Jasmin Annica Kuhn-Keller

67 White matter hyperintensity shape and cognitive decline over 5 years 4 Figure 4.1. Illustration of the calculation of WMH shape markers from brain MRI scans. WMHs are automatically segmented from FLAIR images. The lateral ventricles are automatically segmented from T1 images and inflated in order to aid delineation of periventricular/confluent WMHs. Based on the segmentations, the shape markers (for periventricular/confluent WMH: solidity, convexity, concavity index, and fractal dimension; for deep WMH: eccentricity and fractal dimension) are calculated using the formula’s shown in the figure. 4.3.5 Neuropsychological testing Based on a cognitive test battery the composite scores were calculated for three cognitive domains: memory, executive function, and processing speed.10–12 Immediate- and delayed-recall portions of the California Verbal Learning Test13 were included in the memory composite score. The executive domain score included the Digits Backward Test14, the Spatial Working Memory Test of the Cambridge Neuropsychological Test Automated Battery15, and the Stroop Test, Part III (wordcolor interference). The processing speed domain score included the Digit Symbol Substitution Test14, the Figure Comparison Test16 and the Stroop Test17 Part I (word naming) and Part II (color naming). Domain scores were calculated by converting raw scores on each test to standardized z-scores and averaging the z-scores across the tests in each composite score per participant. Change in cognitive function was determined per domain by subtracting the baseline scores from the follow-up scores. 4.3.6 Statistical analysis To aid comparability of the results, solidity and convexity were inverted for the linear regression analyses. Solidity and WMH volumes were multiplied by 100 and natural log transformed due to non-normal distribution. Z-scores of WMH shape

RkJQdWJsaXNoZXIy MTk4NDMw