Meeting Banner
Abstract #1116

Diagnostic Accuracy of Diffusion Tensor Imaging for Pediatric Cervical Spinal Cord Injury

Nadia Barakat1, MJ Mulcahey2, John Gaughan1, Pallav Shah1, Scott Faro1, Amer Samdani2, Randal Betz2, Jrgen Finsterbusch3, Feroze Mohamed1

1Temple University, Philadelphia, PA, United States; 2Shriners Hospital for Children, Philadelphia, PA, United States; 3University Medical Center Hamburg-Eppendorf, Hamburg, Germany

Few studies have evaluated the diagnostic accuracy of Diffusion Tensor Imaging (DTI) for spinal cord injury (SCI) but none have involved children. In this study the predictive validity of DTI was evaluated by examining its diagnostic accuracy for pediatric cervical SCI in 35 children. The specificity, sensitivity, receiver operating characteristics area under the curve (ROC AUC) and corresponding 95% confidence intervals (CI) of DTI for the ISNCSCI clinical endpoints and MRI level of injury were calculated. Resampling methods were used to validate the estimates from the final models.

Abstract PDF Poster PDF

Keywords

abnormal absent accuracy adjusted among anal analyze anesthesia assent association atrophy attention automatic axial background board bootstrap certified cervical characteristics children classification classified clinical combination complete concurrent confidence considered consistency contra contraction control controls conversion cord correction cover curve deep designation determining developing deviation diagnostic differ difficult diffusion done drawn endpoint endpoints estimation evaluate evaluated evaluating examination examining fast field final findings gating good hamburg hospital impairment important included incomplete indicated informed injury inner intact international intervals jackknife john knowledge locating locations logistic materials measures median models moderate motion motor multivariate muscle neurological novel objective observation operating overall parents particularly patients pediatric predicting predictive predictors pres presence pressure pulse received recent registration regressions relevance reliability reliably remaining reported respiratory sacral scanned scanner sens sensation sensitivity severity shah significantly slice slices sparing specificity spinal spine strength strong strongest structural subjects table temple tensor transition typically underwent univariate validate validity variable variance view