The evolving landscape of biomarkers for vascular dementia

Abstract

Recent progress in cerebrovascular research has revealed new aspects of the pathogenesis of vascular dementia. Newly uncovered molecular mechanisms underlying the physiological and pathological functions of cerebrovasculature have provided an evolving landscape for understanding the pathogenesis of vascular dementia. These findings have, in turn, promoted the development of biomarkers for predicting, diagnosing, and assessing the severity of vascular dementia. Furthermore, the recent approval of an anti-amyloid antibody treatment against Alzheimer’s disease has brought researchers’ attention to the interplay between Alzheimer’s amyloid-β and cerebrovasculature, as amyloid-related imaging abnormalities occurred in a certain number of patients receiving the therapy. Cerebral amyloid angiopathy, the deposition of amyloid-β in the brain’s blood vessels, may play a role in the interplay between Alzheimer’s and vascular dementia. Here, we review recent developments in research on vascular dementia related to the findings concerning imaging, biofluid, and digital biomarkers for diseases.

 1. Introduction

The recent literature on dementia suggests that early diagnosis and intervention can prevent the progression of dementia, which adds weight to objective biomarkers for assessing and diagnosing the condition. Vascular dementia (VaD) is classically defined by a cognitive decline with concomitant evidence of stroke or ischemic changes in brain-image analysis, such as magnetic resonance imaging (MRI) or computerized tomography (CT). Recent progress in basic research on the pathogenesis of vascular dementia has revealed multifactorial aspects of the disease, which have provided a new avenue for biomarker development. Biomarkers associated with the molecular mechanisms of cerebrovascular dysfunction may enable the detection of VaD at the earliest stage. Furthermore, amyloid-related imaging abnormalities (ARIA), which are recognized as a side effect of anti-amyloid antibody therapy for Alzheimer’s disease (AD), have drawn the attention of researchers in the dementia field. Developing biomarkers to predict the risk of ARIA is in high demand for successful treatment with anti-amyloid antibody therapy. In the current review, we summarize the recent progress made in biomarker developments related to VaD, with a particular focus on imaging, biofluid, and digital biomarkers.

 2. Imaging biomarkers for VaD

Various diagnostic criteria and classifications of VaD have been proposed, and all include imaging findings as an important component. Imaging biomarkers play an critical role in the management of vascular cognitive impairment (VCI). Classic diagnostic criteria for VaD include the International Classification of Diseases, Tenth Revision (ICD-10); the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5); the Alzheimer’s Disease Diagnostic and Treatment Centers (ADDTC) criteria; and the National Institute of Neurological Disorders and Stroke (NINDS) - Association Internationale pour la Recherché et l’Enseignement en Neurosciences (AIREN) criteria. The term “VCI” has also been proposed as an umbrella concept that includes the prodromal stages of dementia¹⁾. The NINDS-AIREN classification²⁾ lists the following subtypes: (1) multi-infarct, (2) single-infarct, (3) small-vessel disease, (4) hypoperfusion, (5) hemorrhage, and (6) other mechanisms. The Vienna Impairment of Cognition Classification Consensus Study (VICCCS)³⁾ proposed the following subtypes: (1) poststroke, (2) multi-infarct, (3) subcortical ischemic, and (4) mixed.

A deep understanding of the pathological significance of each imaging finding is crucial for the accurate diagnosis of VaD. In recent years, imaging biomarkers of various modalities have been used in the development and clinical testing of new drugs, leading to a better understanding of the pathogenesis of not only VaD but also other neurodegenerative diseases, including AD. This section outlines the significance of key imaging biomarkers and emerging concepts in VaD pathogenesis.

 2.1. Key imaging biomarkers associated with VaD

 Stroke

Stroke is a major cause of VaD. Post-stroke cognitive impairment (PSCI) occurs in approximately half of all patients within one year after a stroke⁴⁾. Meta-analyses on strategic lesion locations for VCI based on lesion-symptom mapping (the Meta VCI Map consortium) have evaluated the impact of previously proposed infarction in strategic sites on cognitive function⁵⁾. According to the results, left frontal infarcts in the left frontal temporal lobe, left thalamus, and right parietal lobe are strongly associated with PSCI.

In a cohort study⁶⁾ examining the impact of infarcts in cortical, subcortical, cerebellar and whole brain regions on cognitive function, people with new subcortical infarcts had the greatest risk of developing dementia, compared with those without infarcts. This suggests that infarcts resulting from small vessel diseases have a greater impact on the development of dementia than embolic infarcts in large vessels.

 Cortical cerebral microinfarcts

Cortical cerebral microinfarcts (CMI) have been reported as markers of both large and small vessel lesions and are associated with dementia⁷⁾. In brain MRI, CMI is defined by small lesions with hypointense signals on T1-weighted, hyperintense signals on T2-weighted or fluid-attenuated inversion recovery (FLAIR), and isointense signals on T2*-weighted images, which are located in the cortical area with an upper size limit of 4 mm⁸⁾. Development of the 3T MRI has enabled the detection of these lesions.

