Field deployable concussion assessment device

Abstract

A device and method for assessment of traumatic brain injury (TBI) is described. The device is configured to acquire brain electrical signals from a patient's forehead using one or more neurological electrodes. The acquired brain electrical activity data is subjected to artifact rejection and feature extraction, and a subset of features are then combined in at least one classifier function. The classifier functions statistically place a patient in one of four categories related to the extent of brain dysfunction: 1) normal brain electrical activity; 2) abnormal brain electrical activity consistent with non-structural injury with less severe manifestations of functional brain injury; 3) abnormal brain electrical activity consistent with non-structural injury with more severe manifestations of functional brain injury; and 4) abnormal brain electrical activity consistent with structural brain injury.

The invention claimed is: 
     
       1. An apparatus for assessment of traumatic brain injury in a patient, comprising:
 a patient sensor comprising at least one neurological electrode; and 
 a handheld base unit operatively coupled to the patient sensor, the base unit comprising:
 a digital signal processor configured to perform automatic identification and removal of artifacts from brain electrical signals acquired by the at least one neurological electrode, extract one or more features from the acquired brain electrical signals, and execute at least one classification function to classify the patient into one of four a plurality of categories indicative of the presence and severity of traumatic brain injury; 
 wherein the at least one classification function is designed using a training database comprising a population of controls and patients who reportedly sustained closed head injuries and the training database is organized using Standard Assessment of Concussion (SAC) scores. 
 
 
     
     
       2. The apparatus of  claim 1 , wherein the at least one classification function comprises at least three classification functions. 
     
     
       3. The apparatus of  claim 2 , wherein the digital signal processor is configured to execute the at least three classification functions in cascade. 
     
     
       4. The apparatus of  claim 2 , wherein the digital signal processor is configured to execute the at least three classification functions in a sequence, the sequence comprising the steps of:
 executing a first classification function designed to classify patients with structural brain injury from patients who are normal or have only functional brain injury; 
 executing a second classification function designed to classify normal patients from patients having structural and/or functional brain injury; and 
 executing a third classification function designed to classify patients with two separate grades of functional brain injury. 
 
     
     
       5. The apparatus of  claim 2 , wherein the three classification functions comprise:
 a first classification function designed to classify patients with structural brain injury from patients who are normal or have only functional brain injury; 
 a second classification function designed to classify normal patients from patients having structural and/or functional brain injury; and 
 a third classification function designed to classify patients with severe and less severe manifestations of functional injury when no structural injury is present. 
 
     
     
       6. The apparatus of  claim 1 , wherein the at least one classification function comprises at least two classification functions. 
     
     
       7. The apparatus of  claim 6 , wherein the digital signal processor is configured to execute the at least two classification functions in parallel. 
     
     
       8. The apparatus of  claim 7 , wherein the digital signal processor is configured to execute the at least two classification functions independent of each other. 
     
     
       9. The apparatus of  claim 8 , wherein the base unit further comprises a user interface for displaying one or more classification performance measures to enable a clinician to make a decision about a category of the patient. 
     
     
       10. The apparatus of  claim 6 , further comprising a multi-stage classifier, wherein the at least two classification functions are performed by the multi-stage classifier. 
     
     
       11. The apparatus of  claim 1 , wherein the base unit further comprises a display unit for providing an indication of the presence and severity of traumatic brain injury. 
     
     
       12. The apparatus of  claim 11 , wherein the display unit displays the category that the patient is classified into. 
     
     
       13. The apparatus of  claim 1 , wherein the training database is organized using one or more clinical characteristics. 
     
     
       14. The apparatus of  claim 1 , wherein the one or more features comprise linear and/or non-linear quantitative features. 
     
     
       15. The apparatus of  claim 1 , wherein the four plurality of categories comprise:
 abnormal brain electrical activity consistent with structural brain injury; 
 abnormal brain electrical activity consistent with non-structural injury with severe clinical manifestations of functional injury; 
 abnormal brain electrical activity consistent with non-structural injury with less severe manifestations of functional injury; and 
 normal brain electrical activity. 
 
