Abstract
Background: This review explores the potential of virtual reality (VR) and artificial intelligence (AI) to identify preclinical cognitive markers of Alzheimer disease (AD). By synthesizing recent studies, it aims to advance early diagnostic methods to detect AD before significant symptoms occur.
Objective: Research emphasizes the significance of early detection in AD during the preclinical phase, which does not involve cognitive impairment but nevertheless requires reliable biomarkers. Current biomarkers face challenges, prompting the exploration of cognitive behavior indicators beyond episodic memory.
Methods: Using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we searched Scopus, PubMed, and Google Scholar for studies on neuropsychiatric disorders utilizing conversational data.
Results: Following an analysis of 38 selected articles, we highlight verbal episodic memory as a sensitive preclinical AD marker, with supporting evidence from neuroimaging and genetic profiling. Executive functions precede memory decline, while processing speed is a significant correlate. The potential of VR remains underexplored, and AI algorithms offer a multidimensional approach to early neurocognitive disorder diagnosis.
Conclusions: Emerging technologies like VR and AI show promise for preclinical diagnostics, but thorough validation and regulation for clinical safety and efficacy are necessary. Continued technological advancements are expected to enhance early detection and management of AD.
doi:10.2196/62914
Keywords
Introduction
Nowadays, widespread access to health care systems and changes in living conditions have resulted in an aging population, leading to an increase in the prevalence of neurocognitive disorders (NCD) [,]. This phenomenon will lead to a societal change and place an additional burden on health care systems. Therefore, the main challenge at this time lies in the development of therapeutic measures at the pharmacological level that can prevent or halt the progression of Alzheimer disease (AD). However, so far, it has not been possible to find a pharmacological product that meets the safety and efficacy criteria necessary for large-scale use [-].
Conversely, numerous studies have shown that factors such as access to higher education, leading a healthy lifestyle, controlling cardiovascular risk factors, and being socially active can have a preventive effect by delaying the onset of symptoms and disease progression. In this context, it is of vital importance to identify emerging markers of AD that allow a diagnosis to be made in its preclinical stage.
The definition of preclinical AD varies, marked by criteria similarities and differences. It signifies the initial stage in the AD continuum, characterized by an extended asymptomatic phase with evidence of AD pathology, yet lacking cognitive, behavioral, or activities of daily living impairment (
Table 1). Duration varies (6-10 years), contingent on onset age, and progression to mild cognitive impairment (MCI) hinges on factors like age, sex, and apolipoprotein E status. The complexity arises from not all those meeting preclinical AD criteria progressing to MCI or AD dementia, adding nuances to the predictive analysis [].
a| AD continuum | Pathological and anatomical evidence of AD | Behavioral and psychological changes and cognitive impairment | Functional deficit | ||
| IWG-2b | Asymptomatic, at risk or presymptomatic | Prodromal | Mild AD dementia | Moderate AD dementia | Severe AD dementia |
| NIA-AAc | Preclinical | MCId (prodromal AD) | AD with mild dementia | AD with moderate dementia | AD with severe dementia |
| FDAe | Stages 1 and 2 (up to 20 years prior to clinical AD) | Stage 3 (episodic memory, executive function, visuospatial function; disease progression to clinical AD | Stages 4-6 (all domains, in a progressive way) | ||
aAD: Alzheimer disease.
bIWG: International Working Group.
cNIA-AA: National Institute on Aging—Alzheimer’s Association.
dMCI: mild cognitive impairment.
eFDA: Food and Drug Administration.
In this sense then, early biomarkers are crucial for assessing and monitoring AD. These indicators, recommended by the National Institute on Aging and the Alzheimer’s Association, include the assessment of extracellular amyloid beta (Aβ) and hyperphosphorylated tau protein (p-tau) in the brain [,]. The latest guidelines classify neurodegeneration using biomarkers such as amyloid positron emission tomography, Aβ42 levels, tau protein, and neuroimaging techniques. Incorporating these biomarkers aids in the early detection and understanding of AD, aligning with evolving clinical practices and research efforts [].
However, these biomarkers are expensive and not widely available in much of the world, and their results are not always consistent with the clinical manifestations of the patient, since even in the presence of positive markers, symptoms may not develop or may develop incompletely, and the time gap between the appearance of these biomarkers and the onset of symptoms has not yet been fully characterized.
Since the 1990s, there has been an increasing number of publications aiming to identify cognitive-behavioral markers of the shift from “normal” cognition (NC) to early symptoms of the disease [-]. In most studies, episodic memory appears as the domain that is altered in the first instance, reporting variability ranging from 20 [] to 2 years prior to symptom onset [,,-].
More recent studies have explored other domains that could be altered earlier, especially visuospatial function and executive functions; in addition, spatial navigation tasks have appeared as a new domain with characteristics that encompasses the 2 previous ones [-]. Some studies have proposed to evaluate functional (such as modifications in expansive activities of daily living, changes in mobility patterns) and behavioral aspects, as well as conduct a more thorough analysis of companion reports, and even subjective memory complaints. However, there is little agreement on how to measure them and which ones exhibit true early marker behavior of the disease [-]. At this point, it is important to mention that the vast majority of studies are oriented to the diagnosis of NCD due to AD, with only a handful of studies exploring some characteristics of frontotemporal dementias or NCD associated with Parkinson disease, though these tend to have a much lower quality of evidence than studies for AD [-].
The most current approaches have incorporated artificial intelligence (AI) and machine learning (ML) tools with the purpose of developing multivariate models that take advantage of these advanced technologies to integrate neuroimaging results, neuropsychological variables, and biomarkers in the early diagnosis of AD [,,-]. In this context, the use of virtual reality (VR) environments presents a novel and yet underexplored opportunity. So far, they have mainly been used in therapeutic applications, but their potential in the field of diagnostics remains to be investigated [,,-].
This study aims to review prevalent neurocognitive markers for the preclinical stages of tauopathies, particularly AD. It also explores recent research on VR applications in this context and aims to identify the most evidence-supported tests for assessing valuable cognitive domains.