 White matter hyperintensities

A meta-analysis has shown that white matter hyperintensities (WMH) are associated with cognitive decline and dementia development⁹⁾. In a brain MRI, WMH are defined as signal abnormalities of variable sizes in the white matter that are hyperintense on T2-weighted images, without cavitation (a signal different from cerebral spinal fluid (CSF))⁸⁾. WMH reflect demyelination, axon loss, and gliosis¹⁰⁾ and are generally classified as periventricular WMH (PWMH) and deep WMH (DWMH), with PWMH reported to have a stronger association with cognitive impairment¹¹⁾. Damages in the high density of periventricular long association fibers around the periventricular area are thought to be responsible for the resulting executive dysfunction¹²⁾. A number of methods for the automatic or semi-automatic assessment of white matter high signals have been developed in recent years;¹³⁾ however, their reproducibility and comparability have not yet been adequately tested.

 Perivascular space

Enlargement of the perivascular space (PVS) is considered an indicator of its dysfunction¹⁴⁾. PVS plays an important role in the removal of metabolic waste products from neurons in mice¹⁵⁾, and its dysfunction is thought to contribute to various neurological disorders. In a brain MRI, PVS defined as a fluid-filled space along a vessel in the brain; it appears round, ovoid, or linear shape and is typically under 2 mm in diameter⁸⁾. An impaired efflux of amyloid-β via the PVS has been implicated in the pathogenesis of AD^16,17).

 Cerebral microbleeds

A meta-analysis of five studies showed that the presence of a cerebral microbleed (CMB) was not associated with the development of dementia or AD; however, it has been reported as related to the risk of stroke and death⁹⁾. In a brain MRI, a CBM is defined as small (typically 2–5 mm in size) area of signal void with an associated blooming artifact on T2*-weighted images. Neuropathological studies using ex vivo 3T MRI on autopsy brains have confirmed that most lesions meeting these imaging criteria are, indeed, microbleeds on neuropathological examination¹⁸⁾.

 Cortical superficial siderosis

A finding of cortical superficial siderosis (cSS), along with CMB, WMH, and centrum semiovale PVS (CSO-PVS), suggests the existence of cerebral amyloid angiopathy¹⁹⁾. cSS is due to a convexal subarachnoid hemorrhage, superficial cortical hemorrhage resulting from vascular malformation, hemorrhagic changes in the infarct, or trauma. cSS is defined by thin T2*-hypointense areas on an MRI, often localized to one gyrus or sulci, and may sometimes extend to multiple brain regions.

 Microstructural changes based on diffusion tensor images

A diffusion tensor image (DTI) is based on diffusion-weighted imaging (DWI). While DWI assesses diffusion isotropically, DTI takes anisotropy into account, allowing the assessment of fiber bundles and the perivascular space. Several indices have been proposed based on this principle. Diffusion tensor tractography assesses white matter damage by visualizing nerve fiber bundles in three dimensions, which may predict PSCI²⁰⁾. Analysis along the perivascular space (DTI-ALPS) is reported to assess the glymphatic system by visualizing the movement of water molecules in the direction of the perivascular space²¹⁾. This technique has been used to assess amyloid-β efflux pathways in AD research. Recent findings have suggested that glymphatic system impairment assessed by DTI-ALPS is associated with cognitive impairment in patients with small vessel disease²²⁾.

 Changes in regional cerebral blood flow

Brain perfusion single photon emission tomography (SPECT) is capable of assessing the distribution of cerebral blood flow and is useful for the differential diagnosis of dementia and for the assessment of cerebrovascular disease²³⁾. Patients with AD show reduced tracer uptake in the parieto-occipital lobe, whereas those with multi-infarct dementia show reduced uptake in multiple cortical regions²³⁾.

 2.2. Emerging concepts for the pathogenesis of VaD

 Blood-brain barrier disruption

Disruption of the blood-brain barrier (BBB) leads to inflammatory and immune responses and reduced cerebral blood flow, leading to neurodegeneration²⁴⁾. The BBB is composed of vascular endothelial cells, pericytes, basement membranes, and astrocytes and is responsible for maintaining brain homeostasis and providing an optimal environment for neural activity. Dynamic contrast-enhanced MRI using gadolinium contrast has been reported to be useful for evaluating BBB disruption;²⁵⁾ however, there are concerns about the gadolinium’s deposition in the brain and complications in patients with renal dysfunction. Recently, diffusion-prepared arterial spin labeling (DP-ASL)²⁶⁾ has been developed to assess water clearance across the BBB by irradiating radio waves in the carotid artery region to magnetically label water molecules in the vessel and then deliver diffusion-weighted pulses without the use of contrast agents.

 Neurovascular unit

The concept of the neurovascular unit (NVU) was proposed by NINDS in 2001 as the smallest unit of neuronal function; this idea has now been widely accepted²⁷⁾. The NVU is thought to maintain neuronal function through interactions and cross-talk among neurons, oligodendrocytes, microglia, and the extracellular matrix, in addition to BBB component cells. Recent single-cell transcriptomics studies have suggested that there is no single NVU clone and that it constitutes a diverse neurovascular complex²⁸⁾. Microglia²⁹⁾, astrocytes³⁰⁾, and oligodendrocytes³¹⁾ have been shown as involved in the pathogenesis of dementia, and methods for capturing these components as imaging biomarkers may have important implications for therapeutic development.

