Use of Multimodal Magnetic Resonance Imaging Techniques to Explore Cognitive Impairment in Leukoaraiosis

Identifying Leukoaraiosis with Mild Cognitive Impairment by Fusing Multiple MRI Morphological Metrics and Ensemble Machine Learning

Leukoaraiosis (LA) is strongly associated with impaired cognition and increased dementia risk. Determining effective and robust methods of identifying LA patients with mild cognitive impairment (LA-MCI) is important for clinical intervention and disease monitoring. In this study, an ensemble learning method that combines multiple magnetic resonance imaging (MRI) morphological features is proposed to distinguish LA-MCI patients from LA patients lacking cognitive impairment (LA-nCI). Multiple comprehensive morphological measures (including gray matter volume (GMV), cortical thickness (CT), surface area (SA), cortical volume (CV), sulcus depth (SD), fractal dimension (FD), and gyrification index (GI)) are extracted from MRI to enrich model training on disease characterization information. Then, based on the general extreme gradient boosting (XGBoost) classifier, we leverage a weighted soft-voting ensemble framework to ensemble a data-level resampling method (Fusion + XGBoost) and an algorithm-level focal loss (FL)-improved XGBoost model (FL-XGBoost) to overcome class-imbalance learning problems and provide superior classification performance and stability. The baseline XGBoost model trained on an original imbalanced dataset had a balanced accuracy (Bacc) of 78.20%. The separate Fusion + XGBoost and FL-XGBoost models achieved Bacc scores of 80.53 and 81.25%, respectively, which are clear improvements (i.e., 2.33% and 3.05%, respectively). The fused model distinguishes LA-MCI from LA-nCI with an overall accuracy of 84.82%. Sensitivity and specificity were also well improved (85.50 and 84.14%, respectively). This improved model has the potential to facilitate the clinical diagnosis of LA-MCI.

White matter integrity mediates the associations between white matter hyperintensities and cognitive function in patients with silent cerebrovascular diseases

OBJECTIVE

To evaluate the relationships between cognitive function and white matter hyperintensity volume (WMHV) in patients with silent cerebrovascular disease and to investigate whether white matter integrity or brain atrophy play a role in this association.

METHODS

Automated Fiber Quantification and Voxel- based morphometry were used to track and identify the integrity of 20 well-defined white matter tracts and to measure the gray matter volume (GMV). A linear regression model was applied for examining the associations between cognitive function and WMHV and mediation analysis was used to identify the roles of white matter integrity or GMV in the influence of WMHV on cognitive function.

RESULTS

Two hundred and thirty-six individuals were included for analysis. Executive function was linearly associated with fractional anisotropy (FA) of the right interior frontal occipital fasciculus (IFOF) (β = 0.193; 95% CI, 0.126 to 1.218) and with WMHV (β = -0.188; 95% CI, -0.372 to -0.037). Information processing speed was linearly associated with WMHV (β = -0.357; 95% CI, -0.643 to -0.245), FA of the right anterior thalamic radiation (ATR) (β = 0.207; 95% CI, 0.116 to 0.920), and FA of the left superior longitudinal fasciculus (SLF) (β = 0.177; 95% CI, 0.103 to 1.315). The relationship between WMHV and executive function was mediated by FA of the right IFOF (effect size = -0.045, 95% CI, -0.015 to -0.092). Parallel mediation analysis showed that the association between WMHV and information processing speed was mediated by FA of the right ATR (effect size = -0.099, 95% CI, -0.198 to -0.038) and FA of the left SLF (effect size = -0.038, 95% CI, -0.080 to -0.003).

CONCLUSION

These findings suggest a mechanism by which WMH affects executive function and information processing speed by impairing white matter integrity. This may be helpful in providing a theoretical basis for rehabilitation strategies of cognitive function in patients with silent cerebrovascular diseases.

