Biomarkers and Risk Assessment of Alzheimer’s Disease in Low- and Middle-Income Countries

Abstract

Dementia is a chronic syndrome which is common among the elderly and is associated with significant morbidity and mortality for patients and their caregivers. Alzheimer’s disease (AD), the most common form of clinical dementia, is biologically characterized by the deposition of amyloid-β plaques and neurofibrillary tangles in the brain. The onset of AD begins decades before manifestation of symptoms and clinical diagnosis, underlining the need to shift from clinical diagnosis of AD to a more objective diagnosis using biomarkers. Having performed a literature search of original articles and reviews on PubMed and Google Scholar, we present this review detailing the existing biomarkers and risk assessment tools for AD. The prevalence of dementia in low- and middle-income countries (LMICs) is predicted to increase over the next couple of years. Thus, we aimed to identify potential biomarkers that may be appropriate for use in LMICs, considering the following factors: sensitivity, specificity, invasiveness, and affordability of the biomarkers. We also explored risk assessment tools and the potential use of artificial intelligence/machine learning solutions for diagnosing, assessing risks, and monitoring the progression of AD in low-resource settings. Routine use of AD biomarkers has yet to gain sufficient ground in clinical settings. Therefore, clinical diagnosis of AD will remain the mainstay in LMICs for the foreseeable future. Efforts should be made towards the development of low-cost, easily administered risk assessment tools to identify individuals who are at risk of AD in the population. We recommend that stakeholders invest in education, research and development targeted towards effective risk assessment and management.

Keywords

Alzheimer’s disease artificial intelligence biomarkers blood cerebrospinal fluid low- and middle-income countries neuroimaging risk

INTRODUCTION

Dementia is a chronic progressive syndrome characterized by cognitive decline which impacts one’s ability to maintain activities of daily living beyond what is expected for normal aging [1 –3]. It causes impairment in memory, orientation, comprehension, calculation, learning capacity, language, and judgment while sparing consciousness [4]. The global prevalence of dementia is currently about 50 million [5] and will triple by the year 2050 [6]. Low- and middle-income countries (LMICs) currently account for about 60% of the disease burden and are projected to contribute 68% by the year 2050 [5 , 8].

There are wide misconceptions and poor understanding of dementia in many places in LMICs. For instance, dementia is often confused with the normal aging process; some cultures also attribute the illness to witchcraft among other superstitious beliefs which may lead to the abuse of people living with dementia [9 –11].

ALZHEIMER’S DISEASE: EPIDEMIOLOGY AND PATHOGENESIS

Alzheimer’s disease (AD) is the commonest type of dementia [12] affecting more than 27 million people globally and contributing to about 60–70% of dementia cases [13]. The clinical course of AD runs through progressive loss of episodic memory, language and visuospatial impairment, and behavioral disorders [4]. It also results in significant distress and financial strain among caregivers and family members. Among many other factors that contribute to the burden of psychological morbidities among caregivers [14 –17], persons living with dementia can become quite aggressive or apathetic towards family members [18].

Although the pathogenesis of AD is not fully understood, the amyloid cascade hypothesis is the most popular of the theories of AD. It is clear that the disease is associated with accumulations of cytoplasmic neurofibrillary tangles of phosphorylated tau (p-tau) and extracellular amyloid plaques in the brain. The mechanisms of neurocognitive decline, however, are yet to be fully elucidated [19].

AD and other related dementias currently constitute the seventh largest contributor to global mortality [20] and are expected to affect up to 75 million in 2030 and 132 million by the year 2050 [21]. LMICs are projected to experience more growth in the population of senior citizens relative to the high-income countries (HICs) [22], thus should begin to pay more attention to the growing burden of AD for which aging is a major risk factor [1 , 20].

THE NEED FOR A PARADIGM SHIFT IN THE DIAGNOSIS OF ALZHEIMER’S DISEASE

Neurodegenerative processes in AD begin more than 10-20 years before the manifestation of symptoms [23, 24]. Detecting this earlier on in the course of the illness would be highly beneficial in altering the course of the disease. In addition, this could help people living with AD take important decisions a long while before cognitive decline sets in [25]. Based on the currently available evidence, AD is incurable and has an irreversible progression which inevitably culminates in the death of the patient. Monoclonal antibodies like aducanumab and lecanemab are likely candidates for disease-modifying therapy in AD [26]. While there are still concerns about the efficacy and safety of these drugs, their availability in LMICs in the foreseeable future seems yet unlikely. This raises ethical questions as to whether early diagnosis really empowers people to make informed decisions or rather predisposes them to undue psychological distress. There, however, seems to be a consensus that the benefits of revealing the risk of AD to a cognitively unimpaired individual outweigh the potential risk of psychological disturbance or self-harm [27, 28] as this is expected to guide individuals and their families in making important decisions about their final years while still healthy. Such decisions may range from health— adopting lifestyle and health behavior modifications— to legal and financial decisions [25, 29]. These include writing a will; making decisions about power of attorney ever before the impairment of cognition; deciding who their primary caregiver would be and stating ahead of time what decisions they would consider to be in their interest when incapacitated by the disease.

Considering that the onset of the AD pathology long precedes clinical symptoms, there is need for a paradigm shift from clinical diagnosis to more objective methods of diagnosis which are capable to detecting the process much earlier. Clinical diagnosis is often made using the National Institute of Neurological and Communicative Disorders and Stroke and the AD and Related Disorders Association (NINCDS/ADRDA) criteria and a battery of neuropsychological tests [30, 31]. In 2018, the National Institute on Aging and Alzheimer’s Association (NIA-AA) research framework made changes to their diagnostic criteria of AD moving from a clinical to biological diagnosis. The new criteria recognize the asymptomatic stages of AD, during which period significant brain degeneration would have occurred. The new criteria also present the progression of AD as a continuum rather than a disease with three categories, cognitively unimpaired, mild cognitive impairment (MCI), and dementia, as seen under the clinical definition [23, 32]. This updated diagnostic algorithm calls for the use of biomarkers to arrive at diagnosis objectively and follow the disease progression [23]. Atkinson et al. describe a biomarker as a “quantifiable characteristic of biological or pathogenic processes or pharmacological response to a therapeutic intervention” [32, 33]. According to the US Food and Drug Administration (FDA), biomarkers should be specific, sensitive, predictive, robust, simple, accurate, inexpensive, and obtainable from standard biological samples like blood and urine [34].