Methods
The steps followed for article selection are based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) methodology [], as evidenced in recently published literature reviews on similar topics [,]. References were obtained from the search for articles in PubMed and Google Scholar. Articles from the last 10 years were included, although we also included 2 articles published more than 10 years ago because of their importance in the field. In the first instance, the terms used for the search were “pre-dementia Alzheimer’s disease,” “dementia early diagnosis,” “dementia early diagnosis financial,” “benefits of early diagnosis of Alzheimer’s disease cost effectiveness,” “benefits of early diagnosis of Alzheimer’s disease,” and “early diagnosis of Alzheimer’s disease.” However, the articles obtained with these operators corresponded to those with primary treatment objectives or involving exclusively biological measures in different fluids, development of new neuroimaging modalities, or developments of AI and ML algorithms without including neuropsychological parameters. We opted for the search strategy proposed by Bastin et al [] using the following keywords: ((memory AND longitudinal AND Alzheimer’s disease) AND (prodromal OR conversion OR preclinical)), ((mild cognitive impairment AND (Alzheimer’s disease OR dementia) AND neuropsychology AND (prediction OR longitudinal)), and (Alzheimer’s disease AND conversion AND neuropsychology).
When using these terms, none of the papers yielded results that included VR tools for diagnosis, so the pattern ((mild cognitive impairment AND (Alzheimer’s disease OR dementia) AND neuropsychology AND virtual reality)) was incorporated within the search strategy, finding 13 new papers of which 11 were incorporated, including 2 review papers that are discussed towards the end. Two papers were discarded because they were rehabilitation papers. The terms were used in English for most of the search engines, except for Google Scholar where the same terms were used in Spanish. Articles in both languages were included. As inclusion criteria, we selected those papers that made a longitudinal follow-up or an extensive review either through a systematic review or meta-analysis of the proposed topic. Those papers that could show conversion from NC to MCI or from MCI to dementia were weighted.
Based on this, 58 articles were selected based on the reading of the abstract. Studies conducted in animals, those that pursued primary treatment objectives, those conducted in patients with depression or other psychiatric/psychological comorbidity, those whose primary objective was to develop an AI algorithm, those that only used markers from blood, cerebrospinal fluid, urine, saliva, or other biological markers, and research theses were excluded. After reading the full text, those papers whose primary objective was the development of diagnostic imaging methods unrelated to neurocognitive variables (mainly related to functional magnetic resonance imaging [MRI], volumetry, cortical thickness) and without development of neurocognitive assessment (the tests used were not reported, they only showed raw scores of screening tests) were eliminated, as well as those papers that did not include a control group (MCI or normal aging or healthy controls).
Moreover, the Mendeley functionalities were used for the purpose of consolidating all references drawn from the various databases consulted. Subsequently, an integrated matching tool was used to identify and eliminate duplicates. In the initial screening phase, a preliminary selection of papers was made based on their relevance to the research question. To conduct this screening, the titles and abstracts of the papers were used to ascertain whether any of the 7 exclusion criteria set out in Table 2 were met. In instances where a decision was not immediately evident, the article was designated as a potential candidate for the subsequent phase.
| Reason for exclusiona | Values (n=26+17) |
| Screening | |
| 12 | |
| 4 | |
| 3 | |
| 5 | |
| 2 | |
| Eligibility | |
| 5 | |
| 6 | |
| 6 | |
aReasons for excluding papers during the screening (n=26) and eligibility process (n=17).
Subsequently, the full text of all papers that had been deemed eligible following the screening phase was considered. This was considered to be the most appropriate set of papers on the topic, having overcome the exclusion criteria set out in Table 2.
Results
The article selection process is outlined in 
Figure 1. This narrative outlines key discoveries from selected articles (Table 3 and Table 4)—studies without VR—categorizing the type of work and pinpointing the most early, sensitive, and specific neurocognitive variables according to each paper. Multiple cognitive variables, when presented, are organized by diagnostic profile. The chosen tests for variable measurements are specified. For a more detailed analysis of the reviewed articles, please consult the supplementary material (). Moreover, a dedicated section and table are focused on findings related to VR and AI tools utilized for diagnosis (Table 5).
| Reference | Neuropsychological domain | Gold standard | Validation | |
| 1 | [] | Executive functions | Cognitive markers:
Biological markers:
| |
| 2 | [] | Episodic memory | Cognitive markers: delayed memory testing Biological markers: atrophy of the hippocampal formation | |
| 3 | [] | Episodic memory | Cognitive markers:
Biological markers:
|
|
| 4 | [] | Global cognition | Cognitive markers:
Biological markers:
|
|
| 5 | [] | Episodic memory | Cognitive markers: Free and Cued Selective Reminding Test – Free Recall and Total Recall Biological markers: PiB-PET e |
|
| 6 | [] | Episodic memory | Cognitive markers:
Biological markers: cerebrospinal fluid p-tau/amyloid beta 42 |
|
| 7 | [] | Episodic memory and spatial navigation memory | Cognitive markers: navigation time, trajectory errors, and delayed recall of a map |
|
| 8 | [] | Verbal episodic memory | Cognitive markers:
Biological markers: PiB-PET e | |
| 9 | [] | EEG changes | Cognitive markers: neuropsychological assessment Biological markers: | |
| 10 | [] | Global cognition | Cognitive markers:
Biological markers:
| |
| 11 | [] | Episodic memory | Cognitive markers:
|
|
| 12 | [] | Visuospatial function and memory task (delayed recall and immediate recall) | Cognitive markers:
Biological markers:
| |
| 13 | [] | Voice |
| |
| 14 | [] | Hearing loss | Biological markers:
| |
| 15 | [] | Global cognition, verbal episodic memory, and attention shifting/flexibility verbal fluency | Cognitive markers:
| |
| 16 | [] | Verbal memory task: free delayed recall | Cognitive markers: Rey Auditory Learning Verbal Test | |
| 17 | [] | Global cognition and health | Medical records |
|
| 18 | [] | Sleep parameters | Cognitive markers:
Biological markers:
| n=100 |
| 19 | [] | Processing speed, community mobility, driving evaluations, AVD | Cognitive markers:
Biological markers:
| |
| 20 | [] | Sleep disorders | Cognitive markers:
Biological markers:
| |
| 21 | [] | Visuospatial working memory, anosognosia, visuomotor control, cognition spatial | Cognitive markers:
Biological markers:
| |
| 22 | [] | Global cognition | Cognitive markers:
Biological markers:
|
|
aAD: Alzheimer disease.
bAPOE: apolipoprotein E.
