Sensor-Based Technologies for the Detection of Unwanted Loneliness in Older Adults: A Systematic Review.

Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Sensors (Basel) . 2026 Mar 24;26(7):2028. doi: 10.3390/s26072028 Search in PMC Search in PubMed View in NLM Catalog Add to search Sensor-Based Technologies for the Detection of Unwanted Loneliness in Older Adults: A Systematic Review María Mercedes Párraga Vico María Mercedes Párraga Vico 1 University of Jaén, 23071 Jaén, Spain; [email protected] Find articles by María Mercedes Párraga Vico 1 , Juana María Morcillo Martínez Juana María Morcillo Martínez 2 Department of Psychology, University of Jaén, 23071 Jaén, Spain; [email protected] Find articles by Juana María Morcillo Martínez 2 , Juan F Gaitán-Guerrero Juan F Gaitán-Guerrero 3 Department of Computer Science, University of Jaén, 23071 Jaén, Spain; [email protected] Find articles by Juan F Gaitán-Guerrero 3 , Juan Luis Herreros Bódalo Juan Luis Herreros Bódalo 4 Department of Telecommunication Engineering, University of Jaén, 23700 Linares, Spain; [email protected] (J.L.H.B.); [email protected] (J.C.C.M.) Find articles by Juan Luis Herreros Bódalo 4 , Macarena Espinilla Estévez Macarena Espinilla Estévez 3 Department of Computer Science, University of Jaén, 23071 Jaén, Spain; [email protected] Find articles by Macarena Espinilla Estévez 3, * , Juan Carlos Cuevas Martínez Juan Carlos Cuevas Martínez 4 Department of Telecommunication Engineering, University of Jaén, 23700 Linares, Spain; [email protected] (J.L.H.B.); [email protected] (J.C.C.M.) Find articles by Juan Carlos Cuevas Martínez 4 Editor: Toshiyo Tamura Author information Article notes Copyright and License information 1 University of Jaén, 23071 Jaén, Spain; [email protected] 2 Department of Psychology, University of Jaén, 23071 Jaén, Spain; [email protected] 3 Department of Computer Science, University of Jaén, 23071 Jaén, Spain; [email protected] 4 Department of Telecommunication Engineering, University of Jaén, 23700 Linares, Spain; [email protected] (J.L.H.B.); [email protected] (J.C.C.M.) * Correspondence: [email protected] Roles Toshiyo Tamura : Academic Editor Received 2026 Jan 28; Revised 2026 Feb 27; Accepted 2026 Mar 18; Collection date 2026 Apr. © 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license . PMC Copyright notice PMCID: PMC13075024  PMID: 41977813 Abstract Highlights What are the main findings? Passive sensor technologies combined with artificial intelligence and multimodal data fusion have great potential for detecting behavioral markers associated with unwanted loneliness and social isolation in older adults. Artificial intelligence models based on multimodal fusion achieve greater accuracy than unimodal approaches in predicting states of loneliness and isolation in older adults. What are the implications of the main findings? These technologies enable objective assessment, complementing traditional self-reporting tools. Their implementation in real-world settings requires addressing challenges of user acceptance and validation in different samples. Abstract Background: Unwanted loneliness and social isolation in older adults are public health problems with negative effects on physical and mental health. The usual assessment tools, based on self-report questionnaires, have limitations in capturing these phenomena continuously and objectively. Objective : We aimed to critically analyze recent scientific evidence on the use of passive sensor technologies combined with artificial intelligence for the detection of unwanted loneliness and social isolation in older adults. Methods: Studies were reviewed in databases (PubMed, Scopus, Web of Science, and IEEE Xplore) that used wearable devices, environmental sensors in the home, smartphones, and multimodal fusion approaches. This systematic review was conducted following the PRISMA 2020 guidelines. Results : Behavioral variables derived from passive monitoring, such as mobility, time away from home, sleep patterns, and digital interactions, are consistently associated with measures of loneliness and social isolation. Likewise, artificial intelligence models based on the combination of multiple data sources show better predictive performance than unimodal approaches. Conclusions: Sensor-based technologies can complement traditional assessment methods, although their practical application requires overcoming challenges related to methodological validation, user acceptance, and ethical considerations. Keywords: loneliness detection, older adults, passive sensing, machine learning, healthcare IoT, multimodal data fusion 1. Introduction Unwanted loneliness and social isolation are increasingly recognized as significant public health issues among older adults. Persistent loneliness has been consistently associated with adverse mental health effects, including cognitive decline, depression, sleep disturbance, anxiety, dementia, and even suicidal ideation [ 1 , 2 , 3 , 4 ]. In addition, physical health is affected, as there is evidence linking loneliness to cardiovascular disease, hypertension, stroke, frailty, obesity, and functional decline [ 1 , 2 , 3 , 4 , 5 , 6 ]. These effects are partly mediated by health-related behaviors: socially isolated people tend to be less physically active, have poorer diets, smoke more, and experience greater difficulty quitting smoking, further increasing their risk of disease [ 6 , 7 ]. Large cohort studies and meta-analyses indicate that loneliness, social isolation, or living alone may increase the risk of premature mortality by approximately 26–32% [ 8 , 9 ]. Loneliness is also associated with increased use of healthcare