It is estimated that by the year 2050, the prevalence of AD in LMICs will be more than twice that of HICs [5]. Proper risk assessment and early diagnosis will be important game changers in order to forestall these grim predictions, especially in resource-poor settings. Moreover, significant representation of underserved populations in clinical trials is quite germane for developing equitable therapeutic solutions for the disease. A wrong diagnosis resulting from a lack of appropriate biomarkers will hamper the outcome of clinical trials. It was once reported that about 50% of those clinically diagnosed with MCI and 25% of those diagnosed with mild AD had no amyloid deposits [32]. The importance of correctly assigning study participants to the category they belong to cannot be overemphasized during clinical trials as wrong labels during trials will likely translate into ineffective medications. It therefore becomes apparent that the benefit of using biomarkers in the diagnosis AD reaches everyone including patients, clinicians, and researchers.

BIOMARKERS OF ALZHEIMER’S DISEASE

According to the ATN biomarker classification system, validated biomarkers for AD can be grouped into three categories namely: amyloid aggregates (A) like cerebrospinal fluid (CSF) amyloid-β (Aβ)₄₂, Aβ₄₂/Aβ₄₀ ratio and amyloid positron emission tomography (PET); tau aggregates (T) like tau-PET; and markers of neurodegeneration (N) including the neurofilament light polypeptide (NF-L), magnetic resonance imaging (MRI) of the brain and ¹⁸F-fluorodeoxyglucose-positron emission tomography (FDG-PET) [23 , 35].

Imaging biomarkers and associated costs

The above listed biomarkers— except brain MRI, FDG-PET, amyloid PET, and tau PET which are imaging tests— are obtained by assaying the CSF. The T1-weighted MRI sequence which shows good contrast between gray and white matter is capable of detecting neurodegeneration, albeit at later stages of the disease. Brain MRI can also reveal atrophy of the frontal lobe, hippocampus, and entorhinal cortex [36 –38]. Unfortunately, MRI is not readily available in resource-poor countries. A study in West Africa found that only 84 MRI units served a combined population of about 370 million. In fact, none of the MRIs in the entire sub-region had up to a strength of 3 Tesla [39]. Also, the relatively high cost associated with taking a brain MRI in the developing world, where there are millions of people without Universal Health Coverage (UHC) and incur significant out-of-pocket health costs, presents a major challenge for this population [40].

PET is a nuclear imaging investigation. Whereas MRI measures gross anatomical changes, FDG-PET detects neuronal injury when there is impairment of brain glucose metabolism, which is a usual phenomenon that precedes structural changes. Amyloid-PET and tau-PET image the deposition of amyloid and tau respectively in the brain. The problems with the routine use of MRI in the clinic in poor-resource settings also apply to FDG-PET, amyloid-PET, tau-PET [32], and single-photon emission computerized tomography (SPECT) [41]. They are largely unavailable, even in tertiary health centers in LMICs, and come at a high financial cost for patients and their families. For instance, an amyloid-PET scan which costs about US$4000 [42] would be totally unaffordable for most people in LMICs.

Cerebrospinal fluid biomarkers

CSF is the clear, colorless extracellular fluid that bathes the brain and spinal cord. It can be obtained via lumbar puncture, a common and well-tolerated procedure with minimal side effects [43]. Since CSF is in contact with the external milieu of neurons and glial cells, we are able to approximate ongoing events in the brain by assaying certain molecules in the CSF. CSF amyloid and tau biomarkers are less expensive than imaging studies and appear to be better suited for poor settings. CSF Aβ₄₂ abnormality can be detected before amyloid-PET can detect an abnormality, before the onset of neurodegeneration [44].

However, there are still barriers to the use of CSF biomarkers in mainstream clinical settings. Lumbar puncture is rarely done except when required for diagnosis of neurological infections and less commonly demyelinating diseases or leptomeningeal metastasis [45]. It is a procedure that requires considerable level of experience by the physician and can also be associated with some discomfort for patient before and after the procedure. A generally poor reception of lumbar puncture has been reported among patients in sub-Saharan Africa [46].

Whereas CSF biomarkers do well in distinguishing AD from cognitively intact controls, these biomarkers do not perform so well in discriminating between AD and other forms of dementia [47]. There is also the problem of low specificity (an abnormal CSF does not often confirm AD although normal CSF usually confirms its absence) as well as racial disparities in the level of certain CSF biomarkers [48]. Coupled with the fact that there are no standardized cut-offs for biomarker levels, all of these factors are limitations to the widespread deployment of CSF biomarkers in clinics [32, 49].

Potential for blood biomarkers

Blood biomarkers are generally cheaper than CSF biomarkers and do not come with the pain, discomfort, and invasiveness associated with lumbar puncture. Venipuncture is a routine time-effective bedside and laboratory procedure that requires no special preparation, making the blood a tissue of interest in the search for low-cost novel AD biomarkers [50].

Unfortunately, efforts made to get AD blood biomarkers into routine use are fraught with challenges to a greater degree than CSF biomarkers, namely poor specificity, sensitivity, precision, and reproducibility. Because the analytes of interest are present in smaller quantities in the blood than in the CSF, there is a need to improve the technology used to detect and quantify their presence. Standardized frameworks for analyzing samples and interpreting results are also non-existent [51, 52]. This is especially important for the blood tissue which is composed of a lot of other confounding analytes [32]. Should these challenges be overcome, blood biomarkers have the potential to revolutionize the biological diagnosis of AD as they present a more affordable and less-invasive approach.

Some of the blood biomarkers with diagnostic potential include plasma tau protein and NF-L, which are both markers of neurodegeneration, and clusterin, a marker of apoptosis. Recently published work by Gonzalez-Ortiz and colleagues show that the novel brain-derived tau isoform is a more specific marker of neurodegeneration than NF-L, capable of distinguishing AD from other types of dementias [53]. Saliva is another interesting source of potential fluid biomarkers, although there has not been a lot of success with that. All of the candidate biomarkers still require massive validation on a global scale and in different populations for them to be sanctioned for clinical use [32]. An interactive platform (https://www.alzforum.org/alzbiomarker/ad-vs-ctrl) based on the work of Olsson et al. (2016) presents a meta-analysis of years of research fluid biomarkers of AD. The results show that CSF biomarkers are generally more robust and have been validated in more samples than plasma biomarkers [54].

