The COMFORTage project: study protocol for the integration of multiple sources towards personalised preventions at Ace Alzheimer Center Barcelona

Introduction: Ageing is accompanied by gradual biological and cognitive changes that increase vulnerability to chronic diseases and neurodegenerative conditions. As populations age, dementia prevalence continues to rise, highlighting the need for earlier detection and personalised prevention strategies. Against this background, the COMFORTage project, funded by Horizon Europe, brings together a multidisciplinary consortium across 12 countries to advance innovative, scalable solutions for dementia care. By integrating digital platforms, biomarker research, and precision medicine, COMFORTage seeks to develop artificial intelligence (AI)–driven tools that support more precise and adaptive interventions. Central to this effort are the Virtualized AI-Based Healthcare Platform and Patient Digital Twins, which enable personalised monitoring and decision support. Within this framework, Pilot 3 at Ace Alzheimer Center Barcelona focuses on individuals with mild cognitive impairment and mild Alzheimer's disease dementia, evaluating the effects of cognitive and functional stimulation and contributing multimodal data to optimise the AI platform.

Methods: Pilot 3 is a randomised, open-label study involving retrospective and prospective datasets. Participants undergo clinical, genetic, neuropsychological, cerebrospinal fluid (CSF) and plasma biomarker assessments, magnetic resonance imaging (MRI), and spontaneous speech analysis. The primary outcomes assess cognitive decline using composite scores from the Neuropsychological Battery used in Ace (NBACE), targeting attention, memory, visuospatial/perceptual functions, executive functions, and language, over a two-year follow-up. Three digital platforms provided by the consortium will be used as cognitive and functional stimulation tools for participants. The intervention's effects on cognitive decline will be evaluated through changes in NBACE composite scores. Secondary objectives include assessing impacts on physical, psychological, social, and functional well-being; examining associations between biological variables and cognitive changes; and analyzing spontaneous speech as a remote, scalable proxy for cognitive status.

Discussion: Findings from Pilot 3 will contribute to COMFORTage's broader mission, offering critical insights into the scalability and real-world implementation of AI-powered dementia care solutions. This integrated approach highlights the potential of precision medicine and advanced digital tools to elevate global standards in dementia management.

Clinical Trial Registration: identifier NCT07031167.

1 Introduction

Ageing is inherently associated with gradual changes in brain function, metabolism, and physical performance. While some older adults adapt well and maintain cognitive and physical well-being, others experience a decline that increases their vulnerability to health complications. Regardless of baseline health status, these age-related changes can predispose individuals to both cognitive and functional deterioration. Early detection of risk factors is crucial for developing proactive interventions that promote healthy ageing (1). A rapidly ageing population amplifies these challenges, as increasing longevity is accompanied by higher rates of chronic disease, frailty, and cognitive impairment, leading to substantial social and economic consequences. Preserving cognitive reserve and functional independence has therefore become a key priority, underscoring the need for earlier, personalised strategies to detect disease processes before the onset of clinical dementia (2).

As these age-related changes progress, some individuals develop neurodegenerative conditions, most notably Alzheimer's disease (AD), which markedly accelerates cognitive and functional decline. When such disorders coexist with frailty, their combined impact further increases the risk of disability, loss of autonomy, and reduced quality of life (3, 4).

AD is the leading cause of dementia, accounting for 60%–80% of cases worldwide, and its prevalence continues to rise as improvements in healthcare and living standards have led to ageing populations (5). The number of people living with dementia worldwide in 2019 was estimated at 57 million and is projected to increase to 153 million by 2050 (6). In Europe, approximately 9.8 million people were living with dementia in 2018, and this number is projected to nearly double to around 18.8 million by 2050 (7). AD has become not only a growing medical challenge but also a major socio-economic concern (7, 8). Addressing this burden requires integrated and scalable strategies that enable earlier detection and more personalised prevention of dementia.

The disease follows a continuum that begins with asymptomatic pathological changes, progresses through mild cognitive impairment (MCI), and culminates in severe dementia. MCI, which manifests as measurable cognitive decline that does not yet significantly disrupt daily life, is widely recognised as a prodromal stage of AD (9, 10). Given the elevated risk of progression to dementia in certain individuals with MCI, early detection and targeted interventions are critical (11–13). This underscores the importance of developing multimodal approaches for early and precise identification of neurodegenerative changes.

Addressing these growing challenges requires pioneering approaches that transcend traditional healthcare models. The COMFORTage project, funded by the European Union under the Horizon Europe Programme, is an innovative European initiative running from 2024 to 2027 that seeks to address the complex interplay between ageing, dementia, and frailty through the integration of clinical expertise, technological innovation, and community engagement. Bringing together 39 partners across 12 countries, COMFORTage unites expertise from diverse disciplines to foster a truly collaborative and multidisciplinary approach. This multidisciplinary collaboration aims to establish a pan-European framework for community-based, integrated, and person-centered solutions for age-related health challenges (14).

The project supports older adults experiencing cognitive and functional decline due to neurodegenerative disorders, frailty, or other age-related impairments. These conditions are major contributors to disability and dependence, imposing a growing burden on individuals, caregivers, and healthcare systems. Within this framework, COMFORTage applies artificial intelligence (AI) and digital innovation to improve prevention, early detection, and personalised care, while promoting digital inclusion and collaboration among healthcare providers, patients, and caregivers. The project underscores the need for adaptable, multifactorial strategies that address the diverse and evolving needs of ageing populations through inclusive and person-centred interventions.

Several large multicentre initiatives have advanced different aspects of the early-detection-to-care pathway in AD, including the development of AI-based risk models, the validation of digital measures using smartphones and wearables, the creation of well-characterised longitudinal cohorts, and the implementation of lifestyle-based prevention or home-monitoring approaches (15–18). While these efforts have each contributed valuable knowledge, they generally focus on single components of the dementia continuum. In contrast, COMFORTage brings these complementary approaches together within a single, coherent framework that supports early detection, personalised intervention, and coordinated care, and evaluates their real-world feasibility across 13 community-based pilots.

Among the project's key innovations is the Virtualized AI-Based Healthcare Platform (VHP), a cornerstone of COMFORTage designed to improve the management of dementia, frailty, and other age-related conditions through secure, data-driven, and personalised care solutions. The platform centralizes artificial intelligence resources that support risk-factor analysis, early diagnosis, and individualised clinical decision-making across mental and physical domains of ageing.

The VHP integrates data from multiple sources—such as electronic health records, wearable devices, and real-world evidence—into unified Holistic Health Records (HHRs), creating comprehensive and interoperable patient profiles. The HHR concept was originally introduced by the iHELP project for holistic detection, early risk prediction, and monitoring of pancreatic cancer (19). Built on a secure, General Data Protection Regulation–compliant architecture, the VHP ensures data privacy, transparency, and traceability. At its core is the Integrated Care Model Library (ICML), a repository of AI-driven tools and resources designed to address the complexities of age-related diseases. By leveraging genomic, behavioural, and clinical data, the ICML enables precision in prevention, intervention, and treatment, providing clinicians with predictive and explainable tools to support decision-making. The platform also includes Patient Digital Twins, which generate individualised digital models that allow continuous monitoring and personalised adaptation of care plans.

Complementary assistive and educational modules are also integrated into the VHP to strengthen digital engagement and support self-management. These include the Training and Educational Toolkit (TET), which provides accessible learning materials and interactive resources—such as cognitive training games and practical guidance on using digital health tools—to help users build skills relevant to their care. These resources are delivered through the Training and Educational Marketplace (TEM), a secure environment that hosts a curated selection of applications and uses a recommendation system to suggest tailored content based on individual needs and preferences. Together, the TET and TEM enhance digital literacy among older adults, caregivers, and healthcare professionals, reinforcing the VHP's role as an integrated platform for prevention, monitoring, and personalised intervention.

The COMFORTage project employs a framework of 13 pilot studies distributed across five research clusters to evaluate and refine the VHP and ensure its applicability across diverse populations. These pilots serve as test beds for new technologies, methodologies, and interventions, focusing on scalability and adaptability to regional healthcare systems and cultural contexts. The pilots are strategically located across Europe and involve key academic and clinical partners.