services, including more frequent visits to the doctor and emergency room, especially when combined with social isolation [ 5 , 10 ]. Due to their widespread impact, loneliness and social isolation have been labeled “geriatric giants” by public health authorities, underscoring the need for systematic identification and intervention [ 11 ]. Therefore, addressing these issues is a priority not only for individual well-being, but also for health systems and policy planning. Although self-report scales such as the University of California, Los Angeles (UCLA) Loneliness Scale and the Social and Emotional Loneliness Scale for Adults (SESLA) are widely used and show good internal consistency, recent methodological reviews point to limitations in their validity and ability to adequately capture loneliness [ 12 , 13 ]. Very short forms (1–3 items) are practical for large-scale surveys, but lack sufficient detail for a thorough clinical assessment [ 12 , 13 , 14 , 15 ]. In addition, self-reports are subject to social desirability bias and stigma, which may lead older adults to underestimate their feelings of loneliness [ 16 , 17 ]. Validation studies have mainly focused on younger populations with higher education or who use the Internet, limiting their generalizability to older, more frail individuals with cognitive impairment or from culturally diverse backgrounds [ 12 , 13 , 18 ]. It is important to note that subjective indicators of loneliness and objective indicators of social isolation (e.g., social network size, frequency of social contacts, time spent outside the home) are only modestly correlated, reflecting distinct constructs that should be taken into account in research on aging [ 19 , 20 ]. Traditional questionnaires provide episodic “snapshots” and are influenced by memories and current mood, making them unsuitable for capturing everyday dynamics [ 17 ]. Evidence from intelligent sensor systems in homes, known as Ambient Assisted Living (AAL), suggests that continuous, discreet monitoring of behavior—such as activity in the home, time spent outside the home, and computer or cell phone use—can complement self-reports, although sensor-based measurements do not perfectly match questionnaire scores [ 17 ]. In summary, validated loneliness questionnaires remain reliable, but in order to continuously and accurately assess social isolation in older adults, they must be complemented by objective indicators and tools adapted to each context. Advances in sensor technology have opened up new possibilities for the objective monitoring of behaviors associated with social isolation and loneliness in older adults. In general, wearable sensors (activity trackers), smartphones, and environmental sensors can complement self-assessments and clinical assessments by providing objective, continuous measurements of behaviors associated with social isolation, but they do not yet constitute independent, clinically robust detectors [ 21 , 22 ]. The incorporation of Artificial Intelligence (AI) techniques is revolutionizing the monitoring of mental and behavioral health in aging through digital phenotyping and passive detection [ 23 , 24 ]. These approaches enable continuous, real-world assessment. In particular, the fusion of multimodal data from wearables, environmental sensors, and smartphones allows for the construction of more robust predictive models that overcome the limitations of episodic clinical assessments [ 25 , 26 ]. However, recent reviews in the specific field of loneliness highlight that the translation of these systems into clinical practice and care is hampered by critical challenges of methodological standardization, validation in representative samples, interpretability, and ethical considerations [ 21 , 27 , 28 ]. Despite growing interest in sensor-based approaches, the available evidence remains fragmented, methodologically heterogeneous, and rarely focused specifically on older adults. As a result, the field is shaped by isolated studies that require a review capable of going beyond the mere description of devices and critically analyzing the ability of these technologies to infer experiences of loneliness, the potential of multimodal data fusion, and the ethical and practical barriers to their actual implementation in social and healthcare settings. Therefore, the objective of this review is to comprehensively and critically examine the recent scientific literature on the use of sensor-based and artificial intelligence technologies for the detection of unwanted loneliness and social isolation in older adults. Throughout this review, we adopt a conceptual and operational distinction between loneliness and social isolation, following established gerontological frameworks [ 19 , 20 ]. Loneliness is defined as a subjective, negative emotional state arising from a perceived discrepancy between desired and actual social relationships. It is typically measured using validated self-report scales such as the UCLA Loneliness Scale (versions 3, 20, or 3-item) or the Social and Emotional Loneliness Scale for Adults (SESLA). Social isolation is defined as an objective, quantifiable state of reduced social network size, infrequent social contacts, or limited participation in social activities. It is operationalized through indicators such as time spent outside the home, frequency of visits, living alone, or low social interaction frequency, often derived from sensor data or behavioral logs. The manuscript is structured into four main sections. Section 2 describes the materials and methods used to conduct the narrative review. Section 3 presents the results, organized around the main types of sensors, behavioral markers, and analytical approaches used to assess loneliness and social isolation in older adults. Finally, Section 4 discusses these findings in light of previous literature and possible future lines of research. 