However, there is considerable progress in validation of plasma markers; certain isoforms of p-tau (p-tau-217 and p-tau-181) have been found to reasonably correlate with CSF tau, while the same could not be demonstrated for t-tau [55]. Current evidence suggests that the use of a panel of multiple biomarkers seems most feasible for diagnosis and monitoring of disease progression with a reasonable degree of certainty, as there is a no single perfect assay for that purpose. For instance, it has been shown that CSF Aβ₄₂/Aβ₄₀ ratio has a better diagnostic accuracy than a single assay of Aβ₄₀ or Aβ₄₂ [56]. It is best to research into the possible combinations of biomarkers that give the optimum results in diagnosis and use the findings to develop test kits that can be made readily available.

Considering the currently available evidence, the clinical criteria remain the best available option for routine clinical practice in resource-poor settings. LMICs need to be at the forefront increasing efforts in research, testing, and validation of biomarkers so as to prevent further disparities in health outcomes that could result in the face of a global dementiaepidemic.

IMPROVING RISK ASSESSMENT IN LOW- AND MIDDLE - INCOME COUNTRIES

The knowledge of one’s risk of developing a disease later in life can guide informed decisions about possible lifestyle or risk modifications that impact the development of the disease. Between 58% and 79% of the variability in AD phenotype can be explained by genetics [57]. Early onset AD (defined as starting before 65 years of age) is associated with genetic mutations in APP (amyloid precursor protein gene), PSEN1 (presenilin 1 gene), and PSEN2 (presenilin 2 gene). On the other hand, APOE (apolipoprotein E gene), particularly the ɛ4 allele, significantly impacts one’s risk of developing late onset AD (occurring at age 65 and above) [58, 59].

Acquired risk factors for AD include cerebrovascular disease [3]; medical conditions like diabetes, hypertension, obesity, hearing loss, depression, and dyslipidemia. Others include lack of education, previous history of smoking, alcohol misuse, traumatic brain injury; exposure to air pollution; a lifestyle of physical inactivity and social isolation [60].

Reducing or controlling vascular risk could reduce small vessel disease in AD and slow progression [61, 62]. Lifestyle choices and behaviors like smoking and sleep routine also contribute to the risk of AD. Cognitive reserve, physical activity, and healthy diet reduce the risk of AD [3]. The Northern Manhattan Study has developed a useful risk assessment tool that can be deployed in clinical settings to assess risk of late-onset AD in elderly people using some of the factors mentioned above in addition to gender, education, ethnicity, and APOE genotype in settings where this can be done [63].

The Australian National University - Alzheimer’s Disease Risk Index (ANU-ADRI) is a risk assessment tool that can be taken online (https://anuadri.anu.edu.au/index.php) based on 8 years of research and development. Respondents can generate a report which can be presented to their doctors. ANU-ADRI incorporates more risk factors to assess the risk of AD, namely, history of head injury, pesticide exposure, social engagement, and fish intake. This tool which has the advantage of been administered via self-report has been validated independently on three older cohorts. However, it has yet to be established as being predictive of AD among middle aged people [63, 64].

There are ongoing efforts to develop culturally-appropriate and language-sensitive instruments to assess cognitive status [65]. The Community Screening Instrument for Dementia (CSI-D) is well-suited for the uneducated and has been adapted for use in sub-Saharan Africa [5]. The Rowland Universal Dementia Assessment Scale (RUDAS) has been validated for screening in Arabic [66]. The Intervention for Dementia in Elderly Africans (IDEA) Cognitive Screen has also been validated among the Yorubas and Swahilis. The IDEA Cognitive Screen has no reading, writing, drawing, or calculation component, making it suitable for non-literate individuals [67 –69]. Other relevant instruments with good psychometric properties are the Alzheimer’s Disease Assessment Scale-Cognitive subscale (ADAS-Cog) [70] and the IDEA study Instrumental Activities of Daily Living (IDEA–IADL) questionnaire which has been developed into a mobile application for use in rural areas [71, 72].

However, unlike the ANU-ADRI and the tool developed by the Northern Manhattan Study, the above named tools do not predict AD risk, rather they are dementia screening tools which are nonetheless very useful instruments that could enhance access to cognitive screening and thus combat under-diagnosis of AD in LMICs [73]. Nonetheless, there remains a need for similar risk assessment tools based on large cohorts in LMICs. Azar et al. (2021) describe the importance of AD risk stratification and recommend mitigating risks by dealing with them according to the individual’s risk factors [74].

Culture, language, and level of education are important factors that affect the validity of neurocognitive tests [75]. Ancestry, specific genetic factors, play an even much bigger role in risk assessment. Genetic differences are inherent biological characteristics and not just artifacts of the risk assessment method. Anstey et al. (2014) reported a strong effect of APOE ɛ4 allele on AD risk among East Asians, but less significant risk was found in Sub-Saharan Africa [63]. However, it is still not clear how reliably can the possession of two APOE ɛ4 alleles predicts AD in various populations including sub-Saharans Africans. Comparing between African-Americans and the Yoruba population in Nigeria, Hendrie et al. (2014) found that, even though APOE ɛ4 was significantly associated with AD risk in both populations, it was to a far lesser degree in the Yoruba and other sub-Saharan African populations [76]. Similarly, Sayi et al. (1997) and Chen et al. (2010) report among East Africans the limitations of APOE ɛ4 to differentiate between people with AD and normal subjects [77, 78]. More recent evidence identified an African ancestry specific locus at 19q13.31 to be responsible for mitigating the risk due to APOE ɛ4 among people of African descent [79]. These observations highlight the urgent need for large genome-wide studies in the African and other LMIC populations to identify more reliable ancestry-specific genetic risk variants [65].

Risk factors like poverty, malnutrition, infectious diseases, and illiteracy are highly prevalent in LMICs [80]; however, the extended family structure and stronger family ties (less common in HICs) have been found to be protective for people living in LMICs [81]. There is therefore a need to explore population-specific environmental and socioeconomic factors that are relevant for dementia in LMICs and how these interact with genetic risks.

POTENTIAL FOR ARTIFICIAL INTELLIGENCE/MACHINE LEARNING SOLUTIONS

An artificial intelligence (AI) solution is capable of making predictions based on patterns learnt from training data previously fed into the system [82]. There has been a recent interest in using AI/machine learning (ML) to model AD risk and disease progression [83] using various data sources including clinical data and neuroimages [84, 85], as well as genomic data [86 –88]. In the last decade, the world witnessed a major surge in the AI industry and research [89], which was predicated on recent advances in computer power, sophisticated algorithms, and large labelled datasets [90, 91]. AI seems quite promising for screening, diagnosing, and monitoring the progression of AD. AI solutions are non-invasive and have the potential to be deployed electronically in health systems or even personal mobile devices.

