Systems, devices, and methods for assessment of brain injury
Inventors
Fanton, Michael G. • Camarillo, David B. • Laksari, Kaveh • Chun Wu, Lyndia • Kurt, Mehmet • Nguyen, Taylor H.
Assignees
Leland Stanford Junior University
Publication Number
US-12303256-B2
Publication Date
2025-05-20
Expiration Date
2039-12-12
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Systems, methods and devices for detecting a concussive event are provided. A computational classifier may be trained and utilized for detecting a concussive event in real-time. Head kinematics can be measured and a head kinematic metric determined, which can be utilized within the classifier to detect a concussive event.
Core Innovation
The invention provides systems, devices, and methods for detecting concussive events in real time through direct assessment of head kinematics. The core approach utilizes a head-mounted device, such as a helmet, mouthguard, or similar wearable equipped with one or more accelerometers, to continuously measure angular head motion and gather kinematic data during head impacts. This data is then processed in real time using a mathematical mass-spring-damper brain model to compute a brain angle metric—most notably, the Brain Angle Metric (BAM) that quantifies rotational deformation of the brain across anatomical directions.
A computational classifier, such as a trained regression model, is then employed to determine whether the measured brain angle metric corresponds to a concussive (i.e., mild traumatic brain injury) or non-concussive event. The classifier is trained on empirical data, distinguishing injury from non-injury using head kinematic metrics including maximum brain angle values. By integrating these processes within a system, the invention enables timely, automated, and quantitative detection of potentially injurious head impacts.
The addressed problem is the difficulty of timely diagnosing and preventing mild traumatic brain injury due to limited understanding of injury mechanisms and the inadequacy of traditional translation-based head injury metrics to assess all types of brain injury, particularly those caused by rotational acceleration. The invention incorporates real-time assessment, enhancing the ability to make informed decisions about medical intervention by providing immediate feedback through the integration of hardware and analytical models.
Claims Coverage
There are two independent claims, each focusing on the core inventive features of systems and real-time methods for assessment and detection of concussive events using a head-mounted device, a mass-spring-damper brain model, and a trained classifier.
System for real-time detection of concussive events using head-mounted device, brain model, and trained classifier
A system includes: - A head-mounted device worn or secured to the head of an individual, comprising one or more accelerometers for measuring angular head motion. - A computer system communicating with the head-mounted device, including a memory and processor that: - Captures in real time brain angle measurements in one or more anatomical directions using the accelerometers. - Computes in real time a brain angle metric for the anatomical directions using a mathematical mass-spring-damper model. - Determines in real time, using the brain angle metric in a trained classifier, whether a concussive event has occurred, with the classifier trained to differentiate between concussive and non-concussive events.
Real-time method for detection of concussive events using brain angle metric and trained classifier
A real-time method comprises: 1. Fitting a head-mounted device with one or more accelerometers to an individual’s head to measure angular head motion. 2. Capturing in real time brain angle measurements in one or more anatomical directions via the accelerometers. 3. Computing in real time, using the captured measurements and a mathematical mass-spring-damper model within a computer system, a brain angle metric for those anatomical directions. 4. Determining in real time, with the computed brain angle metric as input to a trained classifier on the computer system, whether a concussive event has occurred, with the classifier trained for concussive/non-concussive classification.
The independent claims cover both a system and a real-time method that uniquely integrate wearable head-mounted sensors, real-time computation of a brain angle metric via a mass-spring-damper brain model, and classification using a trained classifier to detect concussive events.
Stated Advantages
Enables real-time detection of concussive events using head kinematics and computational models, allowing for immediate feedback and intervention.
Provides a quantitative and automated assessment of brain injury risk, reducing reliance on subjective evaluation and overcoming limitations of traditional translational metrics.
Allows implementation across various head-mounted devices, such as helmets or mouthguards, making the system adaptable to different users and environments.
Documented Applications
Assessment and detection of concussive events in athletes, fighters, military personnel, and epileptics using wearable head-mounted devices.
Initiation of medical interventions such as field tests, neurological assessment, cognitive testing, medical imaging, observation, physical and mental rest, and pain or anti-inflammatory medicine following detection of a concussive event.
Utilization in field settings, including sporting events and combat environments, to provide real-time monitoring and injury risk classification based on head impacts.
Interested in licensing this patent?