2. Materials and Methods 2.1. Study Design This systematic review is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement [ 29 ] (see Supplementary Materials ). This review was not registered. The objective is to synthesize the existing literature on sensor-based technologies for detecting unwanted loneliness and social isolation in older adults. This methodological choice is justified by the emerging and interdisciplinary nature of the field—which combines engineering, artificial intelligence, and health sciences—and by the considerable methodological heterogeneity of the available studies. 2.2. Study Selection Process The study selection process was conducted in three phases. First, two reviewers (MMPV and JFGG) independently screened titles and abstracts against the eligibility criteria. Disagreements were resolved by consensus or by consulting a third reviewer (MEE). Second, full texts of potentially eligible articles were retrieved and assessed independently by the same two reviewers. Third, references of included studies were manually screened to identify additional relevant records. Duplicates were automatically detected using Zotero 7.0.32 (64-bit) and manually verified. 2.3. Search Strategy and Information Sources The literature search was conducted in the PubMed, Scopus, Web of Science, and IEEE Xplore databases, selected for their complementary coverage of biomedical, technological, and engineering literature. The search focused on articles published between January 2017 and February 2025, a period that coincides with the rise of passive monitoring and smart home approaches applied to aging. Combinations of keywords and controlled terms related to the following were used: loneliness, social isolation, unwanted loneliness; older adults, elderly, aging population; sensors, wearables, smart home, ambient assisted living, passive sensing; machine learning, artificial intelligence, digital phenotyping. The terms were adapted to the specific syntax of each database, and Boolean operators (AND, OR) were used to combine search concepts. The search was conducted between October 2025 and December 2025. The complete search strings used for each database are provided in Appendix A . The flow of the literature search and selection process is summarized in Figure 1 , following the PRISMA 2020 guidelines [ 29 ]. A total of 950 records were identified, of which 250 were duplicates. After screening 700 titles and abstracts, 580 records were excluded. The full text of 120 reports was assessed for eligibility, of which 63 were excluded (25 no sensor-based data, 18 wrong population, 12 review articles, 8 no measurable outcome). A total of 57 studies met the inclusion criteria and were included in this review. Figure 1. Open in a new tab PRISMA 2020 flow diagram of the study selection process. 2.4. Data Extraction and Synthesis From each included study, the following information was extracted: author(s) and year of publication; sample size and population characteristics (e.g., community-dwelling, residential care, clinical condition); study setting (e.g., home, laboratory, population-based); type of sensor technology used (e.g., wearables, environmental sensors, smartphones); behavioral or digital markers extracted from sensor data; target construct (loneliness or social isolation); predictive model or analytical approach; validation method (e.g., cross-validation, hold-out, external validation); and main performance metrics (e.g., accuracy, AUC, correlation coefficients). Data extraction was performed independently by two reviewers (MMPV and JFGG), and discrepancies were resolved through discussion or consultation with a third reviewer (MEE). Due to the heterogeneity of the included studies, a structured narrative synthesis was conducted, organizing findings according to sensor types, behavioral markers, and analytical approaches, as presented in Section 3 . 2.5. Risk of Bias Assessment A formal risk of bias assessment using a standardized tool was not conducted due to the heterogeneity of the included studies and the exploratory nature of this review. However, a narrative summary of study limitations (e.g., sample sizes, validation strategies) is provided in Section 3.6 . 2.6. Eligibility Criteria and Study Selection The following inclusion criteria were established: (1) studies focusing on older adults (≥60 years); (2) use of passive detection technologies (wearables, environmental sensors, smartphones); (3) objective of detecting, predicting, or correlating with loneliness or social isolation; (4) presentation of analyzable quantitative or qualitative results on the validity or usefulness of the technology; (5) studies based solely on self-reports, systematic reviews, opinion articles, studies without validation results, and those whose main focus was not loneliness/isolation were excluded. The identified articles were organized and managed using Zotero software. Table 1 presents the detailed eligibility criteria. Table 1. Eligibility criteria. Criterion Inclusion Criteria Exclusion Criteria Population Older adults (≥60 years) Studies focused on younger adults, caregivers, or the general population without age-specific analysis Technology Passive sensing technologies (e.g., wearables, environmental sensors, smartphones) Active sensing requiring user interaction, questionnaire-only studies, or non-sensor-based methods Outcome Detection, prediction, or correlation with loneliness or social