The practicality of integrating AI into clinical care of AD in LMICs depends on having the right infrastructure in place. It is difficult to imagine implementing AI on clinical data in the absence of a proper electronic health records system. Since this is the situation in many health institutions in LMICs, digitization of clinical data becomes necessary for them to be a key player in the AI revolution [92, 93]. Secondly, not every institution has adequate data infrastructure and computational power to manage big data. Lastly, the body of relevant data available in LMICs for training such diagnostic AI models is yet inadequate. There is paucity of genome-wide association studies of AD in Africa save for a few candidate gene studies [65]; neuroimaging is not widely done for patients and clinical data is not well-curated. There is a risk of algorithmic bias when AI models are trained using insufficient or non-representative data [94]. Notwithstanding, ongoing work in the Caribbeans and Latin America aims to use ML to identify AD risks by combining behavioral, clinical, and imaging data [95]. And perhaps, there is some value in looking beyond clinical data. A notable example of such projects has received funding to investigate social media history as a potential biomarker of cognitive decline in Northern Africa [96, 97].

For ethical and legal reasons, it is important for health practitioners to understand how a piece of technology works before deploying it. This is seldom the case with AI which is thought to function as a “black box” such that even developers cannot always explain their algorithms. AI explainability in clinical decision-making is very important to preserving the core values and principles of medicine, thus remains an active area of interest for developers, physicians, ethicists, and policy-makers [98, 99]. AI/ML solutions are promising and definitely feasible, therefore more time, capital, and political will should be invested to scale them up for clinical utility in LMICs. The last decade (2010-2019) witnessed a record high in the adoption of data protection legislation by 62 countries, the majority being LMICs (https://unctad.org/page/data-protection-and-privacy-legislation-worldwide). Strong implementation of these laws will be vital for data sharing in the AI age [100, 101]. Moreover, FAIR and open science will also keep information on major breakthroughs in the field within the reach of professionals working in LMICs.

CONCLUSION

There are ongoing efforts to make the biological diagnosis of AD less invasive, less expensive, and appropriate for routine use in clinical settings. These efforts are, however, limited by invasiveness, poor sensitivity, specificity, precision, and lack of standardization of biomarker tests. Neuroimaging is relatively expensive. So fluid biomarkers are the cheaper alternatives, of which CSF biomarkers are currently the most robust. Plasma biomarkers, on the other hand, require standardized protocols and further validation in diverse populations in order for them to be representative.

We posit that clinical diagnosis of AD will remain the mainstay in LMICs for at least the next 5 years. It is recommended that LMICs begin to invest in research and development of low-cost, easily accessible diagnostic and risk assessment tools. The right infrastructure for AI/ML in routine diagnosis is not likely to be fully operational at least for the next 10 years as this is capital intensive and requires a lot of political will from stakeholders at all levels of care and governance. Early response in terms of research and proper education will no doubt serve the populace of the LMICs well in the fight against AD.

Footnotes

ACKNOWLEDGMENTS

The authors have no acknowledgments to report.

FUNDING

Our work is supported by previous grants from Alzheimer’s Research UK (ARUK PG2013-22) and Medical Research Council, UK (MRC, G0500247). ROA is further supported by the US National Institutes of Health (U01HG010273, U19AG074865, R01AG072547), the UK Royal Society/African Academy of Sciences (FLR/R1/191813, FCG/R1/201034), and the Alzheimer’s Association (GBHI ALZ UK-21- 24204).

CONFLICT OF INTEREST

The authors have no conflict of interests to report.

References

Sosa-Ortiz

, Acosta-Castillo

, Prince

(2012) Epidemiology of dementias and Alzheimer’s disease. Arch Med Res 43, 600–608.

Reitz

, Brayne

, Mayeux

(2011) Epidemiology of Alzheimer disease. Nat Rev Neurol 7, 137–152.

Mayeux

, Stern

(2012) Epidemiology of Alzheimer disease. Cold Spring Harb Perspect Med 2, a006239.

Bateman

, Xiong

, Benzinger

TLS

, Fagan

, Goate

, Fox

, Marcus

, Cairns

, Xie

, Blazey

, Holtzman

, Santacruz

, Buckles

, Oliver

, Moulder

, Aisen

, Ghetti

, Klunk

, McDade

, Martins

, Masters

, Mayeux

, Ringman

, Rossor

, Schofield

, Sperling

, Salloway

, Morris

(2012) Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. N Engl J Med 367, 795–804.

Prince

, Wimo

, Guerchet

, Ali

, Wu

, Prina

(2015) World Alzheimer Report 2015. The Global Impact of Dementia: An Analysis of Prevalence, Incidence, Cost and Trends Alzheimer’s Disease International, London.

GBD 2019 Dementia Forecasting Collaborators (2022) Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: An analysis for the Global Burden of Disease Study 2019. Lancet Public Health 7, e105–e125.

Prina

, Mayston

, Wu

, Prince

(2019) A review of the 10/66 dementia research group. Soc Psychiatry Psychiatr Epidemiol 54, 1–10.

Prince

(2000) Dementia in developing countries: A consensus statement from the 10/66 Dementia Research Group. Int J Geriatr Psychiatry 15, 14–20.

Agunbiade

(2019) Explanations around physical abuse, neglect and preventive strategies among older Yoruba people (60+) in urban Ibadan Southwest Nigeria: A qualitative study. Heliyon 5, e02888.

10.

Kehoua

, Dubreuil

, Ndamba-Bandzouzi

, Guerchet

, Mbelesso

, Dartigues

, Preux

(2019) People with dementia in Sub-Saharan Africa: From support to abuse by caregivers: Results of EPIDEMCA-FU program in Congo. Dement Geriatr Cogn Dis Extra 9, 163–175.

11.

Mkhonto

, Hanssen

(2018) When people with dementia are perceived as witches. Consequences for patients and nurse education in South Africa. J Clin Nurs 27, e169–e176.

12.

Ballard

, Gauthier

, Corbett

, Brayne

, Aarsland

, Jones

(2011) Alzheimer’s disease. Lancet 377, 1019–1031.

13.

Ferreira

, Perestelo-Pérez

, Westman

, Wahlund

, Sarrisa

, Serrano-Aguilar

(2014) Meta-review of CSF core biomarkers in Alzheimer’s disease: The state-of-the-art after the new revised diagnostic criteria. Front Aging Neurosci 6, 47.

14.

Campbell

, Wright

, Oyebode

, Job

, Crome

, Bentham

, Jones

, Lendon

(2008) Determinants of burden in those who care for someone with dementia. Int J Geriatr Psychiatry 23, 1078–1085.