 Cerebral amyloid angiopathy

Cerebral amyloid angiopathy (CAA) is an amyloidosis of the cerebral vessels, mainly composed of neuron-derived amyloid-β^32,33). Recently, the Boston criteria Ver. 2.0¹⁹⁾ was proposed as diagnostic criteria for CAA. In addition to CMB and cSS as imaging findings contributing to the diagnosis, convexal subarachnoid hemorrhage, severe CSO-PVS, and multi-spot WMH have been included. Patients with CAA are known to be at high risk of amyloid-related imaging abnormalities (ARIA), a side effect of anti-amyloid-β antibody drugs³⁴⁾. This has recently attracted considerable attention in the AD research field.

 Glymphatic system

Iliff et al. proposed a glymphatic system in which CSF flows in from the perivascular space around arteries, mixes with interstitial fluid from astrocytes, and drains into the perivascular space around veins, taking with it waste proteins such as amyloid-β¹⁵⁾. DTI-ALPS was subsequently developed to visualize CSF and interstitial fluid dynamics in humans and has shown a strong correlation with the dynamics of intrathecal gadolinium contrast media³⁵⁾. Researchers are now using DTI-ALPS as an indicator of the glymphatic system in humans.

 2.3. Perspective

This section outlined imaging biomarkers and pathological concepts related to VaD. Advances in imaging techniques have allowed us to uncover the multiple aspects of VaD pathogenesis. Results from imaging biomarker research may also play a critical role in therapeutic development.

 3. Biofluid biomarkers for VaD

The need for fluid biomarkers in VCI and VaD has emerged with the increasing body of evidence demonstrating the complexity of the disease’s pathogenesis. VaD is primarily evaluated using neuroimaging. Recent studies have clarified the molecular mechanisms underlying VaD, and the development of biomarkers for detecting them is in high demand. Here, we review disease-related proteins and extracellular matrix components as candidate biomarkers reflecting the key pathophysiological processes of VaD, including endothelial dysfunction, BBB disruption, inflammation, oxidative stress, and neuronal degeneration.

 3.1. From endothelial dysfunction to BBB disruption

VaD involves multiple interrelated pathophysiological processes³⁶⁾. Atherosclerosis and arteriosclerosis increase the production of asymmetric dimethylarginine (ADMA), which inhibits endothelial nitric oxide synthase (eNOS) and reduces nitric oxide (NO) production, leading to endothelial dysfunction. Consequently, hypoperfusion and hypoxia occur in the tissues, resulting in an increased expression of intercellular adhesion molecule-1 (ICAM-1) and selectins, which enhances vascular permeability. Chronic hypoperfusion, endothelial damage, and increased vascular permeability promote the extravasation of fibrinogen, activating matrix metalloproteinases (MMP)-2, MMP-3, and MMP-9, which disrupt the BBB. The breakdown of the BBB permits the extravasation of red blood cells and neurotoxic proteins, such as albumin, fibrinogen, and low-density lipoprotein (LDL) cholesterol, into the brain parenchyma, leading to oxidative stress, neuronal damage, and ultimately neuronal death.

Molecular markers that reflect the transition of molecules across vessels or their qualitative changes have the potential to serve as biomarkers for BBB disruption and VaD. The CSF–serum albumin quotient (QAlb), reflecting BBB disruption, is often elevated in VaD³⁷⁾, particularly in subcortical ischemic vascular dementia (SIVD), compared to AD³⁸⁾. The CSF–serum ratio of MMPs has also been proposed as a biomarker for VaD;³⁹⁾ however, findings remain inconsistent, as some studies reported no significant differences in CSF MMP-3 levels between patients with SIVD and controls⁴⁰⁾.

 3.2. Mechanisms of inflammation and the role of microglial activation

Inflammation within and outside the vasculature, accompanied by microglial activation, is implicated in VaD. Microglia are activated by extravasated fibrinogen, adopting either pro-inflammatory or anti-inflammatory phenotypes. Pro-inflammatory microglia, induced by IFN-γ, release neurotoxic mediators such as NO, TNF, IL-6, and IL-1β and contribute to neuronal damage. Conversely, anti-inflammatory microglia, induced by IL-4 and IL-13, secrete IL-10, IL-33, TGF-β, and insulin-like growth factor-1 (IGF-1), provide neuroprotective and angiogenic effects. Monitoring these phenotypes via fluid biomarkers could enhance the understanding of VaD pathogenesis.

TGF-β plays a crucial role in angiogenesis, neuroprotection, immune regulation, and synaptic plasticity and is potentially associated with VaD. Recent studies on the cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy (CARASIL) and COL4A1-COL4A2-related disorders, which represent the genetic model of small vessel disease, suggest that TGF-β may serve as a biomarker for VaD⁴¹⁾. CARASIL is characterized by progressive leukoencephalopathy, multiple lacunar infarcts, and osteoarthritis and is caused by mutations in the serine protease HTRA1, which resembles the pathology of severe sporadic VaD. Mutations in HTRA1 are reported to result in increased TGF-β secretion and signaling.