isolation Studies not reporting loneliness or social isolation as an outcome Study type Primary research reporting quantitative or qualitative results on validity, feasibility, accuracy, or usefulness Systematic reviews, opinion articles, editorials, conference abstracts, or studies without validation results Language Publications in English Non-English publications Publication date January 2017–February 2025 Publications before January 2017 Open in a new tab 3. Results The reviewed evidence on sensor-based technologies for the detection of loneliness and social isolation in older adults is structured around the technological workflow illustrated in Figure 2 . This framework integrates three key stages: data acquisition from multiple sensor platforms (wearables, smart home devices, and smartphones), the extraction of behavioral markers associated with loneliness, and analytical approaches based on multimodal data fusion and machine learning models. Figure 2. Open in a new tab Technological flow for the detection of loneliness and social isolation in older adults using sensors. 3.1. Sensors and Monitoring Platforms Wearable devices and smartphones can track physical activity, location, phone and app usage, heart rate, and sleep patterns [ 21 , 30 ], while smart home environmental sensors, such as motion detectors, door contacts, bed and mattress sensors, temperature and humidity monitors, and appliance usage, can provide additional behavioral context [ 27 , 31 ]. New smart textiles integrated into clothing or furniture aim to provide comfortable and continuous monitoring [ 22 , 31 ]. The reviewed literature describes a growing use of wearable devices, environmental sensors, and mobile platforms for monitoring behavioral and physiological variables related to social isolation and loneliness in older adults [ 17 , 21 , 27 , 28 , 30 , 32 , 33 , 34 ]. Table 2 below presents the main types of sensors used, the variables they measure, and examples of their application, according to recent scientific literature. Table 2. Types of Sensors used to Assess Social Isolation and Loneliness in Older Adults. Sensor Type/Platform Main Measured Variables Variables Related to Loneliness/Isolation References Motion sensors (PIR, infrared) Mobility, presence in rooms, daily activity Time spent in each room, activity/inactivity patterns, mobility [ 17 , 21 , 27 , 28 , 30 , 32 , 33 , 34 ] Door contact sensors Home entrances/exits, room usage Frequency of outings, time spent outside the home, interior door usage [ 17 , 21 , 27 , 30 , 32 , 33 , 34 ] Pressure sensors (bed/chair, smart mattress) Bed/chair presence, sleep parameters Bedtime, naps, sleep efficiency, sedentary time [ 28 , 32 , 33 , 35 , 36 ] Environmental sensors (light, electricity, water, temperature, humidity, air quality) Appliance usage, thermal comfort, home routines TV hours, kitchen/bathroom use, shower events, heating patterns [ 17 , 21 , 27 , 30 , 32 , 33 , 34 , 35 ] Actigraphs/portable accelerometers Physical activity, movement patterns Daily activity level, sedentary time, movement changes [ 28 , 32 , 37 , 38 , 39 , 40 ] Smartwatches and fitness trackers Activity, sleep, vital signs Daily steps, sleep, heart rate, physiological variability [ 28 , 31 , 32 , 36 , 37 , 38 , 39 , 41 , 42 ] Smartphone (sensors and usage logs) Communication, mobility, app usage Number/duration of calls, messages, social app usage, GNSS [ 28 , 32 , 37 , 38 , 43 , 44 , 45 ] Proximity sensors (BLE, RFID, tags) Proximity to objects or people Being at home vs. away, movement within the home, social encounters [ 32 , 46 ] Specific physiological sensors Biological indicators (HR, temperature, EDA, EEG, ECG) Heart rate, conductance, temperature, stress associated with loneliness [ 28 , 32 , 37 , 38 , 47 , 48 , 49 ] Smart textile sensors (clothing/furniture) Body activity, posture, comfort Movement, posture, social interaction, continuous monitoring [ 22 , 32 , 38 , 47 , 50 ] Audio and video sensors (NLP, cameras) Verbal interactions, facial expression, language Voice analysis, speech patterns, non-verbal expressions [ 51 , 52 , 53 ] Open in a new tab To complement the descriptive information presented in Table 2 , Table 3 provides a critical assessment of the strengths and limitations of each sensor category, emphasizing their applicability and methodological constraints in loneliness and social isolation detection. Table 3. Strengths and Limitations of Main Sensor Technologies for Detecting Loneliness and Social Isolation in Older Adults. Sensor Category Strengths Limitations Motion Sensors (PIR, Infrared) Fully passive and unobtrusive; enable long-term monitoring; no user burden; capture indoor movement dynamics Limited to indoor spaces; cannot identify individuals; limited sensitivity to subtle behavioral changes Door Contact Sensors Simple and reliable; generate clear binary event data; detect home exits and entries Only capture door events; miss detailed activity outside the home; limited insight into social interactions Pressure Sensors (Bed/Chair Sensors) Passive monitoring of sleep and rest; no wearable required; capture nocturnal behavior Restricted to specific furniture; cannot detect sleep stages; performance affected by multiple occupants Environmental Sensors (Multi-Sensor Home Systems) Capture contextual home activity; support long-term deployment; provide behavioral context for ADLs Provide indirect measures requiring interpretation; sensitive to environmental changes; installation infrastructure required Wearable Sensors Continuous physiological and activity monitoring; high temporal resolution; capture indoor and outdoor movement; commercially available Require charging