c CH: Cognitively Healthy
dCR: Cognitve Reserve
ePiB- PET: Pittsburgh - positron emission tomography
fFDG- PET: Fluorodeoxyglucose - positron emission tomography
gpMCI: Progress Mild Congitive Impairment
hsMCI: Stable Mild Cognitive Impairment
iMRI-s: structural magnetic resonance image
jADCN: AD neuropathological change
| Neuropsychological domain | Analytical method | Number | References |
| Episodic memory | Statistical: 5 | 5 | [,,,,] |
| Global cognition | Statistical: 4; artificial intelligence: 1 | 5 | [,,,,] |
| Verbal episodic memory | Statistical: 2; artificial intelligence: 1 | 3 | [,,] |
| Visuospatial, navigation | Statistical: 3 | 3 | [,,] |
| Sleep | Statistical: 2 | 2 | [,] |
| Electroencephalogram | Statistical: 1 | 1 | [] |
| Executive functions | Statistical: 1 | 1 | [] |
| Voice | Artificial intelligence: 1 | 1 | [] |
| Audio | Statistical: 1 | 1 | [] |
| Reference | Type of study | Neuropsychological domain | Gold standard | VRa cognitive tasks | |
| 1 | [] | Cross-sectional | Allocentric memory, egocentric memory | VR parks and mazes | Complete neuropsychological assessment; magnetic resonance imaging with volumetry |
| 2 | [] | Cross-sectional | Global cognition | Rivermead Behavioral Memory Test versus Montreal Cognitive Assessment; home selection task; VR versus Montreal Cognitive Assessment | Montreal Cognitive Assessment |
| 3 | [] | Cross-sectional | Functional level in Instrumental Activities of Daily Living | Coffee cup preparation task in a virtual environment | Mini Mental State Examination, Frontal Assessment Battery, Instrumental Activities of Daily Living (Lawton and Brody) |
| 4 | [] | Cross-sectional | Functional level with virtual reality | Tasks of daily life in a virtual environment | Complete neuropsychological battery, Magnetic Resonance Image, cognitive evoked potentials |
| 5 | [] | Cross-sectional | Global cognition | Virtual supermarket | Complete neuropsychological battery |
| 6 | [] | Cross-sectional | Ego/allocentric orientation | VR navigation task | Money’s Road Map test to compare the paper and virtual versions |
| 7 | [] | Ego/allocentric memory | VR navigation task | ||
| 8 | [] | Cross-sectional | Navigation task, entorhinal cortex navigation | VR navigation tasks | Digit Symbol, Free and Cued Selective Reminding Test, Mini Mental State Examination, North American National Adult Reading Test, Trail Making Test A and B, magnetic resonance imaging (entorhinal cortex) |
| 9 | [] | Cross-sectional | Comparing traditional paper-based neurocognitive assessment with neurocognitive assessment realized with immersive VR-3D technology. | Precalibrated VR-3D tests, customized 3D-VR testing, traditional 2D digitized tests | 3D tasks, 3D virtual maze, 2D tasks, T-maze |
| 10 | [] | Cross-sectional | Prospective memory | Recall of prospective and retrospective components of 7 interventions in a virtual city | Virtual driving: left or right. Two pedals (gas and brake). Control speed. |
| 11 | [] | Cross-sectional | Global cognition, motor performance, cognitive self-report | VR Functional Capacity Assessment Tool | Rey-Osterrieth Complex Figure, Rey Auditory Verbal Learning Test |
aVR: virtual reality.
A large majority of articles present verbal episodic memory measured by different tests (Free and Cued Selective Reminding Test, California Verbal Learning Test, Rey Auditory Verbal Learning Test) as the earliest measure with the best sensitivity and specificity ratio for the detection of the preclinical stage of AD and the most recent works propose to support this measure with different neuroimaging formats, genetic profiling (apolipoprotein E), or markers (Aβ, tau, p-tau) in cerebrospinal fluid [,,,,,,-]. Among these papers, the one by Gagliardi’s team comparing different neuropsychological tests for the measurement of episodic memory stands out [].
Harrington and collaborators [] focused on executive functions using the Stroop test, conducted on individuals without apparent cognitive impairment. They compared those with positive amyloid in cerebrospinal fluid to those without pathology, revealing executive failure preceding memory issues. The study emphasized the absence of standardized protocols for cognitive neural vulnerability in preclinical stages.
The works of Bastin [] and Gainotti [] are 2 large review papers that attempted to identify early neuropsychological markers in preclinical and conversion AD, comparing cognitively healthy subjects with and without evidence of pathology. Both papers agreed that episodic memory is the earliest marker, followed by semantic memory, although the protocols used for its assessment differed.
One of the most outstanding articles found in this search is the work of Mistridis et al []. The article agrees that verbal episodic memory is an early marker of cognitive impairment, detailing the sequence of decline of various cognitive functions in subjects with NC who progress to MCI. Verbal memory declines about 8 years before MCI, followed by episodic learning, visual memory, and semantic memory about 4 years before MCI. Executive functioning and processing speed declined about 2 years before MCI diagnosis. This suggests that multiple cognitive domains are valuable in assessing preclinical AD.
Another paper that attempted to establish a timeline is that of Soldan et al [], which used a composite score to evaluate 4 study groups of cognitively unimpaired patients, with the definitions of stage 0 (high Aβ and low tau), stage 1 (low Aβ and low tau), stage 2 (low Aβ and high tau), and suspected non-AD pathology (high Aβ and high tau). The article used linear mixed-effects models to estimate longitudinal cognitive composite scores among individuals in 4 preclinical AD groups. Adjusted for baseline factors, stage 2 individuals showed greater impairment and lower baseline scores compared to others, while stage 0, stage 1, and suspected non-AD pathology groups showed no significant differences. Involving 222 NC adults over 11 years, those in stage 2 (low Aβ, high tau/p-tau) showed markedly lower baseline scores and greater cognitive impairment, suggesting that abnormal amyloid and tau levels are necessary for cognitive differences in NC individuals.
Skills related to visuospatial function, which could prove important in the development of VR environments, were addressed in depth in the work of Caselli et al [] and Iliardi et al []. The first study suggested that visuospatial function deteriorates 20 years before clinical signs of AD, aligning with early pathological changes. It aims to sequence neurocognitive alterations in preclinical stages of AD. The second article addressed the importance of assessing visuospatial working memory, anosognosia, and visuomotor control in patients with MCI, who show worse performance compared to healthy controls and sometimes like patients with dementia. Assessment of visuomotor abilities may help distinguish high-risk AD patients from nonrisk AD patients.
Wadley and colleagues [] focused on processing speed impact on patient functionality, utilizing an ecological approach integrating neurocognitive and functional variables from professional observations, informant reports, and self-reports. They correlated these with neuroimaging and genetic markers, suggesting processing speed as a stronger correlation of everyday abilities than MRI patterns consistent with AD, and more accessible for measurement. Although neuroimaging demonstrates AD-related neurodegeneration, processing speed captures combined effects, comorbidities, and cognitive reserve, serving as an early marker for functional outcomes.