The five clusters integrating the different pilots are: Cluster A) focuses on identifying causal risk factors and developing community-based strategies for early diagnosis and prevention; Cluster B) examines the relationship between dementia and comorbid conditions, broadening the understanding of factors that influence progression; Cluster C) develops personalised interventions to delay progression and improve quality of life; Cluster D) addresses physical health and safety, focusing on neuromechanics and fall prevention; and Cluster E) explores the role of digital innovation hubs and living labs in promoting active ageing and social inclusion.

Ace Alzheimer Center Barcelona (Ace) leads Pilot 3 within Cluster A of the COMFORTage project, which focuses on early detection and cognitive stimulation for individuals with MCI and mild AD dementia. The pilot plays a pivotal role in both the clinical validation of the consortium's interventions and the development of the VHP through the contribution of retrospective and prospective datasets.

In this study, participants are randomly assigned to active or control groups and followed longitudinally for two years to monitor cognitive and functional decline. Assessments include clinical, neuropsychological, genetic, cerebrospinal fluid (CSF), plasma biomarker, and magnetic resonance imaging (MRI) data, enabling an integrated analysis of biological and cognitive interactions (12, 20). These multimodal biomarkers offer complementary information across the disease continuum, supporting more accurate characterization of both early pathology and ongoing neurodegeneration (21, 22). Periodic speech-based digital assessments further complement these measures by providing remote and ecologically valid indicators of subtle cognitive and behavioural change (23–25).

By combining traditional and digital measures, the Ace Pilot aims to evaluate the efficacy of cognitive and functional stimulation interventions while simultaneously generating high-quality, multimodal data to train and validate the AI models underpinning the VHP. Primary outcomes focus on cognitive decline, while secondary measures assess quality of life (QoL), physical activity, and satisfaction with the intervention.

2 Methods

2.1 Study design

The Ace Pilot (Pilot 3) is a randomised, open-label, two-year clinical trial comparing outcomes between an active group receiving a structured cognitive and functional stimulation program and a control group receiving standard care. In parallel, retrospective datasets from Ace's existing cohorts, including clinical, neuropsychological, genetic, CSF and plasma biomarkers, and MRI data, will be shared with the consortium to support the development and validation of the VHP's AI algorithms.

2.2 Objectives

The Ace Pilot aims to enhance the quality of life and overall well-being of individuals with MCI and mild AD dementia by prospectively evaluating a structured cognitive and functional stimulation program, while simultaneously generating high-quality retrospective datasets to support the development and validation of the VHP's AI-based models.

2.2.1 Primary clinical outcome

To evaluate cognitive decline as the primary efficacy outcome between the active and control groups after a two-year follow-up. Cognitive change will be quantified using composite scores derived from the Neuropsychological Battery used in Ace (NBACE) (26, 27), covering five cognitive domains: attention, memory, visuospatial/visuoperceptual functions, executive functions, and language (24).

2.2.2 Translational objectives

Patient well-being and intervention impact: To examine the intervention's effects on participants’ physical, psychological, social, and functional well-being, providing a holistic understanding of health, lifestyle, and care needs.

Biological and cognitive interactions: To investigate the relationship between biological markers, including genetic profiles, CSF and plasma biomarkers, and MRI measures, and cognitive decline, and to identify the conditions under which e-health interventions are most effective. Plasma biomarkers will be measured at baseline, year 1, and year 2 to track longitudinal changes.

Spontaneous speech analysis: To employ automated speech processing for analysing spontaneous speech as a remote and scalable proxy for cognitive status, complementing traditional neuropsychological assessments (23, 24).

AI platform development: To contribute to the generation of multidimensional datasets that integrate clinical, biological, cognitive, and patient-reported outcomes in order to refine and validate the VHP's AI models, and to characterize the intervention's efficacy across diverse metrics for the benefit of participants and caregivers.

2.3 Study participants

The study participants for the Ace Pilot will consist of 100 individuals aged 60–85, divided into two groups of 50 participants each (active and control groups), see Figure 1. Eligible participants will be diagnosed with MCI according to Petersen criteria (13, 28) with a Clinical Dementia Rating (CDR) score of 0.5 (29) or mild AD dementia as defined by the National Institute on Aging-Alzheimer's Association (NIA-AA) criteria (30) with a CDR score of 1 (29).

Figure 1

Figure 1. Study design and participant flow for pilot 3 at Ace.

Participants will follow a standardised protocol, with a consensus diagnosis determined by the research team based on a comprehensive evaluation, including neurological, neuropsychological, and social assessments (31). Additionally, participants must show proficiency in using digital devices such as mobile apps, tablets, or computers, as these tools are essential for both the intervention and monitoring processes.

The sample size was defined based on feasibility considerations, reflecting the expected number of eligible individuals who can be recruited during the study period. This estimate considers the typical annual flow of newly assessed patients who undergo the full diagnostic protocol, including neuropsychological evaluation, CSF and MRI, and who are able to attend regular on-site sessions at the centre. As a result, the target of 100 participants represents a realistic and achievable recruitment goal within the operational context of the Memory Unit.

In accordance with Spain's Data Protection Law (Organic Law 3/2018), all participants were informed by a neurologist about the study's goals and procedures before signing an informed consent form. Patient privacy and data confidentiality were safeguarded in compliance with applicable regulations.

Exclusion criteria include a history of medical conditions that could confound cognitive assessments or hinder participation, such as traumatic brain injury, severe depression, stroke, or brain tumors. Participants with significant uncorrected visual or auditory abnormalities, those without access to a digital device or internet connection, and individuals unable to attend regular in-person sessions at Ace will also be excluded.

2.4 Scheduled visits and assessments

Participants will be prospectively recruited at Ace's Memory Unit. The majority of patients in this unit are beneficiaries of the Catalan public health system and are typically referred by their general practitioners. Both the active and control groups will follow identical evaluation procedures, as outlined in Table 1, ensuring consistency and comparability in data collection and assessment across the study population.

Table 1

Table 1. Overview of scheduled visits and assessments in pilot 3.

Clinical, neuropsychological, and biological data will be systematically collected throughout the study to provide a comprehensive evaluation of participants’ progress and the intervention's impact. Clinical and neuropsychological assessments will be conducted in a face-to-face setting at the three critical time points: baseline (pre-intervention) and one-year and two-year follow-ups. These assessments will yield essential insights into participants’ cognitive and functional status over time, enabling a detailed analysis of the intervention's efficacy. Additionally, self-reported data will be gathered via standardised questionnaires during participants’ regular visits as part of the intervention activities. These face-to-face sessions will be carried out during the first year of the project, allowing consistent monitoring of clinical conditions, neuropsychological performances, and subjective experiences, such as quality of life and activity levels. The second-year follow-up is designed to extend the monitoring period and assess the sustainability of the intervention's effects after its conclusion, providing relevant insights into medium-term outcomes. At this final stage, the same clinical, neuropsychological, and self-reported measures are reassessed.

Biological data will also play a central role in this pilot study. Baseline collections will include genetic, CSF, and MRI data to establish a comprehensive understanding of participants’ biological profiles. Furthermore, plasma samples will be obtained not only at baseline but in the first and second year of follow-up, providing valuable longitudinal data to investigate correlations between dynamic variations in plasma biomarkers and cognitive outcomes.

To complement these assessments, spontaneous speech data will be remotely recorded every 3–4 months. This periodic speech sampling offers a scalable and non-invasive approach to monitoring cognitive changes over time and provides an additional source of information alongside face-to-face assessments and self-reported outcomes.

2.4.1 Neurological evaluation

The assessment consists of a structured anamnesis conducted with both the patient and a caregiver (informant). A comprehensive neurological examination is performed, including a detailed medical history alongside the administration of several standardised tests to assess cognitive, functional, and neuropsychiatric status. These include the Mini-Mental State Examination (MMSE) (32), the Memory test of the Spanish version of the 7 Min Screen Test (33), the Hachinski Ischemia Scale (34), the Spanish version of the Neuropsychiatric Inventory Questionnaire (35), the Global Deterioration Scale (GDS) (36), the CDR (29), and the Blessed Scale (37).

Additionally, specific comorbidities are systematically assessed, including neurological, psychiatric, rheumatologic, endocrine-metabolic, cardiovascular, respiratory, nephrological, and autoimmune conditions. Other relevant factors, such as history of alcoholism, prior oncologic diagnoses, and traumatic brain injury, are also recorded. A detailed medication registry is systematically maintained, documenting all medications taken by participants. The assessment includes the Cardiovascular Risk Factors, Aging, and Incidence of Dementia (CAIDE) score (38) and the Cumulative Illness Rating Scale (CRIS) (39). These variables are not only critical for characterizing the patient population, but also for inclusion in exploratory analyses to examine potential interactions and confounding effects on the study outcomes.