and maintenance; adherence may decrease; potential discomfort or abandonment Smartphone-Based Sensing Leverages existing personal devices; captures communication and mobility data; supports multimodal analytics Privacy concerns; battery consumption; digital divide among older adults; platform heterogeneity Physiological Sensors (ECG, EDA, EEG) Capture objective stress-related biomarkers; high measurement precision; potential emotional correlates Require contact-based setup; typically limited to controlled or laboratory environments; complex signal processing Audio and Video Sensors Direct assessment of social interaction; enable linguistic, paralinguistic, and behavioral analysis; rich contextual information Highly intrusive; strong privacy and ethical concerns; computationally intensive; language-dependent Open in a new tab The comparative analysis highlights that passive sensing technologies offer high ecological validity but limited interpretability, whereas wearable and smartphone-based approaches provide richer multimodal data at the expense of usability and privacy concerns. Audio–video and physiological sensing approaches remain promising but are still constrained by ethical, technical, and deployment challenges. Future research should prioritize multimodal sensor fusion, explainable AI models, and longitudinal validation in real-world environments to improve robustness and generalizability. 3.2. Sensor-Derived Behavioral Markers and Predictive Models for Loneliness and Social Isolation Based on the variables measured by the sensors described in Section 3.1 , the literature identifies various behavioral markers associated with loneliness and social isolation, including time spent outside the home, room location patterns, daytime naps, reduced mobility, and sleep disturbances, which have been integrated into predictive models [ 54 , 55 ]. Similarly, variables related to telephone and computer use, as well as the frequency of social visits, have been incorporated into different modeling approaches [ 56 ]. The studies reviewed report performance metrics ranging from moderate to high values, depending on the type of sensor, the behavioral marker analyzed, and the model used. For example, systems based on PIR sensors and door contacts in smart homes have shown correlations with loneliness scores on the UCLA Scale (r ≈ 0.48), while multisensory platforms have achieved R 2 values of ≈ 0.86 using features derived from bed sensors and environmental parameters [ 35 , 55 ]. It should be noted, however, that most of these high-performance metrics have been achieved in controlled or semi-controlled research settings with small samples, and their generalizability to real-world conditions remains to be demonstrated. Notably, certain sensor-derived markers are more conceptually aligned with objective social isolation than with subjective loneliness. For instance, time spent outside the home, frequency of outings, and number of visits detected via door sensors or proximity beacons directly quantify social contact opportunities [ 17 , 33 ]. In contrast, markers such as sleep fragmentation, mobility variability, or linguistic traits in speech have been associated with the subjective experience of loneliness, although the mechanisms linking them remain less understood and require further validation [ 39 , 57 , 58 ]. To reflect the conceptual distinction between subjective loneliness and objective social isolation established in Section 1 , the reviewed studies are organized into two tables. Table 4 includes studies predicting loneliness using validated self-report scales (e.g., UCLA, SESLA). Table 5 compiles studies focusing on objective social isolation, operationalized through sensor-derived behavioral proxies such as mobility patterns, home exits, or living alone. This separation avoids ambiguity in interpreting the evidence, as both constructs, although related, are measured differently. Table 4. Studies Predicting Subjective Loneliness in Older Adults Using Sensors and Predictive Models. Instrument/Scale Sensor/Features Model n Population Setting Validation Metric References UCLA 3-item PIR, door contacts, PC/phone use Multiple linear regression 30 Community, homes Home Hold-out R 2 = 0.35 [ 17 ] EMA + validated scales Sleep, physical activity, health, EMA Gradient Boosting 78 Community, predementia Community Cross-validation AUC = 0.887 [ 39 ] UCLA (4 factors) Call logs, GPS location Multiple classifiers 52 Community Community Not specified Accuracy, sensitivity, specificity by factor [ 59 ] UCLA Linguistic traits (interviews) Explainable AI (XAI) 84 Older adults Laboratory Hold-out Accuracy = 0.889; AUC = 0.80; F1 = 0.80 [ 58 ] UCLA (qual + quant) Linguistic traits (interviews) ML models 104 Community Laboratory Cross-validation Precision = 94%/76%; Sensitivity = 0.90/0.57; Specificity = 1.00/0.899 [ 57 ] Loneliness scale Sociodemographic, functional health Gradient Boosted Trees 4621 Population-based, China Population-based Cross-validation AUC = 0.84 [ 60 ] Loneliness scale Psychosocial predictors, health MLP vs. Logistic Regression 1541 Population-based, Spain Population-based Not specified Accuracy = 92.3%; R 2 Nagelkerke = 0.396 [ 61 ] UCLA Speech analysis SVM, Random Forest 96 Community Laboratory Cross-validation Accuracy = 76.5% [ 59 ] Open in a new tab Table 5. Studies Predicting Objective Social Isolation in Older Adults Using Sensors and Predictive Models. Instrument/Scale Sensor/Features Model n Population Setting Validation Metric References EMA (ecological momentary assessment) Actigraphy (daily physical activity) Random Forest 78 Community, predementia Community 10-fold CV AUC = 0.935; Accuracy = 0.849; F1 = 0.824 [ 39 ] Not applicable (descriptive) Multimodal: wearables, home sensors Descriptive 20 Community, post-fracture Home Not applicable Feasibility outcomes [ 44 ] Not applicable (descriptive) PIR sensors, door contacts Descriptive analysis 60 Community, COVID-19 Home Not applicable Behavioral changes [ 45 ] Experimental task Non-verbal signals (avatar) ML models 40 Older adults Laboratory Cross-validation To be determined [ 51 ] Functional decline scales Multimodal: wearables, sensors Correlation analysis 15 Community, post-fracture Home Not applicable Preliminary correlations [ 36 ] Open in a new tab 3.3. Multimodal Data Fusion and Artificial Intelligence Approaches Combining data from different types of sensors using machine learning techniques allows for the construction of more robust predictive models for detecting loneliness. Mobile and wearable sensors, including accelerometers, heart rate monitors, sleep trackers, GNSS (Global Navigation Satellite System) geolocation, and smartphone usage logs, are widely used to infer mood, stress, depression, anxiety, and daily functioning through passive and personal detection approaches [ 25 , 62 , 63 ]. In controlled or semi-controlled environments, stress and affect classification based on multimodal biological signals has achieved high accuracy, often exceeding 90% [ 25 , 64 ]. The fusion of heterogeneous sensor modalities—including wearables, environmental sensors, and smartphones—has been suggested as a strategy to improve predictive robustness compared to single-sensor approaches, supporting the concept of a ‘behavior’ composed of multiple digital markers [ 26 , 65 , 66 ]. Multimodal fusion techniques have been applied to activity recognition, cardiovascular risk estimation, and the prediction of cognitive and mobility outcomes in older adults, with ensemble and gradient boosting models showing moderate to high correlations with clinical reference assessments in research settings [ 26 , 65 , 67 ]. Figure 3 presents a taxonomic overview of the sensor-based AI landscape identified in this review. The diagram organizes four sensor categories (ambient, wearables, smartphone, audio/visual) against three AI model families (Traditional ML, Deep Learning, NLP), with representative applications derived from the 57 included studies. Figure 3. Open in a new tab Taxonomic overview of sensor platforms and AI models. The reviewed studies employ diverse AI approaches that can be categorized into three main families: Traditional machine learning: Algorithms such as Random Forest, Gradient Boosting, and Support Vector Machines are most commonly used due to their interpretability and good performance with tabular data extracted from sensors [ 39 , 57 , 59 , 60 ]. These models require manual feature engineering but offer better transparency for clinical applications [ 58 ]. Deep learning: Although less frequent, deep learning approaches including multilayer perceptrons and convolutional neural networks have been applied to raw sensor data, particularly for activity recognition and speech analysis [ 58 , 61 ]. These methods can automatically learn features but require larger datasets and raise concerns about interpretability (‘black box’ issue). Natural Language Processing: Specific to audio-based sensing, NLP techniques including explainable AI (XAI) have been used to analyze linguistic traits in interviews and speech patterns associated with loneliness [ 57 , 58 ]. These approaches offer unique insights into subjective experiences but require careful validation across languages and cultures. However, no large-scale systematic studies have yet directly compared unimodal versus multimodal approaches specifically for loneliness detection in older adults. The current evidence, while promising, remains preliminary and derives primarily from controlled or semi-controlled environments with limited sample sizes and internal validation only. 3.4. Digital Phenotyping Applications in Older Adult Populations In the included studies, digital phenotyping has been applied primarily for the continuous assessment of physical activity, mobility, sleep, cognition, and mood in older adults, using wearable devices, smartphones, and home-based detection systems [ 67 , 68 , 69 ]. Several studies have demonstrated the feasibility and acceptability of long-term monitoring using GNSS geolocation and wearable devices in older adults living in the community, including those with mild cognitive impairment (MCI) or early-stage dementia [ 69 , 70 ]. The mobility and living space characteristics derived from sensor data show consistent associations with cognitive function, physical performance, and depressive symptoms [ 70 , 71 ]; variables closely related to social isolation and psychosocial vulnerability in older adults. Similarly, sleep metrics such as fragmentation and efficiency, obtained using portable devices, have been associated with daily fluctuations in depressive symptoms in socially vulnerable older adults [ 23 ]. In addition, multisensory home systems combined with artificial intelligence techniques have demonstrated their potential to distinguish between normal aging, MCI, and early dementia, achieving high classification performance in research settings [ 26 , 67 ]. Table 6 shows the evidence on digital phenotyping in older adults, displaying the dimensions of aging, sensors used, derived variables, and relevant results. Table 6. Dimensions, Sensors, Variables, and Findings in Digital Phenotyping of Aging. Aging Dimension Sensors/Systems Extracted Variables Relevant Findings References Physical activity and mobility/life-space Smartphone (GNSS, accelerometer), wrist