Among the studies that showed other tools for assessing preclinical stages of NCD that complemented cognitive neural vulnerability, there are studies that incorporate high-resolution electroencephalography measurements [], olfactory markers [], auditory [], retinal thickness measurement by optical coherence tomography [], use of ML algorithms for speech analysis [], and home-based sleep studies from specific devices []. Some of these works were intended to assess preclinical stages of dementias other than AD, and in the review done by Khan et al in 2020 [], the proposal of voice analysis in audios and videos allowed for this analysis in several languages. These studies suggest challenges with tests in English, requiring validation for diverse languages and contexts. However, they feature smaller populations, necessitate specific devices, and occur in nonclinical settings.
Literature on the use of VR for early diagnosis of preclinical NCD was limited, unlike its extensive use in cognitive skill training. Papers related to VR for diagnosis or treatment increased from 2008 to 2017, declined until 2020, then started rising again (Table 5).
By 2011, it was confirmed that ego and allocentric memory, linked to the parietal association cortex and medial temporal cortex, could serve as early markers of cognitive decline. VR models, like park or maze navigation, facilitated assessment []. Moreover, a Canadian team compared NC subjects with MCI patients using the Montreal Cognitive Assessment, Rivermead Behavioral Memory Test, and a VR test where subjects choose an apartment. The VR test correlated positively with the Montreal Cognitive Assessment, unlike traditional tests, but the sample size was small [].
The study of Allain et al [] compares cognitively healthy controls with patients with AD (varying stages, Mini Mental State Examination 18‐26 points) regarding functionality using a virtual coffee preparation task. Measures include completion time, achievement, and error scores compared with real-world tasks. With 24 patients with AD and 32 controls, the study demonstrates VR’s feasibility in studying AD deficits in ecologically valid environments, though limited to a single task.
In 2014, Tarnanas et al [] conducted research correlating daily living activities assessment in virtual environments with biomarkers like neuropsychological tests, event-related potentials, and MRI. Their study, robust compared to others, suggests VR performance equals biomarkers, attributed to VR’s cognitive fidelity and rich behavioral data reflecting early-stage neurocognitive processes.
Zygiuris et al [] introduced the virtual supermarket, a VR tool for rehabilitation and screening. The article extensively examined its validity with various measures and set cutoff points for a diagnostic algorithm. Several papers corroborate the apparent sensitivity of visuospatial functions in early NCD detection, evaluating ego and allocentric orientation. Morganti [] compared a paper and VR version of Money’s “Road Map” test, noting the virtual version’s increased difficulty. They hypothesized that participants solve the traditional version through egocentric spatial transformations, while VR decisions are made based on on-screen interactions. This makes VR tools an interesting alternative, but one that needs to be carefully evaluated to determine what factors influence the results. Similarly, Mohammadi et al [] used a VR navigation task to distinguish between monodomain and multidomain amnestic MCI, patients with AD, and normal controls, comparing results with traditional neuropsychological tests. They analyzed correct responses and response times in neighborhood and maze environments, establishing distinctive patterns in orientation (ego/allocentric) and visual/verbal memory.
Howett et al [] published a study correlating entorhinal cortex volume measures, proposed as an early AD marker, with changes in immersive VR tests. They found consistent results with traditional neuropsychological tests, suggesting navigation tasks aid in early AD diagnosis []. Within this group, Lecouvey et al [] utilized VR to evaluate prospective memory through tasks in a virtual city, comparing it with traditional neurocognitive measurements. Results confirmed early prospective memory alterations in AD, suggesting VR as a reliable assessment tool.
Turner’s team [] created a tablet-based tool evaluating real-world task competence in a realistic VR setting, assessing cognition, motor skills, and self-reported cognitive abilities in patients with Parkinson disease.
To conclude, 2 articles in this review elaborated VR protocols to assess neurocognitive domains, revealing the progression from NC to MCI, focusing on memory and visuospatial functions. In 1 study, an Italian team devised a protocol to detect early cognitive signs of conversion in AD, focusing on egocentric and allocentric spatial representations. Previous studies showed alterations, especially in allocentric frames, in patients with amnesia with MCI and AD. Their innovation lies in proposing an intrinsic connection between egocentric/allocentric frames and spatial relations. The results revealed deficits in allocentric coordinated judgments, implying a deviation towards AD in the representation of metric distances []. In the other study, Machado et al [] highlighted the use of VR environments, in particular 3D mazes, as diagnostic tools for MCI or dementia. Their study compares traditional methods with VR environments, combining precalibrated 3D tests with validated 2D neuropsychological assessments, allowing for assessment across several cognitive domains.
Finally, these studies, when paired with advanced AI algorithms, hold promise for early diagnosis. They enable cross-referencing of various variables, enhancing diagnostic accuracy [,,,,].
Discussion
Principal Findings
The objective of the Discussion section is to integrate the findings of this literature review and to explore their implications for advancing early diagnostic methods for AD using VR and AI. This section considers the strengths and limitations of current approaches, while also highlighting the potential role of these emerging technologies. Our findings indicate that while verbal episodic memory is a sensitive preclinical marker of AD, other cognitive domains, such as executive function and processing speed, may also serve as valuable early indicators. Furthermore, we examine the significant untapped potential of VR for the assessment of complex cognitive behaviors. VR has the capacity to simulate real-world environments, thereby enabling the evaluation of subtle deficits in cognition that would otherwise be challenging to capture in traditional assessments. AI, with its capacity to analyze multidimensional data, provides a promising approach for integrating biomarkers, cognitive data, and behavioral patterns. Nevertheless, there are considerable obstacles to the clinical implementation of these technologies, particularly in relation to the processes of validation, regulation, and ethical consideration.
The diagnosis of NCD involves advanced pathological definitions based on biomarkers but lacks changes in treatment perspectives. AD remains a paradigm, guiding NCD studies. Although pharmacological treatments are lacking, lifestyle and cardiovascular health modifications may prevent or delay onset, highlighting the importance of early diagnosis for effective intervention.
For more than 30 years, neuropsychological tests have provided the possibility of a more accurate diagnosis and of generating the basis for more specific pharmacological and nonpharmacological treatment plans. However, in recent years, with the advent of biomarkers, the scenario has become more complex [].
Combining various diagnostic methods like amyloid and tau protein determination in cerebrospinal fluid, positron emission tomography scans with fluorodeoxyglucose or florbetapir, and brain volumetric measurements enhances diagnostic accuracy. However, controversies arise as apparently healthy individuals exhibit positive pathological markers without developing measurable symptoms over extended periods, challenging current diagnostic instruments [,,,-].