15.

Irwin

, Mausbach

, Koo

, Fairman

, Roepke-Buehler

, Chattillion

, Dimsdale

, Patterson

, Ancoli-Israel

, Mills

, Von Känel

, Ziegler

, Grant

(2013) Association between hospice care and psychological outcomes in Alzheimer’s spousal caregivers. J Palliat Med 16, 1450–1454.

16.

Peng

, Chang

(2013) Sleep disturbance in family caregivers of individuals with dementia: A review of the literature. Perspect Psychiatr Care 49, 135–146.

17.

Lavretsky

(2005) Stress and depression in informal family caregivers of patients with Alzheimer’s disease. Aging Health 1, 117–133.

18.

Silva

MVF

, Loures

CMG

, Alves

LCV

, De Souza

, Borges

KBG

, Carvalho

MDG

(2019) Alzheimer’s disease: Risk factors and potentially protective measures. J Biomed Sci 26, 33.

19.

Kang

, Lee

(2017) Metabolism-centric overview of the pathogenesis of Alzheimer’s disease. Yonsei Med J 58, 479–488.

20.

World Health Organization. Dementia, https://www.who.int/news-room/fact-sheets/detail/dementia.

21.

World Health Organization (2017) Global action plan on the public health response to dementia 2017-2025. https://www.who.int/publications/i/item/global-action-plan-on-the-public-health-response-to-dementia-2017-2025.

22.

Shetty

(2012) Grey matter: Ageing in developing countries. Lancet 379, 1285–1287.

23.

Jack

, Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

, Elliott

, Masliah

, Ryan

, Silverberg

(2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

24.

Holtzman

, Morris

, Goate

(2011) Alzheimer’s disease: The challenge of the second century. Sci Transl Med 3, 77sr1.

25.

Nasreddine

, Garibotto

, Kyaga

, Padovani

(2022) The early diagnosis of Alzheimer’s disease: A patient-centred conversation with the care team. Neurol Ther 12, 11–23.

26.

Golde

(2022) Disease-modifying therapies for Alzheimer’s disease: More questions than answers. Neurotherapeutics 19, 209–227.

27.

Grill

, Raman

, Ernstrom

, Sultzer

, Burns

, Donohue

, Johnson

, Aisen

, Sperling

, Karlawish

(2020) Short-term psychological outcomes of disclosing amyloid imaging results to research participants who do not have cognitive impairment. JAMA Neurol 77, 1504–1513.

28.

Erickson

, Chin

, Johnson

, Gleason

, Clark

(2021) Disclosure of preclinical Alzheimer’s disease biomarker results in research and clinical settings: Why, how, and what we still need to know. Alzheimers Dement (Amst) 13, e12150.

29.

Porsteinsson

, Isaacson

, Knox

, Sabbagh

, Rubino

(2021) Diagnosis of early Alzheimer’s disease: Clinical practice in 2021. J Prev Alzheimers Dis 8, 371–386.

30.

Langa

, Plassman

, Wallace

, Herzog

, Heeringa

, Ofstedal

, Burke

, Fisher

, Fultz

, Hurd

, Potter

, Rodgers

, Steffens

, Weir

, Willis

(2005) The aging, demographics, and memory study: Study design and methods. Neuroepidemiology 25, 181–191.

31.

Rajan

, Weuve

, Barnes

, Wilson

, Evans

(2019) Prevalence and incidence of clinically diagnosed Alzheimer’s disease dementia from 1994 to 2012 in a population study. Alzheimers Dement 15, 1–7.

32.

Khoury

, Ghossoub

(2019) Diagnostic biomarkers of Alzheimer’s disease: A state-of-the-art review. Biomark Neuropsychiatry 1, 100005.

33.

Atkinson

, Colburn

, DeGruttola

, DeMets

, Downing

, Hoth

, Oates

, Peck

, Schooley

, Spilker

, Woodcock

, Zeger

(2001) Biomarkers and surrogate endpoints: Preferred definitions and conceptual framework. Clin Pharmacol Ther 69, 89–95.

34.

Karley

, Gupta

, Tiwari

(2011) Biomarker for cancer: A great promise for future. World J Oncol 2, 151–157.

35.

Jack

Jr , Bennett

, Blennow

, Carrillo

, Feldman

, Frisoni

, Hampel

, Jagust

, Johnson

, Knopman

, Petersen

, Scheltens

, Sperling

, Dubois

(2016) A/T/N: An unbiased descriptive classification scheme for Alzheimer disease biomarkers. Neurology 87, 539–547.

36.

Kehoe

, McNulty

, Mullins

, Bokde

ALW

(2014) Advances in MRI biomarkers for the diagnosis of Alzheimer’s disease. Biomark Med 8, 1151–1169.

37.

Golebiowski

, Barcikowska

, Pfeffer

(1999) Magnetic resonance imaging-based hippocampal volumetry in patients with dementia of the Alzheimer type. Dement Geriatr Cogn Disord 10, 284–288.

38.

Juottonen

, Laakso

, Partanen

, Soininen

(1999) Comparative MR analysis of the entorhinal cortex and hippocampus in diagnosing Alzheimer disease. AJNR Am J Neuroradiol 20, 139–144.

39.

Ogbole

, Adeyomoye

, Badu-Peprah

, Mensah

, Nzeh

(2018) Survey of magnetic resonance imaging availability in West Africa. Pan Afr Med J 30, 240.

40.

Lagomarsino

, Garabrant

, Adyas

, Muga

, Otoo

(2012) Moving towards universal health coverage: Health insurance reforms in nine developing countries in Africa and Asia. Lancet 380, 933–943.

41.

McNeill

, Sare

, Manoharan

, Testa

, Mann

DMA

, Neary

, Snowden

, Varma

(2007) Accuracy of single-photon emission computed tomography in differentiating frontotemporal dementia from Alzheimer’s disease. J Neurol Neurosurg Psychiatry 78, 350–355.

42.

Tanaka

, Ruifen

, Nai

, Tan

, Lim

CZJ

, Zhang

, Stephenson

, Hilal

, Saridin

, Gyanwali

, Villaraza

, Robins

, Ihara

, Schöll

, Zetterberg

, Blennow

, Ashton

, Shao

, Reilhac

, Chen

(2021) Head-to-head comparison of amplified plasmonic exosome Aβ42 platform and single-molecule array immunoassay in a memory clinic cohort. Eur J Neurol 28, 1479–1489.

43.

