Cerebral Small Vessel Disease: A Review Focusing on Pathophysiology, Biomarkers, and Machine Learning Strategies

Abstract

Cerebral small vessel disease (cSVD) has a crucial role in lacunar stroke and brain hemorrhages and is a leading cause of cognitive decline and functional loss in elderly patients. Based on underlying pathophysiology, cSVD can be subdivided into amyloidal and non-amyloidal subtypes. Genetic factors of cSVD play a pivotal role in terms of unraveling molecular mechanism. An important pathophysiological mechanism of cSVD is blood-brain barrier leakage and endothelium dysfunction which gives a clue in identification of the disease through circulating biological markers. Detection of cSVD is routinely carried out by key neuroimaging markers including white matter hyperintensities, lacunes, small subcortical infarcts, perivascular spaces, cerebral microbleeds, and brain atrophy. Application of neural networking, machine learning and deep learning in image processing have increased significantly for correct severity of cSVD. A linkage between cSVD and other neurological disorder, such as Alzheimer's and Parkinson's disease and non-cerebral disease, has also been investigated recently. This review draws a broad picture of cSVD, aiming to inculcate new insights into its pathogenesis and biomarkers. It also focuses on the role of deep machine strategies and other dimensions of cSVD by linking it with several cerebral and non-cerebral diseases as well as recent advances in the field to achieve sensitive detection, effective prevention and disease management.

Keywords: Biomarkers; Blood-brain barrier; Machine learning; Neuroimaging; Small vessel disease.

Figure 1.

Alteration in blood-brain barrier (BBB) and endothelial dysfunction in cerebral small vessel disease. (A) Schematic representation of the BBB in normal condition (healthy individual), which consists of the monolayer of endothelial cell, connected by tight junctions and resting on the basal lamina. Circulating blood cells, such as neutrophils and monocytes, are also part of the unit, given the close interaction with the luminal surface of endothelial cells and their role in immune surveillance. Tight junctions consist of three main groups of proteins. They are transmembrane proteins (claudins, occludin, cadherins) and accessory proteins. These proteins interact to form a barrier from which minimal passive extravasation of plasma proteins, inorganic solutes or even water molecules occur. (B) Disassembly of proteins forming tight junction causes disruption of tight junctions leading to increased BBB permeability to small and large macromolecules. (C) Progressive BBB damage and leakiness results in stiffening of the vessel wall due to degradation of basement membrane (BM) and accumulation of extracellular matrix component. Leakiness in BBB also leads to immune cell infilteration and inflammation. TJ, tight junction; AJ, adherence junction; ZO, zonula occludens; MM-9, matrix metalloproteinases-9.

Figure 2.

Molecular mechanism of several hereditary forms of cerebral small vessel disease namely cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy (CARASIL), and incontinentia pigmenti and their pathophysiology in small vessel causing a disturbed blood-brain barrier (image courtesy: AtheroPoint™, Atheropoint, Roseville, CA, USA). SMAD, mothers against decapentaplegic homolog; NEMO, nuclear factor κB (NF-κB) essential modulator; IKK, IκB kinase; TAK, TGF-β–activated kinase; TAB, TGF-β activated kinase 1 (MAP3K7) binding protein; TNFR, tumor necrosis factor receptor; GOM, granular osmophilic material; TGF-βR, transforming growth factor β receptor; ECM, extracellular matrix; LTBP-1, latent transforming growth factor β binding protein 1; LAP, latency-associated protein; HTRA1, high-temperature requirement a serine peptidase 1; N3ECD, extracellular domain of NOTCH3.

Figure 3.

Neuroimaging markers of cerebral small vessel disease. (A) Recent small subcortical infarct on diffusion weighted imaging (arrow). (B) Lacune on fluid-attenuated inversion recovery imaging (FLAIR) (arrow). (C) White matter hyperintensity on FLAIR imaging (arrows). (D) Perivascular spaces on T1-weighted imaging (arrows). (E) Deep microbleeds on gradient recalled echo (GRE) T2 weighted imaging (arrows). (F) Lobar cerebral microbleeds on GRE imaging (arrows).

Figure 4.

Architecture of typical Convolutional Neural Networks (CNN) for image processing. The typical CNN architecture for image processing consists of a series of layers of convolution filters, interspersed with a series of data reduction or pooling layers. Several convolutional and pooling layers are usually stacked on top of each other to form a deep model and retrieve more abstract feature representations. The fully connected layers interpret these feature representations and execute the function of high-level reasoning.