Margarita Bachiller

Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: In today's context, widespread healthcare access and changing living conditions contribute to an aging population, escalating the prevalence of Neurocognitive Disorders (NCD). This shift poses societal challenges and strains healthcare systems, emphasizing the urgent need for pharmacological interventions to address Alzheimer's Disease (AD) progression. Despite efforts, a universally effective pharmacological solution remains elusive. Simultaneously, studies highlight that factors like education, healthy lifestyles, and social engagement can delay AD onset. Identifying early markers, particularly in the preclinical stage, becomes crucial. The preclinical stage, characterized by evidence of AD pathology without cognitive or behavioral impairment, prompts the quest for early biomarkers. Current biomarkers, while valuable, face challenges of cost, limited availability, and inconclusive correlation with clinical manifestations. This underscores the necessity for alternative cognitive-behavioral markers, with recent studies exploring domains beyond episodic memory. Moreover, advancements incorporating Artificial Intelligence (AI) and Machine Learning (ML) aim to integrate neuroimaging, neuropsychological variables, and biomarkers for early AD diagnosis. Additionally, the exploration of Virtual Reality (VR) environments presents a novel avenue for diagnostics Objective: This project systematically reviews neurocognitive markers in preclinical AD diagnosis, emphasizing recent studies involving VR. A secondary objective is to discern tests with the highest evidentiary support for each proposed domain. Methods: Methods: The article selection process involved an initial search using specific keywords, such as "memory AND longitudinal AND Alzheimer's disease," "mild cognitive impairment AND (Alzheimer's disease OR dementia) AND neuropsychology AND (prediction OR longitudinal)," and "Alzheimer's disease AND conversion AND neuropsychology." However, this initial search did not yield any papers discussing virtual reality tools for diagnosis. To address this, additional terms, including "virtual reality," were incorporated into the search strategy. This adjustment resulted in the identification of 13 new papers, with 11 fully integrated into the study. Notably, two review papers from this subset are discussed in the later sections. Two papers were excluded due to their focus on rehabilitation rather than the intended diagnostic scope Results: Results: Selected articles highlight verbal episodic memory, measured by various tests, as the earliest, most sensitive marker for preclinical Alzheimer's Disease (AD). Neuroimaging, genetic profiling (APOE), and CSF markers (A$beta$, tau, p-tau) complement this assessment. Notable studies detail the sequence of cognitive decline, with executive functions preceding memory failure. Speed of processing emerges as a significant correlate, surpassing MRI patterns for functional outcomes. Limited literature explores virtual reality's potential for early diagnosis, with some studies demonstrating its efficacy in assessing visuospatial and memory functions. Promisingly, AI algorithms, coupled with advanced diagnostic tools, offer a multidimensional approach for accurate early diagnosis of neurocognitive disorders. Conclusions: The preclinical diagnosis of neurocognitive disorders is challenging, with emerging tools like VR and AI offering potential but also posing obstacles such as cost, rigorous validation, and the need for expert interpretation. Ethical concerns regarding patient data in virtual environments require attention. Despite challenges, the evolving role of VR in Alzheimer's diagnosis signifies a promising advancement in healthcare. Continued technological refinement is expected to enhance early identification and management, demanding thorough validation and regulation for clinical efficacy and safety.