Incorporating neuropsychological tests into composite diagnostic schemes becomes crucial for reliable conclusions in preclinical disease stages. Verbal episodic memory emerges as the earliest altered neurocognitive domain, followed by visual and semantic memory, despite measurement controversies [,-].
Recent studies suggest executive functions and processing speed alterations as early markers preceding declines in episodic and semantic memory. Improving processing speed through training protocols is vital for maintaining functionality in cognitively demanding tasks like driving and managing finances for individuals with MCI [,,,,].
Enhancing diagnostic accuracy for MCI involves indirectly estimating posteromedial parietal cortex function through simple outpatient neuropsychological tasks. Assessing visuomotor abilities in high-risk AD individuals distinguishes conversion probability. Iliardi et al [] revealed deficits in visuospatial working memory and metamemory as conversion predictors, intricately linked to posteromedial parietal cortex neuronal activity. This innovative approach spans various domains, potentially significantly improving the care and treatment of patients with MCI.
Three recent Spanish review studies focused on identifying cognitive processes assessed via VR and determining commonly used everyday life scenarios in virtual environment design to enhance ecological validity. They noted poor correlation between traditional neuropsychological tests and activities of daily living, prompting the development of technology-based instruments like VR and serious games for evaluation [-]. One of these studies reported that 52.3% of works implemented immersive VR, followed by nonimmersive VR (43.2%), and finally semi-immersive VR (4.5%) [].
VR environments are emerging as new diagnostic tools, but limitations exist. Selected studies were cross-sectional, with few patients evaluated. Gender differences, like navigation strategies in mazes, were noted. Performance in visuospatial tasks is influenced by reference points, instructions, and experimental parameters, highlighting the need for careful consideration in test design [,,,,,].
The translational approach merges animal models and patient evaluations, incorporating complex 3D tasks and standardized neuropsychological tests with automatic analysis, enhancing neuroscience’s cognitive function investigations. A clinical module with preconfigured 2D and 3D tasks simplifies routine patient evaluations [].
Limitations of This Literature Review
However, this literature review on preclinical cognitive markers of AD using VR and AI has several notable limitations. First, many studies were conducted with a limited number of participants, which reduces the generalizability of the findings. This small sample size makes it difficult to draw broad conclusions about the effectiveness of VR and AI for early diagnosis in different populations. Second, there is a lack of psychometric validity in some of the cognitive markers used, which raises concerns about the accuracy and reliability of these tools in assessing preclinical stages of AD. In addition, AI algorithms used in early diagnosis are often subject to bias due to the nature of the training data, potentially leading to biased decision-making that could disproportionately affect certain demographic groups. Ethical considerations also remain a challenge, particularly about privacy, consent, and transparency of AI-driven diagnostic processes. Addressing these limitations is critical to the development of robust, fair, and clinically useful AI- and VR-based tools for early detection of AD.
In conclusion, the integration of VR and AI in AD diagnosis represents a rapidly advancing and promising area in medical and neuropsychological research. Key considerations in this domain may include the following aspects:
- Diagnostic accuracy: The fusion of VR and AI provides an avenue for conducting highly accurate and objective assessments of cognitive functions in general, especially visuospatial functions. These tools can identify nuanced alterations in cognitive performance that traditional assessments frequently struggle to detect.
- Early detection: Detecting AD in its preliminary stages is crucial for early intervention and personalized support. VR and AI technologies can detect early signs of cognitive decline before symptoms manifest in daily activities, facilitating more timely and targeted interventions.
- Personalization: Through AI, VR evaluations can be personalized to match the unique capabilities and requirements of each patient. This enables tests to target specific cognitive areas of deficiency, enhancing diagnostic precision and identifying personalized areas for enhancement.
- Greater immersion: VR offers an immersive environment that replicates real-life scenarios and is ideal for evaluating patients’ proficiency in everyday activities like city navigation or shopping. These tasks are challenging to replicate within traditional clinical settings, making VR assessments particularly valuable as they facilitate an ecological and nonintrusive cognitive evaluation of the person.
- Continuous data acquisition: VR and AI facilitate continuous data gathering, enabling precise monitoring of disease advancement and the efficacy of therapeutic measures over time.
Recommendations for Future Research
Further research is needed to improve the effectiveness of VR and AI practices in early diagnosis of AD, particularly by addressing personalization and diagnostic accuracy. One promising direction is to develop VR assessments that can be tailored to each patient’s unique abilities and needs, ensuring that cognitive assessments are as individualized and responsive as possible. This personalization could contribute to a more accurate understanding of each patient’s cognitive status, thereby improving the sensitivity of early detection. In addition, research should focus on improving diagnostic accuracy by strengthening the psychometric validity of cognitive markers used in VR environments to ensure that these assessments are both reliable and clinically meaningful. In parallel, ongoing efforts are needed to mitigate biases in AI predictions by ensuring that models are trained on diverse datasets that accurately represent the broader population, thereby reducing inequalities in diagnostic outcomes. These future research efforts are essential to refine VR and AI as effective, unbiased, and personalized tools for early detection of AD.
Conclusions
Preclinical diagnosis of NCD remains challenging, with much more exploration needed. Although VR and AI offer benefits, challenges include expensive hardware, rigorous test validation, and results interpretation by trained professionals. Ethical concerns arise from patient data collection in virtual environments, necessitating strict confidentiality measures. Despite these challenges, VR’s use in AD diagnosis marks noteworthy progress in health care and NCD research. Continued technological advances will improve early detection and management. It is therefore essential to validate and regulate clinical safety and efficacy, delving into new preclinical cognitive markers. Further research is needed to improve the efficacy of VR and AI practices in the early diagnosis of AD, in particular addressing personalization and diagnostic accuracy.
Acknowledgments
This work is part of the IDENTIA project. This project is funded by PP2021-009109/MCIN/AEI/10.13039/ 501100011033, Ministry of Science and Innovation of Spain, the State Research Agency, and by the European Union “NextGeneration EU/PRTR.”
Conflicts of Interest
None declared.
Multimedia Appendix 1
Tables with the full description of each item and abbreviations.
