Method and system for detecting concussion
Inventors
Samadani, Rosina • Sanderson, Daniel O. • Sanderson, Joel D.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12575782-B2
Publication Date
2026-03-17
Expiration Date
Abstract
A method for identifying an occurrence of a concussion in a subject involves capturing eye blink raw data from at least one eye of the subject. The method next involves analyzing the eye blink raw data to generate at least a first subject specific eye blink metric and determining that a significant difference exists between the first subject specific eye blink metric and a predefined normal value. Next, the occurrence of the concussion in the subject may be identified, based on determining that the significant difference exists.
Core Innovation
The invention identifies an occurrence of a concussion in a subject using eye blink raw data captured from two eyes in response to a common visual stimulus on a display. The subject is seated in front of the display at a distance of between 10 and 90 cm, and a visual stimulus is presented on the display. An eye tracking device having a camera at a frequency greater than 500 Hz obtains a sample over at least 30 seconds.
The invention analyzes, via a processor, a measure of how often blinks from the two eyes overlap over a period of time when subjected to the common visual stimulus. Based on the measure, the processor identifies the occurrence of a concussion in the subject. In certain implementations, the invention compares a subject-specific eye blink metric to a predefined normal value, including baseline, control, or expected values, to determine a significant difference.
The invention further includes additional eye-blink measures that support the concussion identification, such as counting blinks in a first eye that do not have a corresponding blink in a second eye. The invention may use statistical significance testing such that the concussion occurrence is identified when the difference is significant, including approaches involving a z-score and/or a p value. Output may include a concussion score, audio alert, visual alert, numerical data, or a certainty factor.
Claims Coverage
The partial content identifies two independent claims, each centered on identifying a concussion from eye-blink raw data captured from two eyes while a visual stimulus is presented. Across the independent claims, the core inventive features focus on overlapping blink frequency and counting non-corresponding blinks, used by a processor to identify concussion occurrence.
Concussion identification using overlapping blink measure during common visual stimulus
Seating the subject in front of a display at a distance of between 10 and 90 cm, presenting a visual stimulus on the display, capturing eye blink raw data from two eyes using an eye tracking device with a camera at frequency greater than 500 Hz over at least 30 seconds, analyzing a measure of how often blinks from the two eyes overlap over a period of time subjected to the common visual stimulus, and identifying the occurrence of a concussion based on the measure.
Concussion identification using count of non-corresponding blinks
Presenting a visual stimulus on a display, capturing eye blink raw data from two eyes in response to the visual stimulus using an eye tracking device with a camera at frequency greater than 500 Hz over at least 30 seconds, counting a number of blinks in a first eye that do not have a corresponding blink in a second eye, and identifying the occurrence of a concussion based on the counted number.
Overall claim coverage centers on concussion identification using processor-based analysis of eye blink raw data from two eyes during a shared visual stimulus, using either how often left/right blinks overlap or how many blinks are non-corresponding between eyes.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?