 3.3. Neuronal death and its relationship with white matter lesions and the extracellular matrix

The processes of endothelial dysfunction, inflammation, and BBB disruption culminate in neuronal damage and death, significantly influencing VaD pathogenesis. Inflammatory cytokines released by activated microglia and astrocytes directly injure neurons. MMP-2 and MMP-3 damage oligodendrocytes, disrupting neuronal nutritional pathways, while oxidative stress impairs oligodendrocyte production. These processes lead to myelin degradation and white matter damage, resulting in the release of myelin basic protein (MBP) and neurofilament light chain (NfL)⁴²⁾. The extracellular matrix remodeling caused by these processes contributes to vascular wall fibrosis and cellular edema, leading to the WMH observed on MRI.

VaD-associated WMH often overlaps with CAA, which can worsen dementia and cause non-hypertensive intracerebral hemorrhages. Notably, anti-amyloid antibody therapies for AD are associated with side effects such as ARIA, including edema and hemorrhages linked to CAA. Given the difficulty in distinguishing CAA from VaD in a brain MRI, detecting CAA through fluid biomarkers could significantly increase the safety of anti-amyloid antibody therapies. Recent proteomic analyses of brain tissue from patients with AD have identified proteins such as semaphorin 3G (SEMA3G) as potential CAA-specific markers⁴³⁾, paving the way for their application in clinical practice.

Extracellular matrix changes associated with myelin degradation and white matter damage may also serve as fluid biomarkers for VaD. These pathological changes are known to be evident in patients with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL). CADASIL is characterized by progressive white matter lesions observed on MRI, recurrent subcortical infarcts, depressive symptoms, and VCI. A diagnosis is confirmed through genetic or pathological identification of NOTCH3 mutations; however, some patients with the mutation may not present imaging abnormalities, emphasizing the need for a fluid biomarker to detect them. Plasma or serum NfL levels are reportedly elevated in patients with CADASIL, compared to healthy controls, and correlate with disease severity and progression⁴⁴⁾. Additionally, proteomic analyses of human cerebral arteries in patients with CADASIL have identified significant increases in extracellular matrix proteins, including collagen 1α2, laminin γ1, and clusterin⁴⁵⁾. Biomarker research in genetic models like CARASIL and CADASIL provides insights into VaD pathophysiology and may guide new therapeutic approaches.

 3.4. Perspective

The need for fluid biomarkers to distinguish VaD from other dementias has become increasingly urgent with the approval of anti-amyloid antibody therapies and their associated risks, such as ARIA. Given that individual marker proteins are not sensitive or specifically high enough to efficiently detect VaD, combining multiple proteins and molecules is required to develop a useful biomarker. Moreover, considering the heterogeneity of VCID, it is essential to evaluate biomarkers tailored to different subtypes and stages of the disease. Research on fluid biomarkers for VaD will also contribute significantly to the overall understanding and treatment of dementia.

 4. Digital biomarkers for VaD

Digital biomarkers are emerging as a novel tool for the diagnosis and monitoring of diseases and have attracted increasing attention in recent years; however, what exactly is a digital biomarker? The term “digital” originates from “digit,” which simply refers to information represented as numbers. Does “digital,” in the context of digital biomarkers, merely carry this numerical implication? Srikanth et al. provided a more precise definition: “A digital biomarker is a characteristic or set of characteristics, collected through digital health technologies (DHT), that is measured as an indicator of normal biological processes, a response to an exposure, or the effects of an intervention, including therapeutic interventions⁴⁶⁾.” This definition highlights two essential features of digital biomarkers: they are captured through digital health technologies, and their states can be altered by disease progression or interventions such as therapeutic treatments.

In recent years, advancements in smart devices have made it possible to digitally record and quantify various types of biometric data, such as electroencephalogram (EEG), eye movement, facial expressions, voices, and body movements, which were previously measured using traditional methods. Furthermore, with the progress of artificial intelligence (AI), it has become possible to predict disease onset and assess the features of symptoms by integrating information obtained in regular clinical practices. The following discussion focuses on digital biomarkers used in the evaluation of cerebrovascular disorders and VaD.

Cognitive impairment and various focal neurological symptoms are well-known manifestations of stroke⁴⁷⁾. The symptoms and etiologies of VaD are heterogeneous, necessitating comprehensive evaluations from multiple perspectives. Digital biomarkers offer the potential to predict stroke onset and assess neurological deficits resulting from cerebrovascular disorders. Digital cognitive assessment may enable the quantitative evaluation of multiple cognitive domains that facilitate the differentiation of underlying causes of cognitive dysfunction⁴⁸⁾. Below, we explore the potential of digital biomarkers for evaluating VaD.

 4.1. Assessment of dysarthria and aphasia using speech data

Dysarthria and aphasia are common manifestations in VaD, depending on the location of brain damage. The objective evaluation of these symptoms through speech data analysis is a valuable approach. The effectiveness of various methods, including audio-, image-, video-, and text-based techniques, has been reported in the literature⁴⁹⁾. The use of digitized speech data enables more precise and objective evaluations of dysarthria and aphasia, facilitating earlier detection and timely interventions.