wearables, ECG patches with accelerometer, home motion sensors Time at home, distance traveled, radius of gyration, number of significant locations, circadian routine, PA intensity, steps/day, temporal PA patterns Higher activity and spatial diversity are associated with better cognition, less functional decline, less depression, and greater community engagement; low PA is linked to higher risk of MCI/dementia and worse executive function [ 64 , 68 , 72 , 73 ] Cognition and dementia risk Smartphone (GNSS, app usage, keystrokes), wearables, multisensory home systems (PIR, doors, bed, medication, beacons) Mobility phenotypes, regularity of habits, typing speed and variability, time of first/last phone interaction, high-resolution PA metrics Combinations of digital traits (mobility, PA, device usage, home patterns) discriminate between normal aging and early MCI/dementia with good ML model performance [ 64 , 71 , 72 , 73 ] Mood/depressive symptoms Smartphone (GNSS, calls/app usage), activity and sleep wearables, bed sensors Sleep fragmentation and efficiency, activity variability, daily mobility, volume and temporal pattern of calls/screen usage Lower mobility, more irregular sleep, and certain phone usage patterns are associated with greater severity and variability of depressive symptoms over time [ 64 , 74 , 75 , 76 , 77 ] Sleep Wearables (actigraphy, fitness bands), bed sensors, ECG patches Sleep duration, nocturnal awakenings, efficiency, night-to-night variability, circadian activity rhythms Sleep metrics are related to daily mood fluctuations and variability of depressive symptoms; some sleep traits contribute to prediction models of cognitive decline and social isolation [ 44 , 71 , 74 , 75 , 76 ] Social isolation and community life Home motion and door sensors, proximity beacons, smartphone (GNSS, communication logs) Time away from home, frequency of outings, room presence patterns, call frequency/duration, sociability indicators Patterns of lower community mobility, fewer visits to places, and less social interaction are associated with social isolation, worse functional status, and greater psychosocial vulnerability [ 44 , 64 , 67 , 78 , 79 ] Open in a new tab 3.5. Inferring Social and Contextual Behavior from Sensor Data Beyond individual health metrics, AI-based analysis of sensor data is increasingly being used to infer social and contextual behavior. Computer vision, smartphone sensors, and wearable device data enable the quantification of social interactions, sociability, and living space, including community engagement derived from GNSS, phone communication patterns, and environmental sensing [ 55 , 62 , 63 ]. Although both classical and deep learning methods can recognize interaction patterns, daily routines, and behavioral changes, their generalization and clinical reliability remain limited by small sample sizes, homogeneous cohorts, and heterogeneous methodological choices across studies [ 24 , 63 , 64 ]. 3.6. Summary of Evidence Levels Based on the reviewed studies, the evidence can be categorized into three levels according to methodological rigor, sample size, and validation strategies. Relatively robust evidence: Associations between basic behavioral markers (time outside home, mobility patterns, sleep fragmentation) and loneliness or social isolation are supported by multiple studies with consistent findings across different populations and settings [ 17 , 33 , 35 , 39 ]. These markers have been validated using standardized instruments (e.g., UCLA Loneliness Scale) and show moderate correlations in community-dwelling older adults [ 17 , 35 ]. Promising but preliminary evidence: Multimodal data fusion approaches demonstrate potential for improving predictive performance, with some studies reporting AUC values > 0.85 [ 39 , 80 ]. However, direct comparisons with unimodal approaches are lacking, and most studies are limited by small sample sizes (typically n < 100), lack of external validation, and controlled settings [ 25 , 26 , 65 ]. Exploratory evidence: Emerging approaches including linguistic analysis using natural language processing [ 57 , 58 ] speech pattern recognition [ 59 ] and digital phenotyping for cognitive impairment [ 67 , 69 ] represent innovative methodologies requiring replication in larger, diverse populations before clinical translation. Direct performance comparisons between studies are not feasible due to substantial heterogeneity in dataset characteristics, evaluation metrics, problem formulations, and validation strategies. This heterogeneity underscores the need for standardized benchmarks and reporting guidelines [ 21 , 27 , 28 ]. 4. Discussion Based on the findings presented, the discussion is organized into five subsections: (1) summary of the technological and methodological findings identified; (2) the contribution of sensors and AI to the detection of loneliness; (3) the advantages and limitations of passive monitoring; (4) ethical considerations, privacy, and acceptance by users; and (5) future lines of research. 4.1. Synthesis of Principal Findings This narrative review has identified and organized recent evidence on the use of sensor-based technologies and artificial intelligence (AI) for the detection of unwanted loneliness and social isolation in older adults. The results indicate that there are three main technological components that interact in a continuous flow: (a) sensor platforms (wearables, environmental, and smartphone) that collect raw data; (b) behavioral and digital markers derived from that data (mobility, sleep, social interaction, patterns at home, linguistic traits); and (c) machine learning (ML) and multimodal fusion analytical approaches that transform the markers into predictions or correlations with validated loneliness scales [ 17 , 39 , 53 ]. The literature converges in indicating that, although self-reports remain essential, passive sensors offer objective, continuous, and contextual measurement of behaviors associated with loneliness, overcoming limitations such as social desirability bias or the episodic nature of questionnaires [ 21 , 27 ]. However, the field is still in an emerging phase, with studies presenting considerable methodological heterogeneity, small sample sizes, and limited longitudinal validation [ 37 , 81 ]. 