2.4.2 Neuropsychological assessment

The NBACE is a comprehensive tool designed to assess multiple cognitive, emotional, and functional domains. Table 2 provides an overview of all domains and their corresponding tests, offering a clear and concise summary of the full assessment structure. This table serves as the primary reference for the components included in the NBACE.

Table 2

Table 2. Neuropsychological battery used in Ace (NBACE), assessment structure across cognitive and functional domains.

2.4.3 Nursing evaluation

The evaluation will be performed at the nursing station at Ace and includes the collection of anthropometric data such as weight, height, abdominal circumference, and blood pressure, providing key indicators of participants’ physical health. Blood samples are collected for genetic analyses at baseline and for metabolic and proteomic studies at both baseline and follow-up assessments. Additionally, the study incorporates validated tools to assess health risks and comorbidities. Together, these measures ensure a comprehensive evaluation of participants’ biological and physical health profiles, supporting the exploration of potential associations between these factors and cognitive decline.

2.4.4 MRI acquisition and preprocessing

A Siemens MAGNETOM VIDA 3T scanner (Erlangen, Germany) equipped with a 32-channel head coil will be utilised at Clínica Corachan facilities in Barcelona. Anatomical images for volumetric analyses will be acquired using a T1-weighted 3D MPRAGE sequence, providing high-resolution structural data. Diffusion tensor imaging will be performed using an EPI sequence with 64 diffusion directions to assess white matter integrity. Additional sequences will be employed to detect ischemic damage and microbleeds, enabling a comprehensive evaluation of vascular contributions to cognitive impairment. All acquired images will be processed and analysed at the Neuroimaging Lab at Ace, ensuring consistency and quality in image interpretation and data extraction.

2.4.5 Fluid biomarkers acquisition and measurements

Several biospecimen samples are collected on the same day, forming part of the biobank at Ace, which is officially registered with Instituto de Salud Carlos III (code C.0000299). Ace's samples collection protocols have been approved by the Clinical Research Ethics Commission of the Hospital Clinic (Barcelona, Spain, reference num: HCB/2014/0494) in accordance with the Declaration of Helsinki and the current Spanish regulations in the field of biomedical research (law 14/2007, royal decree 1716/2011).

Blood samples are collected in polypropylene tubes with EDTA (BD Vacutainer). Plasma is separated by centrifugation, aliquoted, and stored at −80°C until use. CSF is obtained through lumbar puncture, performed by an experienced neurologist at Ace following established consensus recommendations. The procedure is carried out on fasting patients, who are seated and locally anesthetised with 1% subcutaneous mepivacaine. A total of 13 mL of CSF is collected using polypropylene tubes (Sarstedt Ref 62.610.018). CSF samples are centrifuged (2,000× g, 10 min, 4°C). The resulting supernatant is divided into aliquots and stored at −80°C for subsequent analyses. This collection protocol strictly adheres to the recommendations outlined by the Alzheimer's Biomarker Standardization Initiative (40), ensuring high-quality and standardised sample handling for biomarker evaluation. AT(N) classification will utilise core CSF biomarkers, including amyloid-β40 (Aβ40), amyloid-β42 (Aβ42), phosphorylated tau (pTau181), and total tau (tTau). Positivity thresholds follow the cut-offs established by the CSF program at Ace Alzheimer Center Barcelona (41), these cut-off values are based on validated population-derived distributions and have been previously applied in clinical and research cohorts to characterize neurodegenerative progression. In addition, plasma biomarkers (pTau217) are measured for research purposes (42).

2.4.6 Genotyping

Genomic DNA is extracted from buffy coat samples using the Maxwell RSC 48 instrument (Promega) with the AS1540 DNA extraction kit, ensuring high-quality samples for genetic analysis, and targeting genotype variants included in the latest Polygenic Risk Score (PRS) for AD, including APOE. APOE genotypes are determined using fluorogenic allele-specific oligonucleotide probes via the TaqMan assay (Life Technologies, Spain) for the APOE genetic variants rs7412 (Test ID: C_904973_10) and rs429358 (Test ID: C_3084793_20). For the TaqMan assays, polymerase chain reaction (PCR) and real-time fluorescence measurements are performed on the QuantStudio3 real-time PCR system (Thermo Fisher Scientific, Spain) utilizing the TaqMan Universal Master Mix system (Ref: 4364341, Life Technologies, Spain), strictly following the manufacturer's protocol. The PCR protocol consists of an initial read step at 60°C for 30 s, followed by 40 cycles comprising an initial denaturation at 95°C for 10 min, a denaturation step at 95°C for 15 s, and annealing at 60°C for 1 min. The process concludes with a final post-read step at 60°C for 30 s. Genotypes are determined through the Genotyping application available on Thermo Fisher Cloud, employing cluster analysis to ensure accurate identification. This standardised methodology provides reliable APOE genotyping for downstream analyses. Additionally, DNA from the same participants is sent for genotyping using the Axiom 815K Spanish biobank array (according to the manufacturer's instructions—Axiom™ 2.0 Assay Manual Workflow, Thermo Fisher) at the Spanish National Center for Genotyping (CeGEN, Santiago de Compostela, Spain). Details on genotyping and quality control procedures are provided elsewhere (43). PRS were calculated following (44), using a list of 83 single nucleotide polymorphisms.

2.4.7 Spontaneous speech

The evaluation of spontaneous speech is an integral part of the assessment protocol provided by Punto Health (45). Ace has extensive experience in administering and processing of spontaneous speech protocols, as demonstrated in recent publications from our group (23, 24).

This assessment is designed to be conducted exclusively in a remote setting, such as the participant's home, ensuring convenience and minimizing disruptions to daily life. The protocol is administered via a mobile application that is downloaded onto the participant's or caregiver's device. The speech assessment involves a picture description task where they describe a given image for one minute, followed by a semantic category fluency task, listing as many items as possible within a given category, such as fruits or professions, within a minute. Finally, open-ended questions prompt participants to describe personal experiences and daily routines. All three tasks are completed within 3 minutes.

2.5 Standardised evaluation protocol

A common evaluation protocol has been developed within the project to ensure methodological consistency and enable meaningful cross-site comparisons (see Table 3 for details on self-administered questionnaires). This standardised approach allows for the integration of data from diverse pilot sites while preserving inter-comparability and scientific rigor.

Table 3

Table 3. Self-administered questionnaires of COMFORTage-Ace (basal, one-year and two-year follow-up).

The collection of genetic data, CSF and plasma samples, as well as MRI scans, is carried out in strict adherence to bioethics committee–approved protocols and in compliance with relevant national and international regulations, including the GDPR. Blood samples are obtained by licenced nurses in a designated clinical area, and CSF procedures are performed by a neurologist or anesthesiologist at Ace. Prior to any procedure, participants receive detailed information and are given sufficient time to ask questions, ensuring informed understanding and comfort.

Given the cognitive challenges faced by many participants, the completion of questionnaires is designed to be flexible: participants are allowed several days and may receive support from trained staff as needed. To minimize any perception of coercion and uphold ethical standards, the Participant Information Sheet explicitly states that participation is voluntary and that individuals may withdraw at any time without consequences to their clinical care. Furthermore, it is made clear that the clinical procedures performed in the context of the project are independent of the study's research objectives. This ensures full respect for participant autonomy and safeguards their rights throughout the research process.

2.6 Intervention plan

The active group (n = 50) will participate in in-person visits, using the cognitive and functional stimulation tools developed by the Consortium and completing follow-up assessments when needed. Sessions will be conducted by a neuropsychologist and delivered twice per week in one-hour format, ensuring sufficient intensity and structure to promote cognitive engagement. Participants will attend these sessions in small groups of approximately five to seven participants, allowing the neuropsychologist to combine individualised attention with the therapeutic benefits of group interaction. The content for each participant will be customised by combining recommendations from the VHP and input from the neuropsychologist conducting the sessions, who will direct the session intervention. The sessions will take place at Ace's facilities, ensuring a controlled and supportive environment tailored to participants’ needs. This structured schedule provides a consistent therapeutic framework during the first year of follow-up. Importantly, needed medication regimens will be maintained throughout the intervention, ensuring no disruption of clinical care.