 4.2. Cognitive assessment using eye-tracking technology

Several studies have reported methods for evaluating brain function using eye-tracking technology. We have been developing an eye-tracking-based cognitive assessment (ETCA) as a rapid screening test for dementia^50,51). In the ETCA, task-based videos are presented to participants while their gaze plot data are recorded using eye-tracking technology. Based on this gaze data, the patient’s cognitive function is scored by calculating their fixation rate on areas of interest within the task video. This approach addresses several challenges of traditional neuropsychological tests, such as reducing inter-rater variability and alleviating the patient’s test-related burden. Additionally, the ETCA provides subscores on each cognitive domain, which is potentially useful in the differential diagnosis of vascular and other types of dementia.

 4.3. Predicting stroke symptoms and cognitive decline through facial analysis

Advancements in digital technology have enabled scientists to analyze facial expressions objectively, and this is increasingly applied to dementia assessment. Facial aging has long been recognized as a general indicator of overall health⁵²⁾, and recent studies have highlighted the potential of facial features to serve as biomarkers for diagnosing dementia. Umeda-Kameyama et al. reported that AI-based facial analyses could accurately distinguish healthy individuals from patients with dementia, achieving a sensitivity of 87.31% and a specificity of 94.57%⁵³⁾. Similarly, studies have shown that machine learning using facial photographs can efficiently identify conditions such as facial nerve paralysis caused by stroke or Bell’s palsy⁵⁴⁾. These findings underscore the potential of facial analysis as a useful tool for detecting neurological conditions and cognitive decline.

 4.4. Behavioral patterns as predictors of cognitive function

The assessment of functional impairment in daily life is essential for diagnosing dementia. The objective measurement of a patient’s behavior at home using digital devices may, therefore, provide valuable information for diagnostic purposes. Information from family members or caregivers about the patient’s behavior in the home environment is critical for diagnosis, yet subjective information can often introduce diagnostic uncertainty.

Recent studies have suggested that monitoring behavioral patterns at home with wearable devices may help detect individuals with dementia⁵⁵⁾. For instance, a decline in the frequency of computer use has been observed in patients with mild cognitive impairment (MCI). Additionally, behavioral and psychological symptoms of dementia (BPSD), such as agitation or apathy, could potentially be quantified through changes in movement frequency⁵⁶⁾. However, wearable devices requiring physical contact may lead to lower adherence rates among older adults with cognitive decline, leading to incomplete data collection⁵⁷⁾.

Given these challenges, selecting appropriate devices to collect biometric data is crucial. If an individual’s behavioral patterns can be measured objectively, it will facilitate more accurate cognitive assessments outside of clinical settings.

 4.5. Preventing cerebrovascular disorders with digital biomarkers

The use of digital technologies before the onset of cerebrovascular disorders may be helpful in stroke prevention. Remote monitoring of the heart rate using a wearable device has been shown to detect atrial fibrillation, which can reduce the risk of stroke onset⁵⁸⁾. Additionally, AI models have demonstrated their effectiveness in predicting post-stroke cognitive decline⁵⁹⁾. These technologies hold promise in predicting and preventing the onset of VaD.

 4.6. Perspective

The development of various modalities has made it possible to obtain previously inaccessible features in clinical practice. While the use of new biomarkers and AI-driven diagnostics has become a topic of interest, challenges remain in the healthcare field, particularly concerning privacy, data ownership, and cost⁶⁰⁾. As digital technologies continue to evolve, efforts will be necessary to address these issues and facilitate the widespread adoption of these new technologies.

 5. Conclusion

Here, we reviewed the recent progress made in the biomarker development for VaD, with a particular focus on imaging, biofluid, and digital biomarkers (Fig. 1). Findings from basic research on the pathogenesis of VaD have revealed multifactorial aspects of the disease, providing a new avenue for biomarker development. Biomarkers associated with the molecular mechanisms of the disease may enable the detection of vascular dementia at the earliest stage.

Fig. 1  Emerging biomarkers in vascular dementia

Recent progress in VaD research has uncovered the molecular bases of VaD and provided a new avenue for imaging- and biofluid-biomarker developments. AI-based digital technologies have emerged as a potential solution for enhancing the diagnosis of VaD.

CSF, cerebrospinal fluid; VaD, vascular dementia

Acknowledgements

This work was supported by JSPS KAKENHI grant 21H02828 (Grant-in-Aid for Scientific Research (B)) (S.T.) and a research grant from the Cell Science Research Foundation (S.T.).

