Introduction

White 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.

Methods

We 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.

Results

Random 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).

Conclusion

Our 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: We previously reported that infants who developed autism spectrum disorder (ASD) had increased cerebrospinal fluid (CSF) in the subarachnoid space (i.e., extra-axial CSF) from 6 to 24 months of age. We attempted to confirm and extend this finding in a larger independent sample. METHODS: A longitudinal magnetic resonance imaging study of infants at risk for ASD was carried out on 343 infants, who underwent neuroimaging at 6, 12, and 24 months. Of these infants, 221 were at high risk for ASD because of an older sibling with ASD, and 122 were at low risk with no family history of ASD. A total of 47 infants were diagnosed with ASD at 24 months and were compared with 174 high-risk and 122 low-risk infants without ASD. RESULTS: Infants who developed ASD had significantly greater extra-axial CSF volume at 6 months compared with both comparison groups without ASD (18% greater than high-risk infants without ASD; Cohens d = 0.54). Extra-axial CSF volume remained elevated through 24 months (d = 0.46). Infants with more severe autism symptoms had an even greater volume of extra-axial CSF from 6 to 24 months (24% greater at 6 months, d = 0.70; 15% greater at 24 months, d = 0.70). Extra-axial CSF volume at 6 months predicted which high-risk infants would be diagnosed with ASD at 24 months with an overall accuracy of 69% and corresponding 66% sensitivity and 68% specificity, which was fully cross-validated in a separate sample. CONCLUSIONS: This study confirms and extends previous findings that increased extra-axial CSF is detectable at 6 months in high-risk infants who develop ASD. Future studies will address whether this anomaly is a contributing factor to the etiology of ASD or an early risk marker for ASD.