Abstract: Personalizing physical activity recommendations for older adults requires understanding not only which dimensions of physical activity and sedentary behaviors (24-h movement behaviors) influence health outcomes but also when, within an individual’s everyday life, these dimensions are most relevant. Current observational and interventional approaches rarely capture the temporal dynamics linking everyday patterns of 24-hour movement behaviors to cognitive and mental health trajectories, 2 key determinants of healthy aging. This study introduces and evaluates PEPHA (Personalized Phenotyping for Aging), an interpretable artificial intelligence (AI) framework designed to identify which dimensions of behavior and when within an observation window are strongly associated with cognitive functioning and depressive symptoms in older adults. We introduce PEPHA, an interpretable AI framework that integrates passive, high-frequency wearable data (physical activity and sleep) with periodic, active, validated cognitive and affective assessments (waves). Using longitudinal data from the Providemus alz cohort (n=67, up to 6 waves and 528 days of wearable data per person), we examined 2 key outcomes representing cognitive functioning and mental health (processing speed and depressive symptoms). PEPHA summarizes daily physical activity and sedentary patterns into slope-based temporal features, optimizes support vector regression parameters through Bayesian optimization, and applies individualized analyses including time order-swap testing and change point detection to identify “potential temporal association windows” (ie, periods within an observation window during which outcomes appear more sensitive to changes in behavioral patterns). Personalized analyses showed that roughly 40% of individuals exhibited moderate or large temporal order effects of 24-hour movement behavior in the outcomes. For 1 exemplar participant (male, above the mean sample age), we localized 2 potential temporal association windows approximately 60 days and 30 days before the assessment of his processing speed, suggesting periods during which this individual may have been more sensitive to favorable or unfavorable behavioral configurations. Across participants, PEPHA revealed distinct 24-hour movement behavioral patterns correlated with cognitive functioning and mental health. Processing speed was best explained by locomotor activity, while depressive symptoms were best explained by sedentary behavior. Five control variables (education, cognitive reserve, diet, sex, and subjective age difference) were noninformative, whereas chronological age had predictive power regarding depressive symptoms. PEPHA demonstrates that continuous passive wearable data can uncover individualized, time-specific behavioral patterns associated with cognitive functioning and mental health. Although exploratory, this framework transforms observational data into interpretable, timing-aware insights that can help identify periods of increased behavioral sensitivity or association, thereby informing future personalized, AI-supported physical activity interventions in aging.

Abstract: Longitudinal, in-the-wild, wearable sensing yields day-level physiology, sleep, activity, and environmental streams, whereas affect and cognition are labeled only episodically (per waves). We recast this cadence mismatch as a temporal representation problem and compare three wave-level mappings from dense histories to sparse labels: levels (within-wave summaries), absolute drift (change across waves), and proportional drift. Using almost a year of data from 82 adults in the Providemus alz study, we model 21 affect and cognition outcomes. Day-scale signals are reduced to compact wave-level descriptors (central tendency, dispersion, and distributional shape) and learned with four regressors under two orthogonal evaluation axes: leave-one-subject-out and leave-one-wave-out. Performance is reported as scaled MAE using both mean and median across folds. Differences emerge: affective states are best predicted by wave-to-wave absolute drift, whereas cognitive performance aligns with within-wave levels, reflecting emotion dynamic theories. Across windowing features, shape descriptors (e.g., minima, kurtosis) carry more signal than simple means/medians. We contribute a representation triad for sparse-label modelling, a wave-level feature schema applicable on-device, and a dual-axis reporting practice that separates cross-participant generalization from temporal robustness. These results convert temporal representation from an implicit preprocessing step into an explicit, testable design choice for real-world affective-computing applications in brain health.

Abstract: Passive digital health technologies (DHTs) are increasingly promoted as scalable tools for detecting Alzheimer’s disease and related dementias (ADRD) earlier than routine clinic visits. We searched six major databases for English-language studies published between January 2014 and July 2024 that used passively collected, real-world DHT data for ADRD screening or diagnosis. Thirty studies met the criteria. Population sizes were highly skewed (median = 87; range 12-82,829), and most designs were longitudinal (53%) and fully passive (68%). Wrist-worn accelerometers and photoplethysmography sensors dominated, though several studies also used gait, sleep, voice, radar, or posture-tracking devices. A cross-study synthesis showed that those two modalities were primarily applied to memory, attention, and language tasks. Nineteen studies reported median accuracy, sensitivity, specificity, and precision between 80-90%, with F1-score and AUC medians approaching 78%, though relying on in-sample cross-validation rather than external cohorts. Reference standards varied widely, data-quality criteria were seldom reported, and fewer than 5% shared datasets publicly. Classification was the predominant modeling strategy, with regression emerging only in recent years. Overall, passive DHTs show promise as low-burden triage tools for population-level ADRD screening, but routine deployment will require more diverse cohorts, harmonized reporting, multimodal privacy-preserving analytics, and rigorous human-factors evaluation.

Abstract: While there has been much discussion around the use of Artificial Intelligence (AI) for multilingual translations in other areas, recommendations pertaining specifically to the use of AI in the context of Clinical Outcome Assessment (COA) translation, linguistic validation, and electronic migration within clinical trials are lacking. Without published recommendations or guidelines, stakeholders involved in the COA translation process may be hesitant to explore or include AI. To address this gap, the AI Working Group of the ISOQOL TCA-SIG conducted a study to assess the landscape of AI in this specific context aimed at proposing recommendations for potential implementation of AI in COA translation, linguistic validation and electronic migration processes. The study consisted of three parts: (1) a literature review targeting studies using AI in COA translation; (2) a survey among relevant stakeholders assessing perceptions of AI use in COA translation; and (3) interviews with AI subject matter experts (SMEs). Survey responses were received from a total of 50 individuals from a wide variety of stakeholder groups, including COA copyright holders, representatives from pharmaceutical company COA/HEOR teams, respondents holding roles associated with the COA translation, eCOA, and AI industries, and authors of the 2005 ISPOR task force article on linguistic validation methodology. Survey data provided detailed feedback regarding the appropriateness of using AI during all reviewed process steps. Results of the literature review and AI expert interviews provided additional depth and nuance, allowing for the generation of detailed recommendations covering the use of AI within linguistic validation and eCOA migration processes. When assessing the potential use of AI tools within the linguistic validation process, it is important to consider not only the capabilities of the technology, but also the degree to which use of AI may or may not align with the spirit and intent of existing linguistic validation guidelines. The recommendations included in this manuscript are designed to balance considerations of technological capability and improved efficiency with concerns related to intellectual property protection, data privacy/security, and the goal of keeping patients at the center of outcomes research.