Classifying Concussion in University Athletes Using MRI and Cognitive Evaluation
Digital Document
Document
Handle |
Handle
http://hdl.handle.net/11134/20002:860655358
|
||||||
---|---|---|---|---|---|---|---|
Persons |
Persons
Creator (cre): Ly, Monica T.
Major Advisor (mja): Chen, Chi-Ming
Associate Advisor (asa): Fein, Deborah
Associated name (asn): Stevens, Michael
|
||||||
Title |
Title
Title
Classifying Concussion in University Athletes Using MRI and Cognitive Evaluation
|
||||||
Origin Information |
Origin Information
|
||||||
Parent Item |
Parent Item
|
||||||
Resource Type |
Resource Type
|
||||||
Digital Origin |
Digital Origin
born digital
|
||||||
Description |
Description
Background: Current methods of concussion assessment lack the objectivity and reliability to detect neurological injury. Advanced neuroimaging methods such as diffusion tensor imaging (DTI) and resting state functional magnetic resonance imaging (fMRI) have found abnormalities indicative of diffuse neuronal damage in concussed patients, while cognitive testing has shown short-term deficits in attention and memory. This multi-site study uses both neuroimaging and cognitive measures to train algorithms to detect the presence of concussion in university athletes. Methods: University athletes (32 concussed, 48 controls) completed symptom reports, brief cognitive evaluation, and MRI within 72 hours of injury. Hierarchical linear regression compared groups on cognitive and neuroimaging measures (mean diffusivity, fractional anisotropy, between- and within-network connectivity) while controlling for gender and site. Control athletes with and without a remote history of concussion were compared on all measures. Logistic regression and support vector machine models were trained using combinations of cognitive and neuroimaging measures and evaluated for overall accuracy, sensitivity, and specificity. Results: Concussed athletes reported greater symptoms than controls, and performed worse on tests of concentration and delayed memory. Concussed athletes showed lower functional connectivity within the frontoparietal and primary visual networks, but did not differ on mean diffusivity and fractional anisotropy. Athletes with a remote history of concussion closely resembled athletes with no history of concussion. Of the cognitive measures, classifiers trained using delayed memory yielded the best performance with overall accuracy of 70%. Of the neuroimaging measures, classifiers trained using mean diffusivity from a subset of tracts yielded similar accuracy. Combining cognitive measures with mean diffusivity increased sensitivity but decreased specificity. Conclusions: Neuroimaging techniques such as DTI and resting state fMRI can detect common neurobiological sequelae of concussion. The integration of neuroimaging and cognitive assessment with machine learning shows promise as an objective, reliable tool in the assessment and diagnosis of concussion.
|
||||||
Genre |
Genre
|
||||||
Organizations |
Organizations
Degree granting institution (dgg): University of Connecticut
|
||||||
Held By | |||||||
Rights Statement |
Rights Statement
|
||||||
Note |
Note
|
||||||
Degree Name |
Degree Name
Doctor of Philosophy
|
||||||
Degree Level |
Degree Level
Doctoral
|
||||||
Degree Discipline |
Degree Discipline
Psychology
|
||||||
Local Identifier |
Local Identifier
S_19573781
|