References

• 1)   O’Brien  JT,  Erkinjuntti  T,  Reisberg  B, et al. Vascular cognitive impairment. Lancet Neurol 2003; 2: 89–98. doi: 10.1016/s1474-4422(03)00305-3.
• 2)   Román  GC,  Tatemichi  TK,  Erkinjuntti  T, et al. Vascular dementia: diagnostic criteria for research studies. Report of the NINDS-AIREN International Workshop. Neurology 1993; 43: 250–260. doi: 10.1212/wnl.43.2.250.
• 3)   Skrobot  OA,  O’Brien  J,  Black  S, et al. The Vascular Impairment of Cognition Classification Consensus Study. Alzheimers Dement 2017; 13: 624–633. doi: 10.1016/j.jalz.2016.10.007.
• 4)   Barbay  M,  Taillia  H,  Nédélec-Ciceri  C, et al. Prevalence of poststroke neurocognitive disorders using National Institute of Neurological Disorders and stroke-Canadian Stroke Network, VASCOG criteria (Vascular Behavioral and Cognitive Disorders), and optimized criteria of cognitive deficit. Stroke 2018; 49: 1141–1147. doi: 10.1161/STROKEAHA.117.018889.
• 5)   Weaver  NA,  Kuijf  HJ,  Aben  HP, et al. Strategic infarct locations for post-stroke cognitive impairment: a pooled analysis of individual patient data from 12 acute ischaemic stroke cohorts. Lancet Neurol 2021; 20: 448–459. doi: 10.1016/S1474-4422(21)00060-0.
• 6)   Sigurdsson  S,  Aspelund  T,  Kjartansson  O, et al. Incidence of brain infarcts, cognitive change, and risk of dementia in the general population: The AGES-Reykjavik study (age gene/environment susceptibility-Reykjavik study). Stroke 2017; 48: 2353–2360. doi: 10.1161/STROKEAHA.117.017357.
• 7)   Hilal  S,  Sikking  E,  Shaik  MA, et al. Cortical cerebral microinfarcts on 3T MRI: a novel marker of cerebrovascular disease. Neurol 2016; 87: 1583–1590. doi: 10.1212/WNL.0000000000003110.
• 8)   Duering  M,  Biessels  GJ,  Brodtmann  A, et al. Neuroimaging standards for research into small vessel disease-advances since 2013. Lancet Neurol 2023; 22: 602–618. doi: 10.1016/S1474-4422(23)00131-X.
• 9)   Debette  S,  Schilling  S,  Duperron  MG, et al. Clinical significance of magnetic resonance imaging markers of vascular brain injury: a systematic review and meta-analysis. JAMA Neurol 2019; 76: 81–94. doi: 10.1001/jamaneurol.2018.3122.
• 10)   Wardlaw  JM,  Smith  C,  Dichgans  M. Small vessel disease: mechanisms and clinical implications. Lancet Neurol 2019; 18: 684–696. doi: 10.1016/S1474-4422(19)30079-1.
• 11)   Bolandzadeh  N,  Davis  JC,  Tam  R, et al. The association between cognitive function and white matter lesion location in older adults: a systematic review. BMC Neurol 2012; 12: 126. doi: 10.1186/1471-2377-12-126.
• 12)   de Groot  JC,  de Leeuw  FE,  Oudkerk  M, et al. Cerebral white matter lesions and cognitive function: The Rotterdam scan study. Ann Neurol 2000; 47: 145–151. doi: 10.1002/1531-8249(200002)47:2<145::aid-ana3>3.3.co;2-g.
• 13)   Balakrishnan  R,  Valdés Hernández  MDC,  Farrall  AJ. Automatic segmentation of white matter hyperintensities from brain magnetic resonance images in the era of deep learning and big data—a systematic review. Comput Med Imaging Graph 2021; 88: 101867. doi: 10.1016/j.compmedimag.2021.101867.
• 14)   Wardlaw  JM,  Benveniste  H,  Nedergaard  M, et al. Perivascular spaces in the brain: anatomy, physiology and pathology. Nat Rev Neurol 2020; 16: 137–153. doi: 10.1038/s41582-020-0312-z.
• 15)   Iliff  JJ,  Wang  M,  Liao  Y, et al. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci Transl Med 2012; 4: 147ra111. doi: 10.1126/scitranslmed.3003748.
• 16)   Saito  S,  Yamamoto  Y,  Ihara  M. Mild cognitive impairment: at the crossroad of neurodegeneration and vascular dysfunction. Curr Alzheimer Res 2015; 12: 507–512. doi: 10.2174/1567205012666150530202508.
• 17)   Saito  S,  Ihara  M. Interaction between cerebrovascular disease and Alzheimer pathology. Curr Opin Psychiatry 2016; 29: 168–173. doi: 10.1097/yco.0000000000000239.
• 18)   van Veluw  SJ,  Scherlek  AA,  Freeze  WM, et al. Different microvascular alterations underlie microbleeds and microinfarcts. Ann Neurol 2019; 86: 279–292. doi: 10.1002/ana.25512.
• 19)   Charidimou  A,  Boulouis  G,  Frosch  MP, et al. The Boston criteria version 2.0 for cerebral amyloid angiopathy: a multicentre, retrospective, MRI-neuropathology diagnostic accuracy study. Lancet Neurol 2022; 21: 714–725. doi: 10.1016/S1474-4422(22)00208-3.
