A Fully Automated Visual Grading System for White Matter Hyperintensities of T2-Fluid Attenuated Inversion Recovery Magnetic Resonance Imaging.

The Fazekas scale is one of the most commonly used visual grading systems for white matter hyperintensity (WMH) for brain disorders like dementia from T2-fluid attenuated inversion recovery magnetic resonance (MR) images (T2-FLAIRs). However, the visual grading of the Fazekas scale suffers from low-intra and inter-rater reliability and high labor-intensive work. Therefore, we developed a fully automated visual grading system using quantifiable measurements. Our approach involves four stages: (1) the deep learning-based segmentation of ventricles and WMH lesions, (2) the categorization into periventricular white matter hyperintensity (PWMH) and deep white matter hyperintensity (DWMH), (3) the WMH diameter measurement, and (4) automated scoring, following the quantifiable method modified for Fazekas grading. We compared the performances of our method and that of the modified Fazekas scale graded by three neuroradiologists for 404 subjects with T2-FLAIR utilized from a clinical site in Korea. The Krippendorff's alpha across our method and raters (A) versus those only between the radiologists (R) were comparable, showing substantial (0.694 vs. 0.732; 0.658 vs. 0.671) and moderate (0.579 vs. 0.586) level of agreements for the modified Fazekas, the DWMH, and the PWMH scales, respectively. Also, the average of areas under the receiver operating characteristic curve between the radiologists (0.80 ± 0.09) and the radiologists against our approach (0.80 ± 0.03) was comparable. Our fully automated visual grading system for WMH demonstrated comparable performance to the radiologists, which we believe has the potential to assist the radiologist in clinical findings with unbiased and consistent scoring.

Rieu Z ，Kim RE ，Lee M ，Kim HW ，Kim D ，Yong J ，Kim J ，Lee M ，Lim H ，Kim J ... -  《Journal of Integrative Neuroscience》

Can white matter hyperintensities based Fazekas visual assessment scales inform about Alzheimer's disease pathology in the population?

Pradeep A ，Raghavan S ，Przybelski SA ，Preboske GM ，Schwarz CG ，Lowe VJ ，Knopman DS ，Petersen RC ，Jack CR Jr ，Graff-Radford J ，Cogswell PM ，Vemuri P ... -  《Alzheimers Research & Therapy》

Quantitative MRI of cerebral white matter hyperintensities: A new approach towards understanding the underlying pathology.

Interest in white matter hyperintensities (WMH), a radiological biomarker of small vessel disease, is continuously increasing. This is, in most part, due to our better understanding of their association with various clinical disorders, such as stroke and Alzheimer's disease, and the overlapping pathology of WMH with these afflictions. Although post-mortem histological studies have reported various underlying pathophysiological substrates, in vivo research has not been specific enough to fully corroborate these findings. Furthermore, post-mortem studies are not able to capture which pathological processes are the driving force of the WMH severity. The current study attempts to fill this gap by non-invasively investigating the influence of WMH on brain tissue using quantitative MRI (qMRI) measurements of the water content (H2O), the longitudinal (T1) and effective transverse relaxation times (T2∗), as well as the semi-quantitative magnetization transfer ratio (MTR), and bound proton fraction (ƒbound). In total, seventy subjects (age range 50-80 years) were selected from a population-based aging cohort study, 1000BRAINS. Normal appearing grey (NAGM) and white matter (NAWM), as well as deep (DWMH) and periventricular (PWMH) white matter hyperintensities, were segmented and characterized in terms of their quantitative properties. The subjects were then further divided into four grades according to the Fazekas rating scale of severity. Groupwise analyses of the qMRI values in each tissue class were performed. All five qMRI parameters showed significant differences between WMH and NAWM (p < 0.001). Importantly, the parameters differed between DWMH and PWMH, the latter having higher H2O, T1, T2∗ and lower MTR and ƒbound values (p < 0.001). Following grading according to the Fazekas scale, DWMH showed an increase in the water content, T1 and a decrease in bound proton fraction corresponding to severity, exhibiting significant changes in grade 3 (p < 0.001), while NAWM revealed significantly higher H2O values in grade 3 compared to grade 0 (p < 0.001). PWMH demonstrated an increase in T2∗ values (significant in grade 3, P < 0.001). These results are in agreement with previous histopathological studies and support the interpretation that both edema and myelin loss due to a possible breakdown of the blood-brain barrier and inflammation are the major pathological substrates turning white matter into DWMH. Edema being an earlier contributing factor to the pathology, as expressed in the elevated water content values in NAWM with increasing severity. In the case of PWMH, an altered fluid dynamic and cerebrospinal fluid leakage exacerbate the changes. It was also found that the pathology, as monitored by qMRI, evolves faster in DWMH than in the PWMH following the severity.

Iordanishvili E ，Schall M ，Loução R ，Zimmermann M ，Kotetishvili K ，Shah NJ ，Oros-Peusquens AM ... -  《-》

Visual Assessment of Age-Related White Matter Hyperintensities Using FLAIR Images at 3 T: Inter- and Intra-Rater Agreement.

Age-related white matter hyperintensities are frequent incidental findings on T2-weighted brain MRI, and they are evaluated in clinical practice using a visual rating scale. To evaluate inter- and intra-rater agreement in MRI visual evaluations of age-related white matter hyperintensities made by two radiologists with different levels of experience using a visual rating scale. Two radiologists of two different levels of experience separately rated age-related white matter hyperintensities in 40 consecutive 3-tesla brain MRI scans using the Fazekas and Schmidt visual rating scale. Ratings were made on axial FLAIR (fluid-attenuated inversion recovery) sequences. Two readings were made by each radiologist. Intra- and inter-rater agreement was statistically determined by using Cohen's weighted kappa analysis. Forty patients (21 females, 19 males; mean age = 57 ± 18.43 years) were included between September and October 2011. Mean values ± SD for visual scores were as follows: periventricular hyperintensities, between 1.175 ± 0.9 and 1.375 ± 0.89; number of deep white matter hyperintensity lesions, between 1.325 ± 1.18 and 1.575 ± 1.15, and extent of deep white matter hyperintensity lesions, between 0.925 ± 0.78 and 1.1 ± 0.74. Intra- and inter-rater agreement was very good (x03BA; values, 0.85-0.91 and 0.80-0.97, respectively) for each of the three visual scale criteria, with significant correlations between ratings (r = 0.95; p < 0.0001) and readings (r = 0.91; p < 0.0001). Visual assessment of age-related white matter hyperintensities by radiologists using a visual scale on FLAIR sequence is reproducible. Differences in experience level do not influence readings. Visual scale use is thus justified in common practice.

Boutet C ，Rouffiange-Leclair L ，Schneider F ，Camdessanché JP ，Antoine JC ，Barral FG ... -  《-》

Increased extracellular free water is related to white matter hyperintensity burden.

Lei X ，Qiu W ，Xu Z ，Yu J ，Lan H ... -  《-》