     
     
       16. A method for assessment of traumatic brain injury in a patient, comprising the steps of:
 connecting at least one neurological electrode to the patient's forehead to acquire brain electrical signals; and   providing a base unit operatively connected to the at least one neurological electrode to process the acquired brain electrical signals;
 wherein the base unit comprises a digital signal processor configured to perform automatic identification and removal of artifacts from brain electrical signals acquired by the at least one neurological electrode, extract one or more features from the acquired brain electrical signals, and execute at least one classification function to classify the patient into one of four categories indicative of the presence and severity of traumatic brain injury, wherein the at least one classification function is designed using a training database comprising a population of controls and patients who reportedly sustained closed head injuries and the training database is organized using Standard Assessment of Concussion (SAC) scores. 
   
     
     
       17. The method of  claim 16 , wherein the at least one classification function comprises at least three classification functions. 
     
     
       18. The method of  claim 17 , wherein the digital signal processor is configured to execute the at least three classification functions in cascade. 
     
     
       19. The method of  claim 17 , wherein the digital signal processor is configured to execute the at least three classification functions in a sequence, the sequence comprising the steps of:
 executing a first classification function designed to classify patients with structural brain injury from patients who are normal or have only functional brain injury;   executing a second classification function designed to classify normal patients from patients having structural and/or functional brain injury; and   executing a third classification function designed to classify patients with two separate grades of functional brain injury but no structural injury.   
     
     
       20. The method of  claim 17 , wherein the three classification functions comprise:
 a first classification function designed to classify patients with structural brain injury from patients who are normal or have only functional brain injury;   a second classification function designed to classify normal patients from patients having structural and/or functional brain injury; and   a third classification function designed to classify patients with severe and less severe manifestations of functional injury when no structural injury is present.   
     
     
       21. The method of  claim 16 , wherein the at least one classification function comprises at least two classification functions. 
     
     
       22. The method of  claim 21 , wherein the digital signal processor is configured to execute the at least two classification functions in parallel. 
     
     
       23. The method of  claim 21 , wherein the digital signal processor is configured to execute the at least two classification functions independent of each other. 
     
     
       24. The method of  claim 23 , wherein the base unit further comprises a user interface for displaying one or more classification performance measures to enable a clinician to make a decision about a category of the patient. 
     
     
       25. The method of  claim 21 , wherein the at least two classification functions are performed by the multi-stage classifier. 
     
     
       26. The method of  claim 16 , wherein the base unit further comprises a display unit for providing an indication of the presence and severity of traumatic brain injury. 
     
     
       27. The method of  claim 26  wherein the display unit displays the category that the patient is classified into. 
     
     
       28. The method of  claim 16 , wherein the four categories comprise:
 abnormal brain electrical activity consistent with structural brain injury; abnormal brain electrical activity consistent with non-structural injury with severe clinical manifestations of functional injury; abnormal brain electrical activity consistent with non-structural injury with less severe manifestations of functional injury; and normal brain electrical activity.   
     
     
       29. The method of  claim 16 , wherein the training database is organized using a series of clinical characteristics. 
     
     
       30. The method of  claim 16 , wherein the training database is used for testing performance of the three classification functions using cross-validation. 
     
     
       31. The method of  claim 30 , wherein said cross-validation is leave-one-out cross-validation. 
     
     
       32. The method of  claim 16 , wherein the one or more features comprise linear and/or non-linear quantitative features. 
     
     
       33. The method of  claim 16 , wherein the one or more quantitative features comprise mutual information features. 
     
     
       34. The method of  claim 16 , wherein the at least one classification function is designed using an evolutionary classifier builder algorithm. 
     
     
       35. The method of  claim 34 , wherein the evolutionary classifier builder algorithm comprises a genetic algorithm. 
     
     
       36. The method of  claim 34 , wherein the evolutionary classifier builder algorithm comprises Modified Random Mutation Hill Climbing algorithm.