DOCX File, 141 KBReferences
- Hugo J, Ganguli M. Dementia and cognitive impairment. Clin Geriatr Med. Aug 2014;30(3):421-442. [CrossRef]
- Prince M, Wimo A, Guerchet M, Gemma-Claire A, Wu YT, Prina M. World Alzheimer Report 2015: the global impact of dementia - an analysis of prevalence, incidence, cost and trends. Alzheimer’s Disease International. 2015. [CrossRef]
- Caselli RJ, Locke DEC, Dueck AC, et al. The neuropsychology of normal aging and preclinical Alzheimer’s disease. Alzheimers Dement. Jan 2014;10(1):84-92. [CrossRef] [Medline]
- Janghel RR, Rathore YK. Deep convolution neural network based system for early diagnosis of Alzheimer’s disease. IRBM. Aug 2021;42(4):258-267. [CrossRef]
- P DE Doctorado En Medicina Y Ciencias De La Salud, Por R, Sánchez-Juan C, Lage Martínez C, Sánchez-Juan P. Comportamiento ocular en el diagnóstico de la enfermedad de alzheimer [Report in Spanish]. URL: https://dialnet.unirioja.es/servlet/tesis?codigo=303662 [Accessed 2025-01-22]
- Porsteinsson AP, Isaacson RS, Knox S, Sabbagh MN, Rubino I. Diagnosis of early Alzheimer’s disease: clinical practice in 2021. J Prev Alz Dis. 2021;01:1-16. [CrossRef]
- Janeiro MH, Ardanaz CG, Sola-Sevilla N, et al. Biomarcadores en la enfermedad de Alzheimer [Article in Spanish]. Adv Lab Med / Av en Med de Lab. Mar 10, 2021;2(1):39-50. [CrossRef]
- Bjerke M, Engelborghs S. Cerebrospinal Fluid Biomarkers for Early and Differential Alzheimer’s Disease Diagnosis. IOS Press; 2018. [CrossRef]
- Khan S, Barve KH, Kumar MS. Recent advancements in pathogenesis, diagnostics and treatment of Alzheimer’s disease. Curr Neuropharmacol. 2020;18(11):1106-1125. [CrossRef] [Medline]
- Almkvist O. Neuropsychological features of early Alzheimer’s disease: preclinical and clinical stages. Acta Neurol Scand. Apr 1996;94(S165):63-71. URL: http://doi.wiley.com/10.1111/ane.1996.94.issue-S165 [CrossRef]
- Battista P, Salvatore C, Berlingeri M, Cerasa A, Castiglioni I. Artificial intelligence and neuropsychological measures: the case of Alzheimer’s disease. Neurosci & Biobehav Rev. Jul 2020;114:211-228. [CrossRef]
- Bastin C, Salmon E. Early neuropsychological detection of Alzheimer’s disease. Eur J Clin Nutr. Nov 2014;68(11):1192-1199. [CrossRef] [Medline]
- Collie A, Maruff P. The neuropsychology of preclinical Alzheimer’s disease and mild cognitive impairment. Neurosci Biobehav Rev. May 2000;24(3):365-374. [CrossRef] [Medline]
- Caselli RJ, Langlais BT, Dueck AC, et al. Neuropsychological decline up to 20 years before incident mild cognitive impairment. Alzheimers Dement. Mar 2020;16(3):512-523. [CrossRef] [Medline]
- Buckley RF, Maruff P, Ames D, et al. Subjective memory decline predicts greater rates of clinical progression in preclinical Alzheimer’s disease. Alzheimer’s & Dementia. Jul 2016;12(7):796-804. [CrossRef]
- Chételat G. Multimodal Neuroimaging in Alzheimer’s Disease: Early Diagnosis, Physiopathological Mechanisms, and Impact of Lifestyle. IOS Press; 2018. [CrossRef]
- Lucey BP, Wisch J, Boerwinkle AH, et al. Sleep and longitudinal cognitive performance in preclinical and early symptomatic Alzheimer’s disease. Brain (Bacau). Oct 22, 2021;144(9):2852-2862. [CrossRef] [Medline]
- Cogné M, Auriacombe S, Vasa L, et al. Are visual cues helpful for virtual spatial navigation and spatial memory in patients with mild cognitive impairment or Alzheimer’s disease? Neuropsychology. May 2018;32(4):385-400. [CrossRef] [Medline]
- Ilardi CR, Chieffi S, Iachini T, Iavarone A. Neuropsychology of posteromedial parietal cortex and conversion factors from mild cognitive impairment to Alzheimer’s disease: systematic search and state-of-the-art review. Aging Clin Exp Res. Feb 2022;34(2):289-307. [CrossRef] [Medline]
- Mohammadi A, Kargar M, Hesami E. Using virtual reality to distinguish subjects with multiple- but not single-domain amnestic mild cognitive impairment from normal elderly subjects. Psychogeriatrics. Mar 2018;18(2):132-142. [CrossRef] [Medline]
- Morganti F. Enacting space in virtual reality: a comparison between money’s road map test and its virtual version. Front Psychol. 2018;9:2410. [CrossRef] [Medline]
- Ruggiero G, Ruotolo F, Iavarone A, Iachini T. Allocentric coordinate spatial representations are impaired in aMCI and Alzheimer’s disease patients. Behav Brain Res. Sep 1, 2020;393:112793. [CrossRef] [Medline]
- Weniger G, Ruhleder M, Lange C, Wolf S, Irle E. Egocentric and allocentric memory as assessed by virtual reality in individuals with amnestic mild cognitive impairment. Neuropsychologia. Feb 2011;49(3):518-527. [CrossRef] [Medline]
- Allain P, Foloppe DA, Besnard J, et al. Detecting everyday action deficits in Alzheimer’s disease using a nonimmersive virtual reality kitchen. J Int Neuropsychol Soc. May 2014;20(5):468-477. [CrossRef] [Medline]
- Tarnanas I, Tsolaki M, Nef T, M. Müri R, Mosimann UP. Can a novel computerized cognitive screening test provide additional information for early detection of Alzheimer’s disease? Alzheimers Dement. Nov 2014;10(6):790-798. [CrossRef]
- Wadley VG, Bull TP, Zhang Y, et al. Cognitive processing speed is strongly related to driving skills, financial abilities, and other instrumental activities of daily living in persons with mild cognitive impairment and mild dementia. J Gerontol A Biol Sci Med Sci. Sep 13, 2021;76(10):1829-1838. [CrossRef] [Medline]
- Bora E, Walterfang M, Velakoulis D. Theory of mind in behavioural-variant frontotemporal dementia and Alzheimer’s disease: a meta-analysis. J Neurol Neurosurg Psychiatry. Jul 2015;86(7):714-719. [CrossRef] [Medline]