The protocol will integrate individualised and group-based approaches, with activities tailored to each participant's functional abilities and progress. For participants with lower digital literacy, face-to-face support will play a pivotal role, ensuring comprehension and fostering a confident and engaging experience. Participants demonstrating sufficient motivation and engagement may have the option to continue activities at home, supported by direct and remote monitoring from assigned professionals. This ensures consistent guidance and task adaptation based on evolving needs. The digital tools will provide multiple outputs (i.e., dosage, time of exposure, performance metrics) which will be analysed during and after the intervention. These outputs will serve a dual purpose: enhancing the dynamic functionality of the VHP and providing critical data to evaluate the intervention's efficacy within this study.

In the Pilot 3 study, the tools used to achieve the functional stimulation targets were carefully selected by the researchers at Ace. They evaluated different options provided by the consortium's technology partners to identify those best suited to the specific needs and characteristics of their participants with cognitive decline. Designed for use by both participants and professionals, these tools facilitate seamless data integration and effectively support the program's goals. The tools used in this pilot study include:

Healthentia (46). Healthentia is a class IIa medical device, intended for supporting subject telemonitoring, decision-making and virtual coaching (47, 48). The collection of physiological data includes heart rate, blood pressure, oxygen saturation, and weight, directly transmitted to care providers via automated electronic means in combination with validated internet of things devices (when available for other clinical partners, not for our pilot). The information collected is further enriched with the transmission of patient's outcomes and outcome scores related to patient's health status, health-affecting factors, health-related quality of life, disease knowledge and adherence to treatment, through validated questionnaires. Participants capable of remotely interacting with the tool beyond face-to-face sessions tend to report more frequently, contributing regular (daily or weekly) data that can inform timely care plan adjustments. All these parameters are visualised (subjects-based dashboards) and mathematically treated (trends analysis, alerts) for supporting experts’ decision-making procedures.

Eligence (49). Eligence is an online platform designed for brain training through interactive games, created by specialists in neuropsychology aimed at individuals in the early stages of dementia, for rehabilitation and prevention of cognitive decline. It features 40+ games and activities that exercise the basic cognitive skills such as memory, reasoning, attention, orientation, executive functions, construction skills, language. Through Eligence, healthcare professionals can create personalised training sessions for individuals and groups, monitor performance in real-time and easily generate detailed evaluation reports. It is accessible from anywhere and any device with an internet connection.

Linguistic Games (50). A novel cognitive and linguistic tool has been designed upon the main objective to exploit multiple cognitive processes that are involved in language performance and speech production and comprehension. The initial functionality of the mock-up tool was in 2D, but also some subtasks have been designed in 3D to introduce higher cognitive-processing demands; however, the tool is not intended as different versions for separate groups but rather as graded levels of difficulty applicable across healthy adults, adults with subjective cognitive decline, and those with mild cognitive impairment. These novel linguistic tests can assess different cognitive abilities through the vehicle of language competence such as attention, memory, language, psychomotor abilities, processing speed and executive functions. The linguistic mock-up has been designed based on Usability and Accessibility Guidelines (51–55), while realistic contexts from older adults’ daily routines have been incorporated for example tasks based on vocabulary or short scenarios related to household chores, daily mobility, travel planning or participation in social events (56).

The data generated by these three digital tools, as well as the clinical, neuropsychological, or biological data relevant to the project—whether retrospective (historical data from Ace) or prospective (related to the follow-up of the two study groups)—will be shared with the Consortium using the VHP. These multimodal data will feed into the platform and serve as input for AI algorithms and PDTs, which aim to predict individual health trajectories and propose tailored interventions. This process exemplifies the integration of digital tools, clinical knowledge, and computational models into a unified, adaptive care system. Audio recordings from the spontaneous speech protocol will not be shared for confidentiality reasons.

The active group will complete the cognitive and functional stimulation intervention at the one-year follow-up. In parallel, the control group will follow the same assessment protocol as the active group, ensuring consistency and comparability in the assessment process. However, participants in the control group will not participate in the stimulation activities provided through the digital platforms developed by the project.

2.7 Analysis plan

The primary objective of Pilot 3 is to compare cognitive decline between the active and control groups after a two-year follow-up period. The neuropsychological structure underlying the primary outcome has been previously defined and validated by an expert panel of neuropsychologists (26, 27). Based on this structure, five composite scores will be derived using Structural Equation Modelling, following the methodology described in (23). To evaluate the impact of the intervention across two conditions on these neuropsychological composites, mixed models will be employed. These models will include three time points (baseline, one-year follow-up and two-year follow-up) and will also be applied to secondary outcome measures, such as quality of life, physical activity, and satisfaction with the intervention. Alterations in plasma biomarkers, including pTau181 and pTau217, will adhere to the same analytical framework, allowing for a comprehensive evaluation of both cognitive and biological markers.

In addition, the study will explore modulating effects, such as biological factors (AD-related biomarkers in CSF, PRS, APOE, MRI parameters, plasma biomarkers, comorbidities, and medication use) through interaction analyses within the mixed models. Associations between spontaneous speech measures, neuropsychological composites, and plasma biomarkers will be examined using regression models and advanced machine learning techniques. This aspect of the study aims to identify the most predictive parameters of pre- and post-intervention slopes across the five neuropsychological composites. By leveraging these predictive models, the study seeks to advance the understanding of how specific variables contribute to cognitive decline and its modulation, ultimately informing precision medicine approaches for dementia care.

Building upon this analytical framework, a comprehensive ML approach will be integrated to enhance predictive modelling and improve the understanding of cognitive decline in response to intervention. In contrast to earlier work that relied primarily on shallow or ensemble learning methods applied to acoustic features, the present study will evaluate a broader range of modelling strategies. Acoustic, prosodic, and temporal speech features will be analysed alongside linguistic features derived from automatic transcription, including lexical, semantic, syntactic, fluency-related, and sentiment-related measures (57). Deep-learning architectures—such as convolutional neural networks and autoencoder-based models—will be explored for both acoustic representations and linguistic embeddings. The potential use of large language models (LLMs) will also be examined to extract contextual and semantic information from transcripts (58).

To manage the complexity of high-dimensional data, feature engineering and dimensionality reduction techniques will be applied. Specifically, wrapper-based feature selection methods and autoencoders will be tested to extract the most relevant predictors while minimizing redundancy and noise. A regression-based predictive framework will be used to evaluate the contribution of cognitive, biological, and speech-derived parameters, incorporating ensemble methods such as Random Forests and Gradient Boosting.

To improve model interpretability, explainable AI techniques, such as SHapley Additive exPlanations (SHAP) (59), will be employed to identify the key factors driving cognitive outcomes. Furthermore, longitudinal analyses will be conducted to track cognitive changes over time, enabling an assessment of individual responses to interventions and facilitating the early identification of cognitive decline. This machine learning-driven approach is expected to complement traditional statistical methods, enhancing both the accuracy and interpretability of predictive models in the context of dementia research.

In addition to these methodological goals, the study will also assess how the integration of digital health tools within the VHP supports clinicians in decision-making processes. The increased frequency, granularity, and clarity of data provided by these tools are expected to improve clinical insight, facilitate earlier detection of cognitive decline, and enable more timely and personalised interventions, thereby reinforcing trust in AI-assisted models as complementary assets in routine care.

3 Discussion

The COMFORTage project exemplifies a transformative and multidisciplinary effort to advance dementia care through the integration of cutting-edge digital technologies, biomarker-based research, and precision medicine strategies. By bringing together a consortium of 39 organizations across 12 countries, this initiative utilises a wide range of expertise and resources, ensuring scalability and external validity of its findings across various healthcare systems.

One of the project's distinguishing features lies in its advanced technological infrastructure. The VHP integrates multidimensional health data, encompassing all key aspects of patient information, supported by the incorporation of PDTs to enhance precision and personalization (60). These strategies enable adaptive decision-making, optimize interventions in real-time, and provide a secure, scalable framework for patient-centred care. Features such as explainable AI, gamified interventions, and educational components further improve engagement and usability (61, 62).