Alcolea

, Martínez-Lage

, Izagirre

, Clerigué

, Carmona-Iragui

, Alvarez

, Fortea

, Balasa

, Morenas-Rodríguez

, Lladó

, Grau

, Blennow

, Lleó

, Molinuevo

(2014) Feasibility of lumbar puncture in the study of cerebrospinal fluid biomarkers for Alzheimer’s disease: A multicenter study in Spain. J Alzheimers Dis 39, 719–726.

44.

Palmqvist

, Mattsson

, Hansson

(2016) Cerebrospinal fluid analysis detects cerebral amyloid-β accumulation earlier than positron emission tomography. Brain 139, 1226–1236.

45.

Straus

, Thorpe

, Holroyd-Leduc

(2006) How do I perform a lumbar puncture and analyze the results to diagnose bacterial meningitis? JAMA 296, 2012–2022.

46.

Thakur

, Mateyo

, Hachaambwa

, Kayamba

, Mallewa

, Nwazor

, Lawal

, Mallum

, Atadzhanov

, Boulware

, Birbeck

, Siddiqi

(2015) Lumbar puncture refusal in sub-Saharan Africa: A call for further understanding and intervention. Neurology 84, 1988–1990.

47.

Molinuevo

, Ayton

, Batrla

, Bednar

, Bittner

, Cummings

, Fagan

, Hampel

, Mielke

, Mikulskis

, O’Bryant

, Scheltens

, Sevigny

, Shaw

, Soares

, Tong

, Trojanowski

, Zetterberg

, Blennow

(2018) Current state of Alzheimer’s fluid biomarkers. Acta Neuropathol 136, 821–853.

48.

Howell

, Watts

, Parker

, Wu

, Kollhoff

, Wingo

, Dorbin

, Qiu

, Hu

(2017) Race modifies the relationship between cognition and Alzheimer’s disease cerebrospinal fluid biomarkers. Alzheimers Res Ther 9, 88.

49.

, Stromswold

, Wilson

, Holder

, Sur

, Laterza

, Savage

, Struyk

, Scheltens

, Teunissen

, Burke

, Macaulay

, Bråthen

, Sando

, White

, Weiss

, Cowes

, Bush

, DeSilva

, Darby

, Rainey-Smith

, Surls

, Sagini

, Tanen

, Altman

, Luthman

, Egan

(2017) A multinational study distinguishing Alzheimer’s and healthy patients using cerebrospinal fluid tau/Aβ42 cutoff with concordance to amyloid positron emission tomography imaging. Alzheimers Dement (Amst) 6, 201–209.

50.

Altuna-Azkargorta

, Mendioroz-Iriarte

(2018) Blood biomarkers in Alzheimer’s disease. Neurologia (Engl Ed) 36, 704–710.

51.

Doecke

, Laws

, Faux

, Wilson

, Burnham

, Lam

, Mondal

, Bedo

, Bush

, Brown

, De Ruyck

, Ellis

, Fowler

, Gupta

, Head

, Macaulay

, Pertile

, Rowe

, Rembach

, Rodrigues

, Rumble

, Szoeke

, Taddei

, Trounson

, Ames

, Masters

, Martins

Alzheimer’s Disease Neuroimaging Initiative; Australian Imaging Biomarker and Lifestyle Research Group (2012) Blood-based protein biomarkers for diagnosis of Alzheimer disease. JAMA Neurol 69, 1318–1325.

52.

O’Bryant

, Mielke

, Rissman

, Lista

, Vanderstichele

, Zetterberg

, Lewczuk

, Posner

, Hall

, Johnson

, Fong

, Luthman

, Jeromin

, Batrla-Utermann

, Villarreal

, Britton

, Snyder

, Henriksen

, Grammas

, Gupta

, Martins

, Hampel

(2017) Blood-based biomarkers in Alzheimer disease: Current state of the science and a novel collaborative paradigm for advancing from discovery to clinic. Alzheimers Dement 13, 45–58.

53.

Gonzalez-Ortiz

, Turton

, Kac

, Smirnov

, Premi

, Ghidoni

, Benussi

, Cantoni

, Saraceno

, Rivolta

, Ashton

, Borroni

, Galasko

, Harrison

, Zetterberg

, Blennow

, Karikari

(2012) Brain-derived tau: A novel blood-based biomarker for Alzheimer’s disease-type neurodegeneration. Brain 139, 16–17.

54.

Olsson

, Lautner

, Andreasson

, Öhrfelt

, Portelius

, Bjerke

, Hölttä

, Rosén

, Olsson

, Strobel

, Wu

, Dakin

, Petzold

, Blennow

, Zetterberg

(2016) CSF and blood biomarkers for the diagnosis of Alzheimer’s disease: A systematic review and meta-analysis. Lancet Neurol 15, 673–684.

55.

Barthélemy

, Horie

, Sato

, Bateman

(2020) Blood plasma phosphorylated-tau isoforms track CNS change in Alzheimer’s disease. J Exp Med 217, e20200861.

56.

Blennow

, Zetterberg

(2018) Biomarkers for Alzheimer’s disease: Current status and prospects for the future. J Intern Med 284, 643–663.

57.

Gatz

, Reynolds

, Fratiglioni

, Johansson

, Mortimer

, Berg

, Fiske

, Pedersen

(2006) Role of genes and environments for explaining Alzheimer disease. Arch Gen Psychiatry 63, 168–174.

58.

Giri

, Lü

, Zhang

(2016) Genes associated with Alzheimer’s disease: An overview and current status. Clin Interv Aging 11, 665–681.

59.

Cacace

, Sleegers

, Van Broeckhoven

(2016) Molecular genetics of early-onset Alzheimer’s disease revisited. Alzheimers Dement 12, 733–748.

60.

Livingston

, Huntley

, Sommerlad

, Ames

, Ballard

, Banerjee

, Brayne

, Burns

, Cohen-Mansfield

, Cooper

, Costafreda

, Dias

, Fox

, Gitlin

, Howard

, Kales

, Kivimäki

, Larson

, Ogunniyi

, Orgeta

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selbæk

, Teri

, Mukadam

(2020) Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet 396, 413–446.

61.

