Background

Brain Injuries

Every 28 seconds, an American suffers a catastrophic brain injury, most commonly stroke or traumatic brain injury (TBI). Irreversible brain damage can begin within minutes of injury. Because this critical window of time is so small, there is a need for immediate evaluation and diagnosis of suspected brain injuries.

Computed Tomography (CT) Scanning

CT scans, which take a series of X-ray images to recreate the image of bones, blood vessels, organs, and other tissues inside the body, are frequently used to identify internal injuries and signs of disease. Currently, the scans are subject to radiologist interpretation and "grading," both of which are subjective and time consuming.

Artificial Intelligence (AI)

Artificial Intelligence employs computer and machine-based technology simulations of intelligent human behavior to accomplish tasks, such as visual perception, decision-making, translation, and image recognition. To create AI technology, engineers expose a program to many versions of training simulations so the program "learns" to recognize input and perform tasks; the program's accuracy increases as it receives more data.

Project

Advances in image recognition, such as those employed by Google and Facebook to recognize faces in images have yet to be applied to medical imaging. Automating recognition of injury in brain images could improve rapid detection of emergencies as well as address both physicians’ time constraints and subjectivity common in these diagnoses. The team successfully trained their AI with over 100,000 CT scans. The AI detected intracranial bleeding with greater than 99% accuracy. This rate is equivalent to the performance of board-certified radiologists.

Additionally, the team integrated this technology into a cloud-based platform, which hospitals can use with several different CT scanning devices and protocols. This innovation allows medical professionals to use this AI anywhere in the world, and update a catalog of potential clinical biomarkers of neurologic injuries. This tool holds the potential to expedite treatment of irreversible damage and reduce subsequent long-term disability and death.

This project also recognizes the need for clinically effective, cloud-based, automated image analysis in under-served areas of the United States and developing world, since TBI and stroke disproportionately affect the populations living in these areas. Providing CT scanning equipment alone can’t meet the needs of these populations, which often lack enough qualified radiologists to interpret patient CT scans.

Once approved for commercial use, the research team plans to make this technology available to meet the needs of Californians as well as patients around the world without rapid access to qualified diagnostic professionals.

To further test and apply their technology to integrative patient care, the team is working with a commercial partner to receive FDA approval for this device as a diagnostic test. Once approved, they will implement this technology with a focus on measurement of patient outcomes and application as the first quantitative biomarker of brain injury in clinical trials.

Research Team and Collaborators

Research Team

  • University of California, San Francisco
    • Arash Afshinnik, MD
    • Claude Hemphill, MD
    • Nerissa Ko, MD
    • Geoff Manley, MD, PhD
    • Pratik Mukherjee, MD, PhD
    • Esther Yuh, MD, PhD
  • University of California, Berkeley
    • Jitendra Malik, PhD

Collaborators

  • Brain Trauma Foundation
  • Community Regional Medical Center in Fresno
  • Stanford University
  • TBI Endpoints Development Project
  • Transforming Research and Clinical Knowledge in Traumatic Brain Injury Consortium
  • UC Berkeley
  • Zuckerberg San Francisco General Hospital and Trauma Center