• 20)   Biesbroek  JM,  Weaver  NA,  Aben  HP, et al. Network impact score is an independent predictor of post-stroke cognitive impairment: a multicenter cohort study in 2341 patients with acute ischemic stroke. NeuroImage Clin 2022; 34, 103018. doi: 10.1016/j.nicl.2022.103018.
• 21)   Taoka  T,  Masutani  Y,  Kawai  H, et al. Evaluation of glymphatic system activity with the diffusion MR technique: diffusion tensor image analysis along the perivascular space (DTI-ALPS) in Alzheimer’s disease cases. Jpn J Radiol 2017; 35: 172–178. doi: 10.1007/s11604-017-0617-z.
• 22)   Tang  J,  Zhang  M,  Liu  N, et al. The association between glymphatic system dysfunction and cognitive impairment in cerebral small vessel disease. Front Aging Neurosci 2022; 14: 916633. doi: 10.3389/fnagi.2022.916633.
• 23)   Frantellizzi  V,  Conte  M,  De Vincentis  G. Hybrid imaging of vascular cognitive impairment. Semin Nucl Med 2021; 51: 286–295. doi: 10.1053/j.semnuclmed.2020.12.006.
• 24)   Sweeney  MD,  Sagare  AP,  Zlokovic  BV. Blood-brain barrier breakdown in Alzheimer disease and other neurodegenerative disorders. Nat Rev Neurol 2018; 14: 133–150. doi: 10.1038/nrneurol.2017.188.
• 25)   Taheri  S,  Gasparovic  C,  Huisa  BN, et al. Blood-brain barrier permeability abnormalities in vascular cognitive impairment. Stroke 2011; 42, 2158–2163. doi: 10.1161/strokeaha.110.611731.
• 26)   Uchida  Y,  Kan  H,  Sakurai  K, et al. APOE ɛ4 dose associates with increased brain iron and β-amyloid via blood-brain barrier dysfunction. J Neurol Neurosurg Psychiatry 2022; 93: 772–778. doi: 10.1136/jnnp-2021-328519.
• 27)   Iadecola  C. The neurovascular unit coming of age: a journey through neurovascular coupling in health and disease. Neuron 2017; 96: 17–42. doi: 10.1016/j.neuron.2017.07.030.
• 28)   Schaeffer  S,  Iadecola  C. Revisiting the neurovascular unit. Nat Neurosci 2021; 24: 1198–1209. doi: 10.1038/s41593-021-00904-7.
• 29)   Keren-Shaul  H,  Spinrad  A,  Weiner  A, et al. A unique microglia type associated with restricting development of Alzheimer’s disease. Cell 2017; 169: 1276–1290.e1217. doi: 10.1016/j.cell.2017.05.018.
• 30)   Habib  N,  McCabe  C,  Medina  S, et al. Disease-associated astrocytes in Alzheimer’s disease and aging. Nat Neurosci 2020; 23: 701–706. doi: 10.1038/s41593-020-0624-8.
• 31)   Kenigsbuch  M,  Bost  P,  Halevi  S, et al. A shared disease-associated oligodendrocyte signature among multiple CNS pathologies. Nat Neurosci 2022; 25: 876–886. doi: 10.1038/s41593-022-01104-7.
• 32)   Saito  S,  Ikeda  Y,  Ando  D, et al. Cerebral amyloid angiopathy presenting as massive subarachnoid haemorrhage: a case study and review of literature. Front Aging Neurosci 2020; 12: 538456. doi: 10.3389/fnagi.2020.538456.
• 33)   Hattori  Y,  Maki  T,  Saito  S, et al. Influence of low-dose aspirin on cerebral amyloid angiopathy in mice. J Alzheimers Dis 2016; 52: 1037–1045. doi: 10.3233/jad-160013.
• 34)   Salloway  S,  Chalkias  S,  Barkhof  F, et al. Amyloid-related imaging abnormalities in 2 phase 3 studies evaluating aducanumab in patients with early Alzheimer disease. JAMA Neurol 2022; 79: 13–21. doi: 10.1001/jamaneurol.2021.4161.
• 35)   Zhang  W,  Zhou  Y,  Wang  J, et al. Glymphatic clearance function in patients with cerebral small vessel disease. Neuroimage 2021; 238: 118257. doi: 10.1016/j.neuroimage.2021.118257.
• 36)   Hosoki  S,  Hansra  GK,  Jayasena  T, et al. Molecular biomarkers for vascular cognitive impairment and dementia. Nat Rev Neurol 2023; 19: 737–753. doi: 10.1038/s41582-023-00884-1.
• 37)   Skillbäck  T,  Delsing  L,  Synnergren  J, et al. CSF/serum albumin ratio in dementias: a cross-sectional study on 1861 patients. Neurobiol Aging 2017; 59: 1–9. doi: 10.1016/j.neurobiolaging.2017.06.028.
• 38)   Kettunen  P,  Bjerke  M,  Eckerström  C, et al. Blood-brain barrier dysfunction and reduced cerebrospinal fluid levels of soluble amyloid precursor protein-β in patients with subcortical small-vessel disease. Alzheimers Dement 2022; 14: e12296. doi: 10.1002/dad2.12296.
• 39)   Candelario-Jalil  E,  Thompson  J,  Taheri  S, et al. Matrix metalloproteinases are associated with increased blood–brain barrier opening in vascular cognitive impairment. Stroke 42, 1345–1350. doi: 10.1161/strokeaha.110.600825.
