Virendra R Mishra1, Xiaowei Zhuang1, Dietmar Cordes1, Aaron Ritter1, and Charles Bernick2
1Cleveland Clinic Lou Ruvo Center for Brain Health, Las Vegas, NV, United States, 2University of Washington - Seattle, Seattle, WA, United States
1Cleveland Clinic Lou Ruvo Center for Brain Health, Las Vegas, NV, United States, 2University of Washington - Seattle, Seattle, WA, United States
Male boxers with neuropsychological impairment can be accurately
identified with prior hypothesis-driven MRI regions while outperforming regions
identified with regression-based techniques or a combination of hypothesis and
regression based methods.
Figure 5: Sensitivity, specificity,
accuracy, and area under the receiver operating curve (AUROC) is shown for each
machine learning (ML) algorithm (RBFN: blue, Linear SVM: orange, Nonlinear SVM:
gray, and random forest (yellow) for various features. Black dotted line
represent the performance of any random classifier and dotted-dash lines
represent 95th percentile of the benchmark measure for the
respective ML algorithm across a various combination of the feature set, and are shown
in the same colors as the bar-plot.
Figure 2: Top: Cluster showing a significant correlation between GM and years of fighting in impaired boxers.
Mean GM density was extracted from each impaired boxer and
plotted as a scatterplot. Middle: Cluster showing a significant
correlation between WM and psychomotor speed
in impaired boxers. Mean WM density was extracted from each impaired boxer and plotted as a scatterplot. Bottom: Cluster showing a significant correlation between WM density and psychomotor
speed in nonimpaired boxers. Mean WM density was extracted from each
nonimpaired boxer and plotted as a scatterplot.