BACKGROUND In today's context, widespread healthcare access and changing living conditions contribute to an aging population, escalating the prevalence of Neurocognitive Disorders (NCD). This shift poses societal challenges and strains healthcare systems, emphasizing the urgent need for pharmacological interventions to address Alzheimer's Disease (AD) progression. Despite efforts, a universally effective pharmacological solution remains elusive. Simultaneously, studies highlight that factors like education, healthy lifestyles, and social engagement can delay AD onset. Identifying early markers, particularly in the preclinical stage, becomes crucial. The preclinical stage, characterized by evidence of AD pathology without cognitive or behavioral impairment, prompts the quest for early biomarkers. Current biomarkers, while valuable, face challenges of cost, limited availability, and inconclusive correlation with clinical manifestations. This underscores the necessity for alternative cognitive-behavioral markers, with recent studies exploring domains beyond episodic memory. Moreover, advancements incorporating Artificial Intelligence (AI) and Machine Learning (ML) aim to integrate neuroimaging, neuropsychological variables, and biomarkers for early AD diagnosis. Additionally, the exploration of Virtual Reality (VR) environments presents a novel avenue for diagnostics OBJECTIVE This project systematically reviews neurocognitive markers in preclinical AD diagnosis, emphasizing recent studies involving VR. A secondary objective is to discern tests with the highest evidentiary support for each proposed domain. METHODS Methods: The article selection process involved an initial search using specific keywords, such as "memory AND longitudinal AND Alzheimer's disease," "mild cognitive impairment AND (Alzheimer's disease OR dementia) AND neuropsychology AND (prediction OR longitudinal)," and "Alzheimer's disease AND conversion AND neuropsychology." However, this initial search did not yield any papers discussing virtual reality tools for diagnosis. To address this, additional terms, including "virtual reality," were incorporated into the search strategy. This adjustment resulted in the identification of 13 new papers, with 11 fully integrated into the study. Notably, two review papers from this subset are discussed in the later sections. Two papers were excluded due to their focus on rehabilitation rather than the intended diagnostic scope RESULTS Results: Selected articles highlight verbal episodic memory, measured by various tests, as the earliest, most sensitive marker for preclinical Alzheimer's Disease (AD). Neuroimaging, genetic profiling (APOE), and CSF markers (A$beta$, tau, p-tau) complement this assessment. Notable studies detail the sequence of cognitive decline, with executive functions preceding memory failure. Speed of processing emerges as a significant correlate, surpassing MRI patterns for functional outcomes. Limited literature explores virtual reality's potential for early diagnosis, with some studies demonstrating its efficacy in assessing visuospatial and memory functions. Promisingly, AI algorithms, coupled with advanced diagnostic tools, offer a multidimensional approach for accurate early diagnosis of neurocognitive disorders. CONCLUSIONS The preclinical diagnosis of neurocognitive disorders is challenging, with emerging tools like VR and AI offering potential but also posing obstacles such as cost, rigorous validation, and the need for expert interpretation. Ethical concerns regarding patient data in virtual environments require attention. Despite challenges, the evolving role of VR in Alzheimer's diagnosis signifies a promising advancement in healthcare. Continued technological refinement is expected to enhance early identification and management, demanding thorough validation and regulation for clinical efficacy and safety.

Some recent studies show that filters in convolutional neural networks (CNNs) have low color selectivity in datasets of natural scenes such as Imagenet. CNNs, bio-inspired by the visual cortex, are characterized by their hierarchical learning structure which appears to gradually transform the representation space. Inspired by the direct connection between the LGN and V4, which allows V4 to handle low-level information closer to the trichromatic input in addition to processed information that comes from V2/V3, we propose the addition of a long skip connection (LSC) between the first and last blocks of the feature extraction stage to allow deeper parts of the network to receive information from shallower layers. This type of connection improves classification accuracy by combining simple-visual and complex-abstract features to create more color-selective ones. We have applied this strategy to classic CNN architectures and quantitatively and qualitatively analyzed the improvement in accuracy while focusing on color selectivity. The results show that, in general, skip connections improve accuracy, but LSC improves it even more and enhances the color selectivity of the original CNN architectures. As a side result, we propose a new color representation procedure for organizing and filtering feature maps, making their visualization more manageable for qualitative color selectivity analysis.

Some recent studies show that filters in convolutional neural networks (CNNs) have low color selectivity in datasets of natural scenes such as Imagenet. CNNs, bio-inspired in the visual cortex, are characterized by their hierarchical learning structure, which appears to gradually transform the representation space. Inspired by the direct connection between the LGN and V4, which allows V4 to handle low-level information closer to the trichromatic input in addition to processed information that comes from V2/V3, we propose the addition of a long skip connection (LSC) between the first and last blocks of the feature extraction stage to allow deeper parts of the network to receive information from shallower layers. This type of connection improves classification accuracy by combining simple-visual and complex-abstract features to create more color-selective ones. We have applied this strategy to classic CNN architectures and quantitatively and qualitatively analyzed the improvement in accuracy while focusing on color selectivity. The results show that, in general, skip connections improve accuracy, but LSC improves it even more and enhances the color selectivity of the original CNN architectures. As a side result, we propose a new color representation procedure for organizing and filtering feature maps, making their visualization more manageable.