The contribution of white matter pathology, hypoperfusion, lesion load, and stroke recurrence to language deficits following acute subcortical left hemisphere stroke

Aphasia, the loss of language ability following damage to the brain, is among the most disabling and common consequences of stroke. Subcortical stroke, occurring in the basal ganglia, thalamus, and/or deep white matter can result in aphasia, often characterized by word fluency, motor speech output, or sentence generation impairments. The link between greater lesion volume and acute aphasia is well documented, but the independent contributions of lesion location, cortical hypoperfusion, prior stroke, and white matter degeneration (leukoaraiosis) remain unclear, particularly in subcortical aphasia. Thus, we aimed to disentangle the contributions of each factor on language impairments in left hemisphere acute subcortical stroke survivors. Eighty patients with acute ischemic left hemisphere subcortical stroke (less than 10 days post-onset) participated. We manually traced acute lesions on diffusion-weighted scans and prior lesions on T2-weighted scans. Leukoaraiosis was rated on T2-weighted scans using the Fazekas et al. (1987) scale. Fluid-attenuated inversion recovery (FLAIR) scans were evaluated for hyperintense vessels in each vascular territory, providing an indirect measure of hypoperfusion in lieu of perfusion-weighted imaging. We found that language performance was negatively correlated with acute/total lesion volumes and greater damage to substructures of the deep white matter and basal ganglia. We conducted a LASSO regression that included all variables for which we found significant univariate relationships to language performance, plus nuisance regressors. Only total lesion volume was a significant predictor of global language impairment severity. Further examination of three participants with severe language impairments suggests that their deficits result from impairment in domain-general, rather than linguistic, processes. Given the variability in language deficits and imaging markers associated with such deficits, it seems likely that subcortical aphasia is a heterogeneous clinical syndrome with distinct causes across individuals.

Pathogenesis and research progress in leukoaraiosis

Leukoaraiosis is a common imaging marker of cerebral small vessel disease. In recent years, with the continuous advances in brain imaging technology, the detection rate of leukoaraiosis is higher and its clinical subtypes are gradually gaining attention. Although leukoaraiosis has long been considered an incidental finding with no therapeutic necessity, there is now growing evidence linking it to, among other things, cognitive impairment and a high risk of death after stroke. Due to different research methods, some of the findings are inconsistent and even contradictory. Therefore, a comprehensive and in-depth study of risk factors for leukoaraiosis is of great clinical significance. In this review, we summarize the literature on leukoaraiosis in recent years with the aim of elucidating the disease in terms of various aspects (including pathogenesis, imaging features, and clinical features, etc.).

[Association of sudden sensorineural hearing loss and its prognosis with the brain white matter hyperintensity]

Objective:To explore the relationship between white matter hyperintensity（WMH） and sudden sensorineural hearing loss（SSHL） and evaluate the influence of WMH on the prognosis of SSHL. Methods:Fifty hospitalized unilateral SSHL patients and 50 age and gender matched routine physical examination individuals without SSHL history from June 2019 to June 2020 were included for a case-control study. All included subjects underwent 3.0 Tesla cranial magnetic resonance examination, and the Fazekas scale was applied to evaluate periventricular white matter hyperintense（PVWMH） and deep white matter hyperintense（DWMH）. Fazekas score and distribution proportions of Fazekas score was compared between SSHL and control. Ordered logistic regression was used to study the relationship between prognosis of SSHL and WMH. Results:The Fazekas score of WMH in SSHL group was significantly higher than that of control group（PVWMH: P=0.004, DWMH: P=0.010）; There was a significant difference in the distribution proportions of Fazekas scores between SSHL and control（PVWMH: P=0.036, DWMH: P=0.047）; The results of ordered logistic regression showed that patients without WMH（Fazekas=0） is an independent predictor of good prognosis in SSHL（P=0.025, OR=12.779）. Conclusion:The prevalence of WMH in SSHL patients was higher than that of control. SSHL patients without WMH has a better prognosis than those with WMH.