- Chicchi Giglioli IA, Pérez Gálvez B, Gil Granados A, Alcañiz Raya M. The virtual cooking task: a preliminary comparison between neuropsychological and ecological virtual reality tests to assess executive functions alterations in patients affected by alcohol use disorder. Cyberpsychol Behav Soc Netw. Oct 2021;24(10):673-682. [CrossRef] [Medline]
- Fang Y, Han D, Luo H. A virtual reality application for assessment for attention deficit hyperactivity disorder in school-aged children. Neuropsychiatr Dis Treat. 2019;15:1517-1523. [CrossRef] [Medline]
- Fornaguera TJ, Zamora CM. Revisión ¿Es la pérdida de la audición una posible medida para detectar de manera precoz la enfermedad de Alzheimer y la enfermedad de Parkinson? [Website in Spanish]. 2020. URL: https://www.medigraphic.com/cgi-bin/new/resumenI.cgi?IDARTICULO=101654 [Accessed 2023-07-11]
- Nir-Hadad SY, Weiss PL, Waizman A, Schwartz N, Kizony R. A virtual shopping task for the assessment of executive functions: validity for people with stroke. Neuropsychol Rehabil. Jul 2017;27(5):808-833. [CrossRef] [Medline]
- Serino S, Baglio F, Rossetto F, et al. Picture Interpretation Test (PIT) 360°: an innovative measure of executive functions. Sci Rep. Nov 22, 2017;7(1):16000. [CrossRef] [Medline]
- Fathi S, Ahmadi M, Dehnad A. Early diagnosis of Alzheimer’s disease based on deep learning: a systematic review. Comput Biol Med. Jul 2022;146:105634. [CrossRef] [Medline]
- Spasov S, Passamonti L, Duggento A, Liò P, Toschi N. A parameter-efficient deep learning approach to predict conversion from mild cognitive impairment to Alzheimer’s disease. Neuroimage. Apr 1, 2019;189:276-287. [CrossRef] [Medline]
- Pereira T, Ferreira FL, Cardoso S, et al. Neuropsychological predictors of conversion from mild cognitive impairment to Alzheimer’s disease: a feature selection ensemble combining stability and predictability. BMC Med Inform Decis Mak. Dec 19, 2018;18(1):137. [CrossRef] [Medline]
- Tuena C, Mancuso V, Stramba-Badiale C, et al. Egocentric and allocentric spatial memory in mild cognitive impairment with real-world and virtual navigation tasks: a systematic review. J Alzheimers Dis. 2021;79(1):95-116. [CrossRef] [Medline]
- Turner TH, Atkins A, Keefe RSE. Virtual Reality Functional Capacity Assessment Tool (VRFCAT-SL) in Parkinson’s disease. J Parkinsons Dis. 2021;11(4):1917-1925. [CrossRef] [Medline]
- Zhu K, Lin R, Li H. Study of virtual reality for mild cognitive impairment: a bibliometric analysis using CiteSpace. Int J Nurs Sci. Jan 2022;9(1):129-136. [CrossRef] [Medline]
- Nolin P, Banville F, Cloutier J, Allain P. Virtual reality as a new approach to assess cognitive decline in the elderly. AJIS. Oct 1, 2013;2(8):612-616. [CrossRef]
- Caggianese G, Chirico A, De Pietro G, et al. Towards a virtual reality cognitive training system for mild cognitive impairment and Alzheimer’s disease patients. Presented at: 2018 32nd International Conference on Advanced Information Networking and Applications Workshops (WAINA); May 16-18, 2018:663-667; Krakow. [CrossRef]
- Mohammadi A, Hesami E, Kargar M, Shams J. Detecting allocentric and egocentric navigation deficits in patients with schizophrenia and bipolar disorder using virtual reality. Neuropsychol Rehabil. Apr 2018;28(3):398-415. [CrossRef] [Medline]
- Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC. Declaración PRISMA 2020: una guía actualizada para la publicación de revisiones sistemáticas [Article in Spanish]. Rev Esp Cardiol. 2021;74(9):790-799. [CrossRef] [Medline]
- Xie B, Tao C, Li J, Hilsabeck RC, Aguirre A. Artificial intelligence for caregivers of persons with Alzheimer’s disease and related dementias: systematic literature review. JMIR Med Inform. Aug 20, 2020;8(8):e18189. [CrossRef] [Medline]
- Choudhury A, Asan O. Role of artificial intelligence in patient safety outcomes: systematic literature review. JMIR Med Inform. Jul 24, 2020;8(7):e18599. [CrossRef] [Medline]
- Harrington MG, Chiang J, Pogoda JM, et al. Executive function changes before memory in preclinical Alzheimer’s pathology: a prospective, cross-sectional, case control study. PLoS ONE. 2013;8(11):e79378. [CrossRef] [Medline]
- Gainotti G, Quaranta D, Vita MG, Marra C. Neuropsychological Predictors of Conversion from Mild Cognitive Impairment to Alzheimer’s Disease. IOS Press; 2014. [CrossRef]
- Soldan A, Pettigrew C, Cai Q, et al. Hypothetical preclinical Alzheimer disease groups and longitudinal cognitive change. JAMA Neurol. Jun 1, 2016;73(6):698-705. [CrossRef] [Medline]
- Soldan A, Pettigrew C, Cai Q, et al. Cognitive reserve and long-term change in cognition in aging and preclinical Alzheimer’s disease. Neurobiol Aging. Dec 2017;60:164-172. [CrossRef] [Medline]
- Papp KV, Rentz DM, Mormino EC, et al. Cued memory decline in biomarker-defined preclinical Alzheimer disease. Neurol (ECronicon). Apr 11, 2017;88(15):1431-1438. [CrossRef] [Medline]
- Schindler SE, Jasielec MS, Weng H, et al. Neuropsychological measures that detect early impairment and decline in preclinical Alzheimer disease. Neurobiol Aging. Aug 2017;56:25-32. [CrossRef] [Medline]
- Gagliardi G, Epelbaum S, Houot M, et al. Which episodic memory performance is associated with Alzheimer’s disease biomarkers in elderly cognitive complainers? Evidence from a longitudinal observational study with four episodic memory tests (Insight-PreAD). J Alzheimers Dis. 2019;70(3):811-824. [CrossRef] [Medline]
- Gaubert S, Raimondo F, Houot M, et al. EEG evidence of compensatory mechanisms in preclinical Alzheimer’s disease. Brain (Bacau). Jul 1, 2019;142(7):2096-2112. [CrossRef]
- Ricci M, Todino V, Magarelli M, et al. Spect-neuropsychology correlations in very mild Alzheimer’s disease and amnesic mild cognitive impairment. Arch Gerontol Geriatr. 2020;89:104085. [CrossRef] [Medline]