Although the COMFORTage project addresses a wide range of age-related conditions and populations, Pilot 3, led by Ace, focuses specifically on individuals with MCI and mild AD dementia. This pilot assesses the effectiveness of tailored cognitive and functional stimulation interventions, aiming to enhance patient outcomes and inform scalable care strategies. Through a randomised, open-label trial design, participants are divided into active intervention and control groups to rigorously assess outcomes.

By combining retrospective datasets—including thousands of clinical, neuropsychological, genetic, and imaging records—with prospective longitudinal assessments, Pilot 3 will generate a rich dataset for predictive modelling. These assessments incorporate both traditional biomarkers, such as plasma and MRI, and non-invasive biomarkers, including spontaneous speech analysis, which detects subtle cognitive changes through linguistic parameters such as lexical fluency and syntactic complexity (23, 63). Additionally, genetic variables such as PRS and the presence of the APOE ε4 allele are included, further enriching the dataset and enhancing its predictive capabilities (43).

Findings for Pilot 3 are contextualised within the broader COMFORTage framework, which encompasses complementary pilots addressing other aspects of age-related health challenges. These include systemic approaches to care delivery, physical health and mobility, and prevention strategies targeting broader ageing populations. Together, these efforts refine precision medicine approaches, ensuring interventions are not only effective but also scalable and adaptable to various healthcare systems and patient demographics (64, 65). Although the Ace dataset will include 100 participants, its contribution gains substantial relevance when combined with the broader multimodal datasets generated across the COMFORTage consortium. These larger and more diverse cohorts provide a powerful framework for validating AI-driven models and for situating the Ace data within a wider European context. Importantly, the consortium-wide datasets complement established public cohorts—such as ADNI and OASIS (66, 67)—by contributing real-world clinical information, fluid biomarkers, spontaneous speech recordings, and digital-platform activity collected within community-based intervention settings.

Studies suggest that multimodal stimulation programs combining cognitive, social, and digital activities may stabilise or modestly enhance cognitive trajectories in individuals with MCI and early AD, particularly in executive and attentional domains (68, 69). Likewise, plasma biomarkers show strong associations with longitudinal cognitive decline and may be expected to correlate with neuropsychological slopes (70). MRI-based measures have also demonstrated additive predictive value when integrated with neuropsychological data (71). Given these trends, we anticipate that the multimodal dataset generated in Pilot 3 will support more accurate, personalised prediction models within the VHP.

Despite its innovative and comprehensive design, the COMFORTage project faces several key challenges that may influence its overall impact and scalability (72). Among these, the digital divide remains a significant concern, particularly in terms of unequal access to technology and varying levels of digital literacy among older adults. To mitigate this risk, structured onboarding sessions and guided pre-training may be used to help participants become familiar with digital platforms, reducing early practice effects and enhancing long-term engagement.

Likewise, regional differences in healthcare infrastructure and resource availability may affect the feasibility of implementing standardised digital platforms at scale. Another important consideration involves ethical and legal aspects related to the management of sensitive data, especially when dealing with genetic information and biomarker profiles. Ensuring data privacy, patient consent, and compliance with regulatory frameworks like GDPR is crucial to maintain public trust and protect participants’ rights.

From a methodological perspective, Pilot 3 incorporates several overlapping digital interventions—such as Healthentia, Eligence, and the Linguistic Games tools—that deliver cognitive stimulation, functional training, and patient monitoring. Although this multimodal approach enhances the realism and relevance of the intervention, it also poses analytical challenges: it is difficult to disentangle the specific contribution of each platform to the observed outcomes. Improvements in cognitive function or quality of life may stem from the synergistic effects of multiple factors, including physical activity, digital engagement, social interaction, and cognitive challenge.

This limitation is intrinsic to many complex, integrative interventions and reflects the multifactorial nature of neurodegenerative conditions. While the study does not allow for precise attribution of effects to single components, it aligns with the project's overarching goal: to promote holistic, patient-centred care through scalable, real-world strategies. Importantly, the availability of extensive clinical, demographic, and biological data enables us to account for alternative explanations and rule out several potential confounders when interpreting the observed effects. Future research may benefit from experimental designs such as factorial or dismantling trials to further isolate and quantify the contributions of individual intervention elements within broader care models.

In conclusion, while COMFORTage addresses a diverse array of age-related health challenges, Pilot 3's emphasis on cognitive and functional stimulation interventions, supported by advanced biomarker analyses, speech-derived metrics, and digital health tools, is expected to make a significant contribution to the project's overarching goals. By integrating multimodal data within a precision-medicine framework, this pilot is well positioned to inform scalable, evidence-based strategies for early intervention in cognitive decline. Collectively, the interconnected pilots offer a cohesive foundation for future innovation in dementia care, supporting real-world implementation across diverse populations and healthcare settings.

Author contributions

SV: Conceptualization, Writing – original draft, Writing – review & editing. AM: Writing – original draft, Writing – review & editing. JB-F: Writing – review & editing. BC: Writing – review & editing. MA: Conceptualization, Writing – review & editing. AS: Writing – review & editing. GM: Conceptualization, Writing – review & editing. AA: Conceptualization, Writing – review & editing. NA: Writing – review & editing. DP: Writing – review & editing. SS: Writing – review & editing. AM: Writing – review & editing. NT: Writing – review & editing. MR-R: Writing – review & editing. AC: Writing – review & editing. MF: Writing – review & editing. PS-C: Writing – review & editing. MG: Conceptualization, Writing – review & editing. LT: Writing – review & editing. MB: Conceptualization, Writing – review & editing. AR: Conceptualization, Writing – review & editing. MM: Conceptualization, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the European Union (Grant No 101137301) and Innovative UK (Grant No 10103541). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Health and Digital Executive Agency (HADEA). Neither the European Union nor the granting authority can be held responsible for them.

COMFORTage consortium members

Please see the Supplementary Material for consortium members.

Acknowledgments

The authors acknowledge the contributions of the developers of the digital platforms integrated into this study.

Conflict of interest

AR is member of the scientific advisory board of Landsteiner Genmed and Grifols SA. AR has stocks of Landsteiner Genmed. MB has consulted for Araclon, Avid, Grifols, Lilly, Nutricia, Roche, Eisai and Servier. She received fees from lectures and funds for research from Araclon, Biogen, Grifols, Nutricia, Roche and Servier. She reports grants/research funding from Abbvie, Araclon, Biogen Research Limited, Bioiberica, Grifols, Lilly, S.A, Merck Sharp & Dohme, Kyowa Hakko Kirin, Laboratorios Servier, Nutricia SRL, Oryzon Genomics, Piramal Imaging Limited, Roche Pharma SA, and Schwabe Farma Iberica SLU, all outside the submitted work. She has not received personal compensations from these organizations. MM has consulted for F. Hoffmann-La Roche Ltd and is a member of the Scientific Advisory Board of Biomarkers of Araclon. AA was employed by company Stremble Ventures LTD. NA was employed by company Innovation Sprint srl. DP was employed by company Maggioli Spa.

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fdgth.2025.1633507/full#supplementary-material

References

1. Sabbagh MN, Perez A, Holland TM, Boustani M, Peabody SR, Yaffe K, et al. Primary prevention recommendations to reduce the risk of cognitive decline. Alzheimers Dement. (2022) 18(8):1569–79. doi: 10.1002/alz.12535

PubMed Abstract | Crossref Full Text | Google Scholar

2. Kivipelto M, Mangialasche F, Ngandu T. Lifestyle interventions to prevent cognitive impairment, dementia and alzheimer disease. Nat Rev Neurol. (2018) 14(11):653–66. doi: 10.1038/s41582-018-0070-3

PubMed Abstract | Crossref Full Text | Google Scholar

3. de Morais Fabrício D, Chagas MHN, Diniz BS. Frailty and cognitive decline. Transl Res. (2020) 221:58–64. doi: 10.1016/j.trsl.2020.01.002

Crossref Full Text | Google Scholar

4. Sheng ZH, Liu JY, Ma JY, Mi YC, Wang H, Guo F, et al. Frailty increases the risk of alzheimer’s disease in non-demented individuals: a longitudinal cohort study. J Alzheimers Dis. (2025) 103(4):1023–35. doi: 10.1177/13872877241309081

PubMed Abstract | Crossref Full Text | Google Scholar

5. 2025 alzheimer’s disease facts and figures. Alzheimers Dement. (2025) 21(4):e70235. doi: 10.1002/alz.70235

Crossref Full Text | Google Scholar

6. Nichols E, Steinmetz JD, Vollset SE, Fukutaki K, Chalek J, Abd-Allah F, et al. 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. (2022) 7(2):e105–25. doi: 10.1016/S2468-2667(21)00249-8

PubMed Abstract | Crossref Full Text | Google Scholar

7. Europe A. Dementia in Europe Yearbook 2019: Estimating the Prevalence of Dementia in Europe Luxembourg. Luxemburg: Alzheimer Europe (2019).