The relationship between diabetes-related cognitive dysfunction and leukoaraiosis

Cognitive dysfunction is a degenerative disease of the central nervous system, which often associates with ageing brain as well as neurodegenerative diseases. A growing body of evidence suggests that patients with diabetes mellitus (DM) have a significantly higher risk of cognitive impairment. In recent years, studies have found that patients with diabetes-related cognitive dysfunction have an increased burden of leukoaraiosis (LA), and larger white matter hyperintensity (WMH) volume. With the recent advancement of technologies, multimodal imaging is widely exploited for the precise evaluation of central nervous system diseases. Emerging studies suggest that LA pathology can be used as a predictive signal of white matter lesions in patients with diabetes-related cognitive dysfunction, providing support for early identification and diagnosis of disease. This article reviews the findings, epidemiological characteristics, pathogenesis, imaging features, prevention and treatment of LA pathophysiology in patients with diabetes-related cognitive dysfunction.

Claudin-1 and Claudin-3 as Molecular Regulators of Myelination in Leukoaraiosis Patients

OBJECTIVES

Leukoaraiosis is described as white matter lesions that are associated with cognitive dysfunction, neurodegenerative disorders, etc. Myelin depletion is a salient pathological feature of, and the loss of oligodendrocytes is one of the most robust alterations evident in, white matter degeneration. Recent studies have revealed that claudin proteins are aberrantly expressed in leukoaraiosis and regulate oligodendrocyte activity. However, the roles of claudin-1 and claudin-3 in oligodendrocytes and leukoaraiosis are still not well-defined.

METHODS

Quantitative polymerase chain reaction was used to measure the expression of claudin-1 (CLDN1), claudin-3 (CLDN3), and myelinogenesis-related genes such as myelin basic protein (MBP), proteolipid protein (PLP), oligodendrocyte transcription factor 2 (OLIG2), and SRY-box transcription factor 10 (SOX10) in leukoaraiosis patients (n=122) and healthy controls (n=122). The expression of claudin-1 and claudin-3 was either ectopically silenced or augmented in Oli-neu oligodendrocytes, and colony formation, apoptosis, and migration assays were performed. Finally, the expression of myelin proteins was evaluated by western blotting.

RESULTS

Our results revealed that in addition to SOX10, the expression levels of claudin-1, claudin-3, and myelinogenesis-related proteins were prominently downregulated in leukoaraiosis patients, compared to those in healthy controls. Furthermore, the growth and migration of Oli-neu cells were downregulated upon silencing claudin-1 or claudin-3. However, the overexpression of claudin-1 or claudin-3 resulted in the reduction of the degree of apoptosis in Oli-neu cells. In addition, claudin-1 and claudin-3 promoted the expression of MBP, OLIG2, PLP, and SOX10 at the translational level.

CONCLUSION

Our data has demonstrated that the abnormal expression of claudin-1 and claudin-3 regulates the pathological progression of leukoaraiosis by governing the viability and myelination of oligodendrocytes. These findings provide novel insights into the regulatory mechanisms underlying the roles of claudin-1 and claudin-3 in leukoaraiosis.

Brain White Matter: A Substrate for Resilience and a Substance for Subcortical Small Vessel Disease

Age-related brain white matter disease is a form of small vessel disease (SVD) that may be associated with lacunar and other small subcortical infarcts, cerebral microbleeds, and perivascular spaces. This common form of cerebrovascular disease may manifest clinically as cognitive impairment of varying degrees and difficulty with mobility. Whereas some persons show cognitive decline and mobility failure when there are brain white matter hyperintensities (WMH) and acute stroke, others recover, and not everyone with brain white matter disease is disabled. Thus, repair or compensation of brain white matter may be possible, and furthermore, certain vascular risks, such as raised blood pressure, are targets for prevention of white matter disease or are administered to reduce the burden of such disease. Vascular risk modification may be useful, but alone may not be sufficient to prevent white matter disease progression. In this chapter, we specifically focus on WMH of vascular origin and explore white matter development, plasticity, and enduring processes of myelination across the health span in the context of experimental and human data, and compare and contrast resilient brain white matter propensity to a diseased white matter state. We conclude with thoughts on novel ways one might study white matter resilience, and predict future healthy cognitive and functional outcomes.