Premium
Comparing Two Processing Pipelines to Measure Subcortical and Cortical Volumes in Patients with and without Mild Traumatic Brain Injury
Author(s) -
Reid Matthew W.,
Hannemann Nathan P.,
York Gerald E.,
Ritter John L.,
Kini Jonathan A.,
Lewis Jeffrey D.,
Sherman Paul M.,
Velez Carmen S.,
Dren Ann Marie,
Bolzenius Jacob D.,
Tate David F.
Publication year - 2017
Publication title -
journal of neuroimaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.822
H-Index - 64
eISSN - 1552-6569
pISSN - 1051-2284
DOI - 10.1111/jon.12431
Subject(s) - medicine , traumatic brain injury , intraclass correlation , reliability (semiconductor) , concurrent validity , concussion , physical medicine and rehabilitation , medical physics , poison control , injury prevention , clinical psychology , psychometrics , emergency medicine , psychiatry , power (physics) , physics , quantum mechanics , internal consistency
PURPOSE To compare volumetric results from NeuroQuant® and FreeSurfer in a service member setting. Since the advent of medical imaging, quantification of brain anatomy has been a major research and clinical effort. Rapid advancement of methods to automate quantification and to deploy this information into clinical practice has surfaced in recent years. NeuroQuant® is one such tool that has recently been used in clinical settings. Accurate volumetric data are useful in many clinical indications; therefore, it is important to assess the intermethod reliability and concurrent validity of similar volume quantifying tools. METHODS Volumetric data from 148 U.S. service members across three different experimental groups participating in a study of mild traumatic brain injury (mTBI) were examined. Groups included mTBI ( n = 71), posttraumatic stress disorder ( n = 22), or a noncranial orthopedic injury ( n = 55). Correlation coefficients and nonparametric group mean comparisons were used to assess reliability and concurrent validity, respectively. RESULTS Comparison of these methods across our entire sample demonstrates generally fair to excellent reliability as evidenced by large intraclass correlation coefficients (ICC = .4 to .99), but little concurrent validity as evidenced by significantly different Mann‐Whitney U comparisons for 26 of 30 brain structures measured. CONCLUSION While reliability between the two segmenting tools is fair to excellent, volumetric outcomes are statistically different between the two methods. As suggested by both developers, structure segmentation should be visually verified prior to clinical use and rigor should be used when interpreting results generated by either method.