Google Scholar

8. Wimo A, Seeher K, Cataldi R, Cyhlarova E, Dielemann JL, Frisell O, et al. The worldwide costs of dementia in 2019. Alzheimers Dement. (2023) 19(7):2865–73. doi: 10.1002/alz.12901

PubMed Abstract | Crossref Full Text | Google Scholar

9. Petersen RC. Mild cognitive impairment. Continuum (Minneap Minn). (2004) 10(1):9–28. doi: 10.1212/01.CON.0000293545.39683.cc

Crossref Full Text | Google Scholar

10. Hampel H, Toschi N, Babiloni C, Baldacci F, Black KL, Bokde ALW, et al. Revolution of alzheimer precision neurology. Passageway of systems biology and neurophysiology. J Alzheimers Dis. (2018) 64(s1):S47–S105. doi: 10.3233/JAD-179932

PubMed Abstract | Crossref Full Text | Google Scholar

11. Petersen R, Caracciolo B, Brayne C, Gauthier S, Jelic V, Fratiglioni LF. Mild cognitive impairment: a concept in evolution. J Intern Med. (2014) 275(3):214–28. doi: 10.1111/joim.12190

PubMed Abstract | Crossref Full Text | Google Scholar

12. Jack CRJ, Therneau TM, Weigand SD, Wiste HJ, Knopman DS, Vemuri P, et al. Prevalence of biologically vs clinically defined alzheimer spectrum entities using the national institute on aging–alzheimer’s association research framework. JAMA Neurol. (2019) 76(10):1174–83. doi: 10.1001/jamaneurol.2019.1971

PubMed Abstract | Crossref Full Text | Google Scholar

13. Espinosa A, Alegret M, Valero S, Vinyes-Junqué G, Hernández I, Mauleón A, et al. A longitudinal follow-up of 550 mild cognitive impairment patients: evidence for large conversion to dementia rates and detection of major risk factors involved. J Alzheimers Dis. (2013) 34(3):769–80. doi: 10.3233/JAD-122002

PubMed Abstract | Crossref Full Text | Google Scholar

14. COMFORTage. Available online at: https://comfortage.eu (Accessed February 2025).

15. Haraldsen IH, Hatlestad-Hall C, Marra C, Renvall H, Maestú F, Acosta-Hernández J, et al. Intelligent digital tools for screening of brain connectivity and dementia risk estimation in people affected by mild cognitive impairment: the AI-mind clinical study protocol. Front Neurorobot. (2024) 17:1289406. doi: 10.3389/fnbot.2023.1289406

PubMed Abstract | Crossref Full Text | Google Scholar

16. Saunders S, Gregory S, Clement MHS, Birck C, van der Geyten S, Ritchie CW. The European prevention of alzheimer’s dementia programme: an innovative medicines initiative-funded partnership to facilitate secondary prevention of alzheimer’s disease dementia. Front Neurol. (2022) 13:1051543. doi: 10.3389/fneur.2022.1051543

PubMed Abstract | Crossref Full Text | Google Scholar

17. Kivipelto M, Mangialasche F, Snyder HM, Allegri R, Andrieu S, Arai H, et al. World-wide FINGERS network: a global approach to risk reduction and prevention of dementia. Alzheimers Dement. (2020) 16:1078–94. doi: 10.1002/alz.12123

PubMed Abstract | Crossref Full Text | Google Scholar

18. lethe. Available online at: https://www.lethe-project.eu/ (Accessed February 2025).

19. Manias G, Akker H, Azqueta-Alzúaz A, Burgos D, Capocchiano N, Crespo B, et al. iHELP: personalised health monitoring and decision support based on artificial intelligence and holistic health records. IEEE Symposium on Computers and Communications (ISCC); Athens, Greece (2021). p. 1–8

Google Scholar

20. Rodriguez-Gomez O, Sanabria A, Perez-Cordon A, Sanchez-Ruiz D, Abdelnour C, Valero S, et al. FACEHBI: a prospective study of risk factors, biomarkers and cognition in a cohort of individuals with subjective cognitive decline. Study rationale and research protocols. J Prev Alzheimers Dis. (2017) 4(2):100–8. doi: 10.14283/jpad.2016.122

PubMed Abstract | Crossref Full Text | Google Scholar

21. Hunter TR, Santos LE, Tovar-Moll F, De Felice FG. Alzheimer’s disease biomarkers and their current use in clinical research and practice. Mol Psychiatry. (2024) 30(1):272–84. doi: 10.1038/s41380-024-02709-z

PubMed Abstract | Crossref Full Text | Google Scholar

22. Wang Y, Chen HJ, Cheng Y, Xie Y, Cheng Y, Zhao S, et al. Multimodal integration of plasma biomarkers, MRI, and genetic risk to predict cerebral amyloid burden in alzheimer’s disease. Neuroimage. (2025) 322:121550. doi: 10.1016/j.neuroimage.2025.121550

PubMed Abstract | Crossref Full Text | Google Scholar

23. García-Gutiérrez F, Alegret M, Marquié M, Muñoz N, Ortega G, Cano A, et al. Unveiling the sound of the cognitive status: machine learning-based speech analysis in the alzheimer’s disease spectrum. Alzheimers Res Ther. (2024) 16(1):26. doi: 10.1186/s13195-024-01394-y

Crossref Full Text | Google Scholar

24. García-Gutiérrez F, Marquié M, Muñoz N, Alegret M, Cano A, de Rojas I, et al. Harnessing acoustic speech parameters to decipher amyloid status in individuals with mild cognitive impairment. Front Neurosci. (2023) 17:1221401. doi: 10.3389/fnins.2023.1221401

PubMed Abstract | Crossref Full Text | Google Scholar

25. García AM, de Leon J, Tee BL, Blasi DE, Gorno-Tempini ML. Speech and language markers of neurodegeneration: a call for global equity. Brain. (2023) 46(12):4870–9. doi: 10.1093/brain/awad253

Crossref Full Text | Google Scholar

26. Alegret M, Espinosa A, Vinyes-Junqué G, Valero S, Hernández I, Tárraga L, et al. Normative data of a brief neuropsychological battery for Spanish individuals older than 49. J Clin Exp Neuropsychol. (2012) 34(2):209–19. doi: 10.1080/13803395.2011.630652

PubMed Abstract | Crossref Full Text | Google Scholar

27. Alegret M, Espinosa A, Valero S, Vinyes-Junque G, Ruiz A, Hernandez I, et al. Cut-off scores of a brief neuropsychological battery (NBACE) for Spanish individual adults older than 44 years old. PLoS One. (2013) 8(10):e76436. doi: 10.1371/journal.pone.0076436

PubMed Abstract | Crossref Full Text | Google Scholar

28. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment: clinical characterization and outcome. Arch Neurol. (1999) 56(3):303–8. doi: 10.1001/archneur.56.3.303

PubMed Abstract | Crossref Full Text | Google Scholar

29. Morris JC. The clinical dementia rating (CDR): current version and scoring rules. Neurology. (1993) 61(2):165–70. doi: 10.1212/WNL.43.11.2412-a

Crossref Full Text | Google Scholar

30. McKhann GMJ, Knopman DS, Chertkow H, Hyman BT, Jack CRJ, Kawas CH, et al. The diagnosis of dementia due to alzheimer’s disease: recommendations from the national institute on aging-alzheimer’s association workgroups on diagnostic guidelines for alzheimer’s disease. Alzheimers Dement. (2011) 7(3):263–9. doi: 10.1016/j.jalz.2011.03.005

PubMed Abstract | Crossref Full Text | Google Scholar

31. Boada M, Tárraga L, Hernández I, Valero S, Alegret M, Ruiz A, et al. Design of a comprehensive alzheimer’s disease clinic and research center in Spain to meet critical patient and family needs. Alzheimers Dement. (2014) 10(3):409–15. doi: 10.1016/j.jalz.2013.03.006