Sweeney

, Montagne

, Sagare

, Nation

, Schneider

, Chui

, Harrington

, Pa

, Law

, Wang

DJJ

, Jacobs

, Doubal

, Ramirez

, Black

, Nedergaard

, Benveniste

, Dichgans

, Iadecola

, Love

, Bath

, Markus

, Salman

, Allan

, Quinn

, Kalaria

, Werring

, Carare

, Touyz

, Williams

SCR

, Moskowitz

, Katusic

, Lutz

, Lazarov

, Minshall

, Rehman

, Davis

, Wellington

, Gonzélez

, Yuan

, Lockhart

, Hughes

, Chen

CLH

, Sachdev

, O’Brien

, Skoog

, Pantoni

, Gustafson

, Biessels

, Wallin

, Smith

, Mok

, Wong

, Passmore

, Barkof

, Muller

, Breteler

MMB

, Román

, Hamel

, Seshadri

, Gottesman

, van Buchem

, Arvanitakis

, Schneider

, Drewes

, Hachinski

, Finch

, Toga

, Wardlaw

, Zlokovic

(2019) Vascular dysfunction - the disregarded partner of Alzheimer’s disease. Alzheimers Dement 15, 158–167.

62.

Kalaria

, Sepulveda-Falla

(2021) Cerebral small vessel disease in sporadic and familial Alzheimer disease. Am J Pathol 191, 1888–1905.

63.

Anstey

, Eramudugolla

, Dixon

(2014) Contributions of a risk assessment approach to the prevention of Alzheimer’s disease and dementia. J Alzheimers Dis 42, S463–SS473.

64.

Andrews

, Eramudugolla

, Velez

, Cherbuin

, Easteal

, Anstey

(2017) Validating the role of the Australian National University Alzheimer’s Disease Risk Index (ANU-ADRI) and a genetic risk score in progression to cognitive impairment in a population-based cohort of older adults followed for 12 years. Alzheimers Res Ther 9, 16.

65.

Akinyemi

, Yaria

, Ojagbemi

, Guerchet

, Okubadejo

, Njamnshi

, Sarfo

, Akpalu

, Ogbole

, Ayantayo

, Adokonou

, Paddick

, Ndetei

, Bosche

, Ayele

, Damas

, Coker

, Mbakile-Mahlanza

, Ranchod

, Bobrow

, Anazodo

, Damasceno

, Seshadri

, Pericak-Vance

, Lawlor

, Miller

, Owolabi

, Baiyewu

, Walker

, Gureje

, Kalaria

, Ogunniyi A; African Dementia Consortium (AfDC) (2022) Dementia in Africa: Current evidence, knowledge gaps, and future directions. Alzheimers Dement 18, 790–809.

66.

Chaaya

, Phung

TKT

, El Asmar

, Atweh

, Ghusn

, Khoury

, Prince

, Nielsen

, Waldemar

(2015) Validation of the Arabic Rowland Universal Dementia Assessment Scale (A-RUDAS) in elderly with mild and moderate dementia. Aging Ment Health 20, 880–887.

67.

Gray

, Paddick

, Kisoli

, Dotchin

, Longdon

, Chaote

, Samuel

, Jusabani

, Walker

(2014) Development and validation of the identification and intervention for dementia in elderly Africans (IDEA) study dementia screening instrument. J Geriatr Psychiatry Neurol 27, 110–118.

68.

Gray

, Paddick

, Collingwood

, Kisoli

, Mbowe

, Mkenda

, Lissu

, Rogathi

, Kissima

, Walker

, Mushi

, Chaote

, Ogunniyi

, Dotchin

(2016) Community validation of the IDEA study cognitive screen in rural Tanzania. Int J Geriatr Psychiatry 31, 1199–1207.

69.

Paddick

, Gray

, Ogunjimi

, Lwezuala

, Olakehinde

, Kisoli

, Kissima

, Mbowe

, Mkenda

, Dotchin

, Walker

, Mushi

, Collingwood

, Ogunniyi

(2015) Validation of the Identification and Intervention for Dementia in Elderly Africans (IDEA) cognitive screen in Nigeria and Tanzania. BMC Geriatr 15, 53.

70.

Paddick

, Kisoli

, Mkenda

, Mbowe

, Gray

, Dotchin

, Ogunniyi

, Kisima

, Olakehinde

, Mushi

, Walker

(2017) Adaptation and validation of the Alzheimer’s Disease Assessment Scale - Cognitive (ADAS-Cog) in a low-literacy setting in sub-Saharan Africa. Acta Neuropsychiatr 29, 244–251.

71.

Collingwood

, Paddick

, Kisoli

, Dotchin

, Gray

, Mbowe

, Mkenda

, Urasa

, Mushi

, Chaote

, Walker

(2014) Development and community-based validation of the IDEA study Instrumental Activities of Daily Living (IDEA-IADL) questionnaire. Glob Health Action 7, 25988.

72.

Stone

, Heward

, Paddick

, Dotchin

, Walker

, Collingwood

, Thornton

, Yarwood

, McCartney

, Irwin

, Mkenda

, Kissima

, Swai

, Gray

(2018) Screening for instrumental activities of daily living in Sub-Saharan Africa: A balance between task shifting, simplicity, brevity, and training. J Geriatr Psychiatry Neurol 31, 248–255.

73.

Llibre Rodriguez

, Ferri

, Acosta

, Guerra

, Huang

, Jacob

, Krishnamoorthy

, Salas

, Sosa

, Acosta

, Dewey

, Gaona

, Jotheeswaran

, Li

, Rodriguez

, Kumar

, Valhuerdi

, Prince M; 10/66 Dementia Research Group (2008) Prevalence of dementia in Latin America, India, and China: A population-based cross-sectional survey. Lancet 372, 464–474.

74.

Azar

, Salama

, Chidambaram

, Al-Balushi

, Essa

, Qoronfleh

(2021) Precision health in Alzheimer disease: Risk assessment-based strategies. Precis Med Sci 10, 54–70.

75.

Rock

, Price

(2019) Identifying culturally acceptable cognitive tests for use in remote northern Australia. BMC Psychol 7, 62.

76.

Hendrie

, Murrell

, Baiyewu

, Lane

, Purnell

, Ogunniyi

, Unverzagt

, Hall

, Callahan

, Saykin

, Gureje

, Hake

, Foroud

, Gao

(2014) APOE ɛ4 and the risk for Alzheimer disease and cognitive decline in African Americans and Yoruba. Int Psychogeriatr 26, 977.

77.

Sayi

, Patel

, Premkumar

, Adem

, Winblad

, Matuja

, Mtui

, Gatere

, Friedland

, Koss

, Kalaria

(1997) Apolipoprotein E polymorphism in elderly east Africans. East Afr Med J 74, 668–670.

78.

Chen

, Mizuno

, Elston

, Kariuki

, Hall

, Unverzagt

, Hendrie

, Gatere

, Kioy

, Patel

, Friedland

, Kalaria

(2010) A comparative study to screen dementia and APOE genotypes in an ageing East African population. Neurobiol Aging 31, 732–740.

