Search

IntroductionWhite matter hyperintensities (WMHs) are areas of abnormal signal on magnetic resonance images (MRIs) that characterize various types of histopathological lesions. The load and location of WMHs are important clinical measures that may indicate the presence of small vessel disease in aging and Alzheimer's disease (AD) patients. Manually segmenting WMHs is time consuming and prone to inter-rater and intra-rater variabilities. Automated tools that can accurately and robustly detect these lesions can be used to measure the vascular burden in individuals with AD or the elderly population in general. Many WMH segmentation techniques use a classifier in combination with a set of intensity and location features to segment WMHs, however, the optimal choice of classifier is unknown.MethodsWe compare 10 different linear and nonlinear classification techniques to identify WMHs from MRI data. Each classifier is trained and optimized based on a set of features obtained from co-registered MR images containing spatial location and intensity information. We further assess the performance of the classifiers using different combinations of MRI contrast information. The performances of the different classifiers were compared on three heterogeneous multi-site datasets, including images acquired with different scanners and different scan-parameters. These included data from the ADC study from University of California Davis, the NACC database and the ADNI study. The classifiers (naïve Bayes, logistic regression, decision trees, random forests, support vector machines, k-nearest neighbors, bagging, and boosting) were evaluated using a variety of voxel-wise and volumetric similarity measures such as Dice Kappa similarity index (SI), Intra-Class Correlation (ICC), and sensitivity as well as computational burden and processing times. These investigations enable meaningful comparisons between the performances of different classifiers to determine the most suitable classifiers for segmentation of WMHs. In the spirit of open-source science, we also make available a fully automated tool for segmentation of WMHs with pre-trained classifiers for all these techniques.ResultsRandom Forests yielded the best performance among all classifiers with mean Dice Kappa (SI) of 0.66±0.17 and ICC=0.99 for the ADC dataset (using T1w, T2w, PD, and FLAIR scans), SI=0.72±0.10, ICC=0.93 for the NACC dataset (using T1w and FLAIR scans), SI=0.66±0.23, ICC=0.94 for ADNI1 dataset (using T1w, T2w, and PD scans) and SI=0.72±0.19, ICC=0.96 for ADNI2/GO dataset (using T1w and FLAIR scans). Not using the T2w/PD information did not change the performance of the Random Forest classifier (SI=0.66±0.17, ICC=0.99). However, not using FLAIR information in the ADC dataset significantly decreased the Dice Kappa, but the volumetric correlation did not drastically change (SI=0.47±0.21, ICC=0.95).ConclusionOur investigations showed that with appropriate features, most off-the-shelf classifiers are able to accurately detect WMHs in presence of FLAIR scan information, while Random Forests had the best performance across all datasets. However, we observed that the performances of most linear classifiers and some nonlinear classifiers drastically decline in absence of FLAIR information, with Random Forest still retaining the best performance.

Background and Objectives: There is no disease-modifying treatment of corticobasal syndrome (CBS) and progressive supranuclear palsy (PSP), 2 disorders characterized by their striking phenotypic, and, in CBS, pathologic heterogeneity. Seed amplification assays (SAAs) could enable the detection of neuropathologic processes, such as α-synuclein (αSyn) copathology, that affect the success of future disease-modifying treatment strategies. The primary objective was to assess possible αSyn copathology in CBS and PSP, as detected in CSF using an αSyn SAA (αSyn-SAA). Secondary objectives were to evaluate the association of αSyn-SAA positivity with other biomarkers including of Alzheimer disease (AD), and with clinical presentation. We hypothesized that αSyn-SAA positivity would be detectable in CBS and PSP and that it would be associated with AD biomarker positivity and β-amyloid (Aβ) 42 levels, neurodegeneration as assessed by neurofilament light chain (NfL) levels, and symptoms associated with synucleinopathies., Methods: This cross-sectional observational study included patients clinically diagnosed with CBS and PSP who underwent a lumbar puncture between 2012 and 2021 (Toronto Western Hospital, Canada). CSF was tested for αSyn-SAA positivity, AD biomarkers, and NfL levels. Clinical data were derived from medical records., Results: We tested the CSF of 40 patients with CBS (19 female patients, 65.9 ± 8.6 years) and 28 with PSP (13 female patients, 72.5 ± 8.7 years old), mostly White (n = 50) or Asian (n = 14). αSyn-SAA positivity was observed in 35.9% patients with CBS and 28.6% with PSP. In young-onset, but not late-onset patients, αSyn-SAA positivity and AD positivity were associated (odds ratio [OR] 8.8, 95% CI 1.2-82.6, p < 0.05). A multivariable linear regression analysis showed a significant interaction of αSyn-SAA status by age at onset on CSF Aβ42 levels (β = 0.3 ± 0.1, p < 0.05). Indeed, age at onset was positively related to Aβ42 levels only in αSyn-SAA-positive patients, as shown by slope comparison. A logistic regression analysis also suggested that REM sleep behavior disorder was associated with αSyn-SAA positivity (OR 60.2, 95% CI 5.2-1,965.8; p < 0.01)., Discussion: We detected a frequency of αSyn-SAA positivity in CBS and PSP in line with pathologic studies, highlighting the usefulness of SAAs for in vivo detection of otherwise undetectable neuropathologic processes. Our results also suggest that AD status (specifically low Aβ42) and older age at onset may contribute to αSyn-SAA positivity. This opens new perspectives for the stratification of patients in clinical trials.