PubMed Abstract | Crossref Full Text | Google Scholar

32. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. (1975) 12(3):189–98. doi: 10.1016/0022-3956(75)90026-6

PubMed Abstract | Crossref Full Text | Google Scholar

33. del Ser T, Sanchez F, Yébenes M, Otero A, Zunzunegui MV, Muñoz D. Spanish version of the 7 minute screening neurocognitive battery. Normative data of an elderly population sample over 70. Neurologia. (2004) 19(7):344–58.15273881

PubMed Abstract | Google Scholar

34. Hachinski V, Lassen NA, Marshall JC. Multi-infarct dementia. A cause of mental deterioration in the elderly. Lancet. (1974) 2(7874):207–10. doi: 10.1016/S0140-6736(74)91496-2

PubMed Abstract | Crossref Full Text | Google Scholar

35. Boada M, Cejudo JC, Tarraga Mestre L, López OL, Kaufer D. Neuropsychiatric inventory questionnaire (NPI-Q): Spanish validation of an abridged form of the neuropsychiatric inventory (NPI). Neurologia. (2002) 17(6):317–23.12084358

PubMed Abstract | Google Scholar

36. Reisberg B, Ferri SH, Leon M, Crook T. The global deterioration scale for assessment of primary degenerative dementia. Neurologia. (1982) 139(9):1136–9.

Google Scholar

37. Blessed G, Tomlinson BE, Roth M. The association between quantitative measures of dementia and of senile change in the cerebral grey matter of elderly subjects. Br J Psychiatry. (1968) 114(512):797–811. doi: 10.1192/bjp.114.512.797

PubMed Abstract | Crossref Full Text | Google Scholar

38. Kivipelto M, Ngandu T, Laatikainen T, Winblad B, Soininen H, Tuomilehto J. Risk score for the prediction of dementia risk in 20 years among middle aged people: a longitudinal, population-based study. Lancet Neurol. (2006) 5(9):735–41. doi: 10.1016/S1474-4422(06)70537-3

PubMed Abstract | Crossref Full Text | Google Scholar

39. Linn BS, Linn MW, Gurel L. Cumulative illness rating scale. J Am Geriatr Soc. (1968) 16(5):622–6. doi: 10.1111/j.1532-5415.1968.tb02103.x

PubMed Abstract | Crossref Full Text | Google Scholar

40. Molinuevo JL, Blennow K, Dubois B, Engelborghs S, Lewczuk P, Perret-Liaudet A, et al. The clinical use of cerebrospinal fluid biomarker testing for alzheimer’s disease diagnosis: a consensus paper from the alzheimer’s biomarkers standardization initiative. Alzheimers Dement. (2014) 10(6):808–17. doi: 10.1016/j.jalz.2014.03.003

PubMed Abstract | Crossref Full Text | Google Scholar

41. Orellana A, García-González P, Valero S, Montrreal L, de Rojas I, Hernández I, et al. Establishing in-house cutoffs of CSF alzheimer’s disease biomarkers for the AT(N) stratification of the alzheimer center Barcelona cohort. Int J Mol Sci. (2022) 23(13):6891. doi: 10.3390/ijms23136891

PubMed Abstract | Crossref Full Text | Google Scholar

42. Cano A, Capdevila M, Puerta R, Arranz J, Montrreal L, de Rojas I, et al. Clinical value of plasma pTau181 to predict alzheimer’s disease pathology in a large real-world cohort of a memory clinic. EBioMedicine. (2024) 108:105345. doi: 10.1016/j.ebiom.2024.105345

PubMed Abstract | Crossref Full Text | Google Scholar

43. Moreno-Grau S, De Rojas I, Hernández I, Quintela I, Montrreal L, Alegret M, et al. Genome-wide association analysis of dementia and its clinical endophenotypes reveal novel loci associated with alzheimer’s disease and three causality networks: the GR@ACE project. Alzheimers Dement. (2019) 15(10):1333–47. doi: 10.1016/j.jalz.2019.06.4950

PubMed Abstract | Crossref Full Text | Google Scholar

44. Bellenguez C, Küçükali F, Jansen IE, Kleineidam L, Moreno-Grau S, Amin N, et al. New insights into the genetic etiology of Alzheimer’s disease and related dementias. Nat Genet. (2022) 54(4):412–36. doi: 10.1038/s41588-022-01024-z

PubMed Abstract | Crossref Full Text | Google Scholar

45. punto. Available online at: https://puntohealth.com/ (Accessed February 2025).

46. Healthentia. Available online at: https://healthentia.com/ (Accessed February 2025).

47. Pnevmatikakis A, Kanavos S, Matikas G, Kostopoulou K, Cesario A, Kyriazakos S. Risk assessment for personalized health insurance based on real-world data. Risks. (2021) 9(3):46. doi: 10.3390/risks9030046

Crossref Full Text | Google Scholar

48. Kyriazakos S, Pnevmatikakis A, Kostopoulou K, Ferrière L, Thibaut K, Giacobini E, et al. Benchmarking the clinical outcomes of healthentia SaMD in chronic disease management: a systematic literature review comparison. Front Public Health. (2024) 12:1488687. doi: 10.3389/fpubh.2024.1488687

PubMed Abstract | Crossref Full Text | Google Scholar

49. eligence. Available online at: https://eligence.eu/portal/index.php/en/home-english/ (Accessed February 2025).

50. Hellas CCfR&T. Linguistic Games. Thessaloniki: Information Technologies Institute (ITI), Centre for Research & Technology Hellas (CERTH) (2025). Available online at: https://linguistic-games.dev.iti.gr./home (Accessed February 2025).

Google Scholar

51. Huang M, McGrew G, Codinge C. Social Media apps: a paradigm for examining usability of mobile apps for working-age adults with mild-moderate cognitive disabilities. ACM Trans Access Comput. (2025) 18(1):1–38. doi: 10.1145/3709144

Crossref Full Text | Google Scholar

52. IEEE. Usability and Accessibility. New York, NY: Institute of Electrical and Electronics Engineers (IEEE) (2021). Available online at: https://brand-experience.ieee.org/guidelines/digital/style-guide/usability-and-accessibility/

Google Scholar

53. ISO. ISO 9241-11:2018(en) Ergonomics of Human-System Interaction — Part 11: Usability: Definitions and Concepts. Geneva: International Organization for Standardization (ISO) (2018). Available online at: https://www.iso.org/obp/ui/#iso:std:iso:9241:-11:ed-2:v1:en

Google Scholar

54. Guidelines WCA. Guidelines. Doetinchem: WCAG.eu (2024). Available online at: https://wcag.eu/knowledge/guidelines/

Google Scholar

55. (WAI) WAI. WCAG 2 Overview. Cambridge: World Wide Web Consortium (W3C) (2005). Available online at: https://www.w3.org/WAI/standards-guidelines/wcag/

Google Scholar

56. Meng J. Enhancing older Adults’ motivation for social interaction: exploring design principles of social Media Mobile applications through a self-determination theory (SDT) approach. In: Gao Q, Zhou J, editors. Human Aspects of IT for the Aged Population. Washington, DC: Springer Nature Switzerland (2024).

Google Scholar

57. Jahan Z, Bhatia Khan S, Saraee M. Early dementia detection with speech analysis and machine learning techniques. Discov Sustain. (2024) 5:65. doi: 10.1007/s43621-024-00217-2

Crossref Full Text | Google Scholar

58. Ding K, Chetty M, Noori Hoshyar A, Bhattacharya T, Klein B. Speech based detection of alzheimer’s disease: a survey of AI techniques, datasets and challenges. Artif Intell Rev. (2024) 57(325). doi: 10.1007/s10462-024-10961-6

PubMed Abstract | Crossref Full Text | Google Scholar

59. Lundberg SM, Lee SI. A unified approach to interpreting model predictions. NIPS’17: Proceedings of the 31st International Conference on Neural Information Processing Systems; USA: Long Beach California (2017). p. 4768–77

Google Scholar

60. Vallée A. Digital twin for healthcare systems. Front Digit Health. (2023) 7(5):1253050. doi: 10.3389/fdgth.2023.1253050

Crossref Full Text | Google Scholar

61. Malek M, Nitsche J, Dinand C, Ehlers J, Lissek V, Böhm P, et al. Interprofessional needs analysis and user-centred prototype evaluation as a foundation for building individualized digital education in dementia healthcare supported by artificial intelligence: a study protocol. Healthcare. (2023) 11(10):1508. doi: 10.3390/healthcare11101508