79.

Rajabli

, Beecham

, Hendrie

, Baiyewu

, Ogunniyi

, Gao

, Kushch

, Lipkin-Vasquez

, Hamilton-Nelson

, Young

, Dykxhoorn

, Nuytemans

, Kunkle

, Wang

, Jin

, Liu

, Feliciano-Astacio

, Schellenberg

, Dalgard

, Griswold

, Byrd

, Reitz

, Cuccaro

, Haines

, Pericak-Vance

, Vance

(2022) A locus at 19q13.31 significantly reduces the ApoE ɛ4 risk for Alzheimer’s disease in African ancestry. PLoS Genet 18, e1009977.

80.

Kalaria

, Maestre

, Arizaga

, Friedland

, Galasko

, Hall

, Luchsinger

, Ogunniyi

, Perry

, Potocnik

, Prince

, Stewart

, Wimo

, Zhang

, Antuono

(2008) Alzheimer’s disease and vascular dementia in developing countries: Prevalence, management, and risk factors. Lancet Neurol 7, 812–826.

81.

You

, Henneberg

(2022) Large household reduces dementia mortality: A cross-sectional data analysis of 183 populations. PLoS One 17, e0263309.

82.

Miller

, Brown

(2018) Artificial intelligence in medical practice: The question to the answer? Am J Med 131, 129–133.

83.

Kumar

, Oh

, Schindler

, Lai

, Payne

PRO

, Gupta

(2021) Machine learning for modeling the progression of Alzheimer disease dementia using clinical data: A systematic literature review. JAMIA Open 4, ooab052.

84.

Odusami

, Maskeliūnas

, Damaševi˘čius

, Krilavičuius

(2021) Analysis of features of alzheimer’s disease: Detection of early stage from functional brain changes in magnetic resonance images using a finetuned resnet18 network. Diagnostics (Basel) 11, 1071.

85.

Beltrán

, Wahba

, Hose

, Shasha

, Kline

; Alzheimer’s Disease Neuroimaging Initiative (2020) Inexpensive, non-invasive biomarkers predict Alzheimer transition using machine learning analysis of the Alzheimer’s Disease Neuroimaging (ADNI) database. PLoS One 15, e0235663.

86.

Kong

, Giovanello

, Wang

, Lin

, Lee

, Fan

, Murali Doraiswamy

, Zhu

(2015) Predicting Alzheimer’s disease using combined imaging-whole genome SNP data. J Alzheimers Dis 46, 695–702.

87.

Jia

, Li

, Wei

, Zhu

, Qu

, Qin

, Tang

, Shen

, Wang

, Shen

, Li

, Peng

, Tan

, Luo

, Guo

, Tang

, Du

, Zhang

, Lyu

, Li

, Zhou

, Wang

, Chu

, Song

, Wu

, Zuo

, Han

, Liang

, Wang

, Jin

, Wang

, Lü

, Li

, Zhou

, Zhang

, Liao

, Qiu

, Li

, Kong

, Li

, Jiao

, Lu

, Jia

(2021) Prediction of Alzheimer’s disease using multi-variants from a Chinese genome-wide association study. Brain 144, 924–937.

88.

Erdogan

, Aydin Son

(2014) Predicting the disease of Alzheimer with SNP biomarkers and clinical data using data mining classification approach: Decision tree. Stud Health Technol Inform 205, 511–515.

89.

Savage

(2020) The race to the top among the world’s leaders in artificial intelligence. Nature 588, S102–S104.

90.

Hosny

, Parmar

, Quackenbush

, Schwartz

, Aerts

HJWL

(2018) Artificial intelligence in radiology. Nat Rev Cancer 18, 500–510.

91.

Bohr

, Memarzadeh

(2020) The rise of artificial intelligence in healthcare applications. Artif Intell Healthc 2020, 25–60.

92.

Ahuja

(2019) The impact of artificial intelligence in medicine on the future role of the physician. PeerJ 7, e7702.

93.

Odekunle

, Odekunle

, Shankar

(2017) Why sub-Saharan Africa lags in electronic health record adoption and possible strategies to increase its adoption in this region. Int J Health Sci (Qassim) 11, 59–64.

94.

Suresh

, Guttag

(2019) A framework for understanding sources of harm throughout the machine learning life cycle. arXiv:1901.10002. https://doi.org/10.48550/arXiv.1901.10002.

95.

Ibanez

, Yokoyama

, Possin

, Matallana

, Lopera

, Nitrini

, Takada

, Custodio

, Sosa Ortiz

, Avila-Funes

, Behrens

, Slachevsky

, Myers

, Cochran

, Brusco

, Bruno

, Brucki

SMD

, Pina-Escudero

, Okada de Oliveira

, Donnelly Kehoe

, Garcia

, Cardona

, Santamaria-Garcia

, Moguilner

, Duran-Aniotz

, Tagliazucchi

, Maito

, Longoria Ibarrola

, Pintado-Caipa

, Godoy

, Bakman

, Javandel

, Kosik

, Valcour

, Miller

(2021) The Multi-Partner Consortium to Expand Dementia Research in Latin America (ReDLat): Driving multicentric research and implementation science. Front Neurol 12, 631722.

96.

Alzheimer’s Association. Funded Studies Details, https://www.alz.org/research/for_researchers/grants/funded-studies-details?FundedStudyID=2360.

97.

Global Brain Health Institute. Pilot Projects, Facebook Interaction as a Potential Marker of Cognitive Decline, https://www.gbhi.org/projects/facebook-interaction-potential-marker-cognitive-decline.

98.

Amann

, Blasimme

, Vayena

, Frey

, Madai

(2020) Explainability for artificial intelligence in healthcare: A multidisciplinary perspective. BMC Med Inform Decis Mak 20, 310.

99.

Danso

, Zeng

, Muniz-Terrera

, Ritchie

(2021) Developing an explainable machine learning-based personalised dementia risk prediction model: A transfer learning approach with ensemble learning algorithms. Front Big Data 4, 613047.

100.

Greenleaf

, Cottier

(2020) 2020 Ends a Decade of 62 New Data Privacy Laws. 163 Privacy Laws &Business International Report 24-26, Available at SSRN: https://ssrn.com/abstract=3572611.

101.

Greenleaf

(2019) Global Tables of Data Privacy Laws and Bills (6th Ed January 2019). Supplement to 157 Privacy Laws & Business International Report (PLBIR) 16 pgs, Available at SSRN: https://ssrn.com/abstract=3380794.