• 40)   Erhardt  EB,  Adair  JC,  Knoefel  JE, et al. Inflammatory biomarkers aid in diagnosis of dementia. Front Aging Neurosci 2021; 13: 717344. doi: 10.3389/fnagi.2021.717344.
• 41)   Yamamoto  Y,  Craggs  L,  Baumann  M, et al. Review: molecular genetics and pathology of hereditary small vessel diseases of the brain. Neuropathol Appl Neurobiol 2011; 37: 94–113. doi: 10.1111/j.1365-2990.2010.01147.x.
• 42)   Khalil  M,  Teunissen  CE,  Otto  M, et al. Neurofilaments as biomarkers in neurological disorders. Nat Rev Neurol 2018; 14: 577–589. doi: 10.1038/s41582-018-0058-z.
• 43)   Leitner  D,  Kavanagh  T,  Kanshin  E, et al. Differences in the cerebral amyloid angiopathy proteome in Alzheimer’s disease and mild cognitive impairment. Acta Neuropathol 2024; 148: 9. doi: 10.1007/s00401-024-02767-1.
• 44)   Chen  CH,  Cheng  YW,  Chen  YF, et al. Plasma neurofilament light chain and glial fibrillary acidic protein predict stroke in CADASIL. J Neuroinflammation 2020; 17: 124. doi: 10.1186/s12974-020-01813-5.
• 45)   Yamamoto  Y,  Liao  YC,  Lee  YC, et al. Update on the epidemiology, pathogenesis, and biomarkers of cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy. J Clin Neurol 2023; 19: 12. doi: 10.3988/jcn.2023.19.1.12.
• 46)   Vasudevan  S,  Saha  A,  Tarver  ME, et al. Digital biomarkers: convergence of digital health technologies and biomarkers. NPJ Digit Med 2022; 5: 36. doi: 10.1038/s41746-022-00583-z.
• 47)   Hilkens  NA,  Casolla  B,  Leung  TW, et al. Stroke. Lancet 2024; 403: 2820–2836. doi: 10.1016/s0140-6736(24)00642-1.
• 48)   Looi  JC,  Sachdev  PS. Differentiation of vascular dementia from AD on neuropsychological tests. Neurology 1999; 53: 670–670, doi: 10.1212/wnl.53.4.670.
• 49)   Al-Ali  A,  Al-Maadeed  S,  Saleh  M, et al. The detection of dysarthria severity levels using AI models: a review. IEEE Access 2024; 12: 48223–48238. doi: 10.1109/access.2024.3382574.
• 50)   Oyama  A,  Takeda  S,  Ito  Y, et al. Novel method for rapid assessment of cognitive impairment using high-performance eye-tracking technology. Sci Rep 2019; 9: 12932. doi: 10.1038/s41598-019-49275-x.
• 51)   Katsuhisa  M,  Oyama  A,  Ito  Y, et al. Eye-tracking-based cognitive assessment efficiently detects mild cognitive decline in the predementia stage. Dement Geriatr Cogn Disord 2024; 54: 29–39. doi: 10.1159/000541235.
• 52)   Gunn  DA,  Rexbye  H,  Griffiths  CE, et al. Why some women look young for their age. PLoS One 2009; 4: e8021. doi: 10.1371/journal.pone.0008021.
• 53)   Umeda-Kameyama  Y,  Kameyama  M,  Tanaka  T, et al. Screening of Alzheimer’s disease by facial complexion using artificial intelligence. Aging 2021; 13: 1765–1772. doi: 10.18632/aging.202545.
• 54)   Umirzakova  S,  Ahmad  S,  Mardieva  S, et al. Deep learning-driven diagnosis: a multi-task approach for segmenting stroke and Bell’s palsy. Pattern Recognit 2023; 144: 109866. doi: https://doi.org/10.1016/j.patcog.2023.109866.
• 55)   Cote  AC,  Phelps  RJ,  Kabiri  NS, et al. Evaluation of wearable technology in dementia: a systematic review and meta-analysis. Front Med (Lausanne) 2020; 7: 501104. doi: 10.3389/fmed.2020.501104.
• 56)   Au-Yeung  WM,  Miller  L,  Beattie  Z, et al. Monitoring behaviors of patients with late-stage dementia using passive environmental sensing approaches: a case series. Am J Geriatr Psychiatry 2022; 30: 1–11. doi: 10.1016/j.jagp.2021.04.008.
• 57)   Paolillo  EW,  Lee  SY,  VandeBunte  A, et al. Wearable use in an observational study among older adults: adherence, feasibility, and effects of clinicodemographic factors. Front Digit Health 2022; 4: 884208. doi: 10.3389/fdgth.2022.884208.
• 58)   Jang  JP,  Lin  HT,  Chen  YJ, et al. Role of remote monitoring in detection of atrial arrhythmia, stroke reduction, and use of anticoagulation therapy—A systematic review and meta-analysis. Circ J 2020; 84, 1922–1930. doi: 10.1253/circj.CJ-20-0633.
• 59)   Hasan  F,  Muhtar  MS,  Wu  D, et al. Web-based artificial intelligence to predict cognitive impairment following stroke: a multicenter study. J Stroke Cerebrovasc Dis 2024; 33: 107826. doi: 10.1016/j.jstrokecerebrovasdis.2024.107826.
• 60)   Mitchell  M,  Kan  L. Digital technology and the future of health systems. Health Syst Reform 2019; 5: 113–120. doi: 10.1080/23288604.2019.1583040.