PubMed Abstract | Crossref Full Text | Google Scholar

62. Berglund A, Jaarsma T, Berglund E, Strömberg A, Klompstra L. Understanding and assessing gamification in digital healthcare interventions for patients with cardiovascular disease. Eur J Cardiovasc Nurs. (2022) 21(6):630–8. doi: 10.1093/eurjcn/zvac048

PubMed Abstract | Crossref Full Text | Google Scholar

63. Chlasta K, Wolk K. Towards computer-based automated screening of dementia through spontaneous speech. Front Psychol. (2021) 11:623237. doi: 10.3389/fpsyg.2020.623237

PubMed Abstract | Crossref Full Text | Google Scholar

64. Palapar L, Blom JW, Wilkinson-Meyers L, Lumley T, Kerse N. Preventive interventions to improve older people’s health outcomes: systematic review and meta-analysis. Br J Gen Pract. (2024) 74(741):e208–18. doi: 10.3399/BJGP.2023.0180

PubMed Abstract | Crossref Full Text | Google Scholar

65. Alruwaili MM, Shaban M, Elsayed Ramadan OM. Digital health interventions for promoting healthy aging: a systematic review of adoption patterns, efficacy, and user experience. Sustainability. (2023) 15(23):16503. doi: 10.3390/su152316503

Crossref Full Text | Google Scholar

66. Petersen RC, Aisen PS, Beckett LA, Donohue MC, Gamst AC, Harvey DJ, et al. Alzheimer’s disease neuroimaging initiative (ADNI). Neurology. (2010) 74(3):201–9. doi: 10.1212/WNL.0b013e3181cb3e25

PubMed Abstract | Crossref Full Text | Google Scholar

67. Medicine W. Open Access Series of Imaging Studies (OASIS). St. Louis: Washington University in St. Louis (2007). Available online at: https://sites.wustl.edu/oasisbrains/

Google Scholar

68. Gkintoni E, Vassilopoulos SP, Nikolaou G, Vantarakis A. Neurotechnological approaches to cognitive rehabilitation in mild cognitive impairment: a systematic review of neuromodulation, EEG, virtual reality, and emerging AI applications. Brain Sci. (2025) 15(6):582. doi: 10.3390/brainsci15060582

PubMed Abstract | Crossref Full Text | Google Scholar

69. Paolillo EW, Saloner R, VandeBunte A, Lee S, Bennett DA, Casaletto KB. Multimodal lifestyle engagement patterns support cognitive stability beyond neuropathological burden. Alzheimers Res Ther. (2023) 15(1):221. doi: 10.1186/s13195-023-01365-9

PubMed Abstract | Crossref Full Text | Google Scholar

70. Park MK, Ahn J, Kim YJ, Lee JW, Lee JC, Hwang SJ, et al. Predicting longitudinal cognitive decline and alzheimer’s conversion in mild cognitive impairment patients based on plasma biomarkers. Cells. (2024) 13(13):1085. doi: 10.3390/cells13131085

PubMed Abstract | Crossref Full Text | Google Scholar

71. Bonarota S, Caruso G, Di Domenico C, Sperati S, Tamigi FM, Giulietti G, et al. Integration of automatic MRI segmentation techniques with neuropsychological assessments for early diagnosis and prognosis of alzheimer’s disease. A systematic review. Neuroimage. (2025) 314:121264. doi: 10.1016/j.neuroimage.2025.121264

PubMed Abstract | Crossref Full Text | Google Scholar

72. Lee C, Coughlin JF. PERSPECTIVE: older adults’ adoption of technology: an integrated approach to identifying determinants and barriers. J Prod Innov Manage. (2015) 32(5):747–59. doi: 10.1111/jpim.12176

Crossref Full Text | Google Scholar

73. Wechsler D. Wechsler Adult Intelligence Scale (WAIS-III). 3rd Ed. Washington, DC: American Psychological Association (APA) (1997).

Google Scholar

74. Wechsler D. Wechsler Preschool and Primary Scale of Intelligence (WPPSI-III). San Antonio, TX: Psychological Corp (2002).

Google Scholar

75. Goodglass H, Kaplan E, Barresi B. Boston Naming Test. 2nd Ed. Austin: Pro-Ed (2001).

Google Scholar

76. Lezak MD. Neuropsychological Assessment. USA: Oxford University Press (2004).

Google Scholar

77. Sala S, Laiacona M, Spinnler H, Trivelli C. Poppelreuter-Ghent’s overlapping figures test: its sensitivity to age, and its clinical use. Arch Clin Neuropsychol. (1995) 10(6):511–34. doi: 10.1093/arclin/10.6.511

PubMed Abstract | Crossref Full Text | Google Scholar

78. Golden CJ. In reply to Adams’s “in search of luria’s battery: a false start.”. J Consult Clin Psychol. (1980) 48(4):517–21. doi: 10.1037/0022-006X.48.4.517

PubMed Abstract | Crossref Full Text | Google Scholar

79. Pillon B, Dubois B, Bonnet AM, Esteguy M, Guimaraes J, Vigouret JM, et al. Cognitive slowing in parkinson’s disease fails to respond to levodopa treatment: the 15-objects test. Neurology. (1989) 39(6):762–8. doi: 10.1212/WNL.39.6.762

PubMed Abstract | Crossref Full Text | Google Scholar

80. Erzigkeit H. The SKT — a short cognitive performance test as an instrument for the assessment of clinical efficacy of cognition enhancers. In: Bergener M, Reisberg B, editors. Diagnosis and Treatment of senile Dementia. Heidelberg: Springer (1989). p. 164–74.

Google Scholar

81. Freedman M, Leach L, Kaplan E, Winocur G, Shulman KI, Delis DC. Clock Drawing: A Neuropsychological Analysis. New York, NY: Oxford University Press (1994).

Google Scholar

82. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. (1983) 67(6):361–70. doi: 10.1111/j.1600-0447.1983.tb09716.x

PubMed Abstract | Crossref Full Text | Google Scholar

83. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged: the index of ADL: a standardized measure of biological and psychosocial function. JAMA. (1963) 185(12):914–9. doi: 10.1001/jama.1963.03060120024016

PubMed Abstract | Crossref Full Text | Google Scholar

84. Lawton M, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. (1969) 9(3):179–86. doi: 10.1093/geront/9.3_Part_1.179

PubMed Abstract | Crossref Full Text | Google Scholar

85. Logsdon RG, Gibbons LE, McCurry SM, Teri L. Quality of life in alzheimer’s disease: patient and caregiver reports. J Ment Health Aging. (1999) 5(1):21–32.

Google Scholar

86. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res. (1982) 17(1):37–49. doi: 10.1016/0022-3956(82)90033-4

PubMed Abstract | Crossref Full Text | Google Scholar

87. Buysse DJ, Reynolds CF III, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res. (1989) 28(2):193–213. doi: 10.1016/0165-1781(89)90047-4

PubMed Abstract | Crossref Full Text | Google Scholar

88. Chin J, Diehl V, Norman K. Development of an instrument measuring user satisfaction of the human-computer interface. International Conference on Human Factors in Computing Systems (1988). p. 213–8

Google Scholar

89. Saunders JB, Aasland OG, Babor TF, De la Fuente JR, Grant M. Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption-II. Addiction. (1993) 88(6):791–804. doi: 10.1111/j.1360-0443.1993.tb02093.x

PubMed Abstract | Crossref Full Text | Google Scholar

90. Russell D, Peplau LA, Cutrona CE. The revised UCLA loneliness scale: concurrent and discriminant validity evidence. J Pers Soc Psychol. (1980) 39(3):472. doi: 10.1037/0022-3514.39.3.472

PubMed Abstract | Crossref Full Text | Google Scholar

91. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. (2003) 35(8):1381–95. doi: 10.1249/01.MSS.0000078924.61453.FB

PubMed Abstract | Crossref Full Text | Google Scholar

92. Guigoz Y, Vellas B, Garry PJ. Assessing the nutritional status of the elderly: the mini nutritional assessment as part of the geriatric evaluation. Nutr Rev. (1996) 54(1 Pt 2):S59–65. doi: 10.1111/j.1753-4887.1996.tb03793.x

PubMed Abstract | Crossref Full Text | Google Scholar