- Nedelec T, Couvy-Duchesne B, Monnet F, et al. Identifying health conditions associated with Alzheimer’s disease up to 15 years before diagnosis: an agnostic study of French and British health records. Lancet Digit Health. Mar 2022;4(3):e169-e178. [CrossRef] [Medline]
- Smirnov DS, Ashton NJ, Blennow K, et al. Plasma biomarkers for Alzheimer’s disease in relation to neuropathology and cognitive change. Acta Neuropathol. Apr 2022;143(4):487-503. [CrossRef] [Medline]
- Zygouris S, Giakoumis D, Votis K, et al. Can a virtual reality cognitive training application fulfill a dual role? Using the virtual supermarket cognitive training application as a screening tool for mild cognitive impairment. J Alzheimers Dis. 2015;44(4):1333-1347. [CrossRef] [Medline]
- Howett D, Castegnaro A, Krzywicka K, et al. Differentiation of mild cognitive impairment using an entorhinal cortex-based test of virtual reality navigation. Brain (Bacau). Jun 1, 2019;142(6):1751-1766. [CrossRef] [Medline]
- Machado ML, Lefèvre N, Philoxene B, et al. New software dedicated to virtual mazes for human cognitive investigations. J Neurosci Methods. Nov 1, 2019;327:108388. [CrossRef] [Medline]
- Lecouvey G, Morand A, Gonneaud J, et al. An impairment of prospective memory in mild Alzheimer’s disease: a ride in a virtual town. Front Psychol. 2019;10:241. [CrossRef] [Medline]
- Bondi MW, Edmonds EC, Salmon DP. Alzheimer’s disease: past, present, and future. J Int Neuropsychol Soc. Oct 2017;23(9-10):818-831. [CrossRef] [Medline]
- Frisoni GB, Boccardi M, Barkhof F, et al. Strategic roadmap for an early diagnosis of Alzheimer’s disease based on biomarkers. Lancet Neurol. Aug 2017;16(8):661-676. [CrossRef] [Medline]
- Mortamais M, Ash JA, Harrison J, et al. Detecting cognitive changes in preclinical Alzheimer’s disease: a review of its feasibility. Alzheimer’s & Dementia. Apr 2017;13(4):468-492. [CrossRef]
- Mistridis P, Krumm S, Monsch AU, Berres M, Taylor KI. The 12 years preceding mild cognitive impairment due to Alzheimer’s disease: the temporal emergence of cognitive decline. J Alzheimers Dis. 2015;48(4):1095-1107. [CrossRef] [Medline]
- An overview of new and emerging technologies for early diagnosis of Alzheimer disease. Canada’s Drug Agency. URL: https://www.cda-amc.ca/overview-new-and-emerging-technologies-early-diagnosis-alzheimer-disease [Accessed 2025-01-22]
- Tarnanas I, Tsolaki M, Nef T, M Müri R, Mosimann UP. Can a novel computerized cognitive screening test provide additional information for early detection of Alzheimer’s disease? Alz Dement. Nov 2014;10(6):790-798. [CrossRef] [Medline]
- Langbaum JB, Zissimopoulos J, Au R, et al. Recommendations to address key recruitment challenges of Alzheimer’s disease clinical trials. Alzheimers Dement. Feb 2023;19(2):696-707. [CrossRef] [Medline]
- Altuna-Azkargorta M, Mendioroz-Iriarte M. Blood biomarkers in Alzheimer’s disease. Neurol Ed. 2021;36(9):704-710. [CrossRef] [Medline]
- Wei ZYD, Shetty AK. Can mild cognitive impairment and Alzheimer’s disease be diagnosed by monitoring a miRNA triad in the blood? Aging Cell. Jun 2022;21(6):e13627. [CrossRef] [Medline]
- Jack CR Jr, Knopman DS, Weigand SD, et al. An operational approach to National Institute on Aging-Alzheimer’s Association criteria for preclinical Alzheimer disease. Ann Neurol. Jun 2012;71(6):765-775. [CrossRef] [Medline]
- Delgado-Reyes AC, Sánchez Lopez JV. Escenarios virtuales para la evaluación neuropsicológica: una revisión de tema [Virtual scenarios for neuropsychological assessment: a topic review]. Panam J Neuropsychol. 2021;15:2-196. [CrossRef]
- Sánchez-Escudero JP, Galvis-Herrera AM, Sánchez-Trujillo D, et al. Virtual reality and serious videogame-based instruments for assessing spatial navigation in Alzheimer’s disease: a systematic review of psychometric properties. Neuropsychol Rev. Feb 26, 2024. [CrossRef] [Medline]
- Valladares-Rodríguez S, Fernández-Iglesias MJ, Anido-Rifón LE, Pacheco-Lorenzo M. Evaluation of the predictive ability and user acceptance of Panoramix 2.0, an AI-based e-health tool for the detection of cognitive impairment. Electronics (Basel). 2022;11(21):3424. [CrossRef]
- Valladares-Rodriguez S, Perez-Rodriguez R, Facal D, Fernandez-Iglesias MJ, Anido-Rifon L, Mouriño-Garcia M. Design process and preliminary psychometric study of a video game to detect cognitive impairment in senior adults. PeerJ. 2017;5:e3508. [CrossRef] [Medline]
Abbreviations
| AD: Alzheimer disease |
| AI: artificial intelligence |
| Aβ: amyloid beta |
| MCI: mild cognitive impairment |
| ML: machine learning |
| MRI: magnetic resonance imaging |
| NC: normal cognition |
| NCD: neurocognitive disorders |
| p-tau: hyperphosphorylated tau protein |
| PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses |
| VR: virtual reality |
Edited by Arriel Benis; submitted 04.06.24; peer-reviewed by E Harriss, Komal Kumar Raja; final revised version received 04.11.24; accepted 03.12.24; published 28.01.25.
Copyright© María de la Paz Scribano Parada, Fátima González Palau, Sonia Valladares Rodríguez, Mariano Rincon, Maria José Rico Barroeta, Marta García Rodriguez, Yolanda Bueno Aguado, Ana Herrero Blanco, Estela Díaz-López, Margarita Bachiller Mayoral, Raquel Losada Durán. Originally published in JMIR Medical Informatics (https://medinform.jmir.org), 28.1.2025.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Medical Informatics, is properly cited. The complete bibliographic information, a link to the original publication on https://medinform.jmir.org/, as well as this copyright and license information must be included.