Medicine-Neurology
  • Brian Lerner, Pratt-Electrical and Computer Engineering-Ph.D. Student
  • Pranav Manjunath, Pratt-Biomedical Engineering-Ph.D. Student

/yfaculty/staff Team Members

  • Michael Cary, School of Nursing
  • Timothy Dunn, Pratt School of Engineering-Biomedical Engineering
  • Deborah Koltai, School of Medicine-Psychiatry and Behavioral Sciences;Neurology
  • Tolulope Oyesanya, School of Nursing
  • Karin Reuter-Rice, School of Nursing