4.2. The Role of Sensor Technologies and AI in Loneliness Detection Passive sensors enable the capture of rich, multidimensional digital phenotypes of aging. The review shows that markers such as reduced time spent outside the home, low morning mobility, sleep fragmentation, and decreased frequency of telephone communications show consistent, albeit moderate, associations with loneliness scores [ 35 , 39 , 59 ]. Multimodal data fusion (e.g., combining data from wearables, environmental sensors, and smartphones) is emerging as a key strategy for improving the robustness and predictive performance of models, overcoming the limitations of unimodal approaches [ 25 , 26 ]. Machine learning algorithms, particularly ensemble models such as Random Forest and Gradient Boosting, have shown promise in modeling the complex relationship between these digital markers and loneliness/isolation constructs, achieving performance metrics such as AUC > 0.85 in some studies [ 39 , 80 ]. However, most studies have been conducted in controlled or semi-controlled environments, and the interpretability of the models (‘black box’ issue) remains a challenge for their clinical acceptance [ 58 ]. 4.3. Advantages and Current Limitations of Passive Sensing Passive monitoring using sensors offers distinct advantages for assessing loneliness in older adults. Its main strength lies in its ability to capture real-world behavioral dynamics objectively, continuously, and discreetly [ 64 , 68 ]. This is particularly valuable for populations with difficulties in frequent self-reporting or with cognitive impairment. Smart home systems (AAL), for example, have proven useful for inferring patterns of isolation from metrics such as intra-domestic mobility and use of spaces [ 17 , 45 ]. The reviewed evidence indicates that these systems can identify patterns of behavior—such as reduced mobility, use of domestic space, or decreased outings—that show significant associations with standardized loneliness scales, pointing to their potential usefulness as objective indicators [ 21 , 27 ]. The fusion of multimodal data (wearables, environmental, smartphones) is emerging as a key strategy for improving predictive robustness compared to unimodal approaches [ 26 ]. However, despite these promising results, the field is still in its early stages, characterized by small sample sizes, heterogeneous methods, inconsistent definitions of loneliness versus social isolation, and limited longitudinal validation [ 32 , 37 , 81 ]. The strength of this evidence is still limited. Most studies are based on small samples, short follow-up periods, and lack population diversity [ 21 , 27 ]. There is a notable lack of longitudinal studies with consistent methodologies in the field of aging [ 21 ]. In addition, privacy, data security, and acceptability, particularly with regard to camera monitoring, are issues of great concern; older people tend to respond positively, but are cautious about potential misuse and lack of human interaction [ 32 , 81 ]. This position is a conceptual discrepancy between objective markers of isolation (e.g., time spent at home) and the subjective experience of loneliness, which do not always correlate [ 81 ]. This divergence underscores that sensor technologies should be viewed as a valuable complement to, rather than a replacement for, traditional psychosocial assessments, requiring careful integration of both approaches for a holistic assessment of social well-being in older adults. From an implementation perspective, the technologies reviewed exhibit varying levels of practical readiness. To provide a more structured perspective on their maturity, it is useful to consider the Technology Readiness Levels (TRL) scale, a standardized framework widely used to assess how mature a technology is before it can be integrated into systems or deployed in real-world settings. Applying this scale to the reviewed technologies: TRL 7–9 (Mature systems, ready for deployment): Basic sensors (PIR, door contacts, pressure mats) have reached TRL 9, with demonstrated commercial maturity and deployment in real-world pilot studies [ 17 , 44 ]. Commercial wearables (actigraphy, activity trackers) are at TRL 8–9, widely available and validated for activity and sleep monitoring in general populations [ 37 , 38 , 40 ]. TRL 4–6 (Systems in validation phase): Integrated multimodal systems for loneliness detection (combining wearables + environmental sensors + smartphones) are at TRL 4–5, with validation in controlled or semi-controlled environments, but require demonstration in real-world conditions at scale and overcoming integration challenges [ 26 , 44 , 67 ]. TRL 1–3 (Proof of concept): Audio and video analysis using NLP for loneliness detection, as well as advanced physiological sensors (EEG, EDA) applied to this specific domain, are at TRL 2–3, currently limited to laboratory settings with small samples and controlled conditions [ 51 , 52 , 57 , 58 ]. Smart textiles for loneliness monitoring are in early development stages (TRL 2–3) [ 22 , 47 ]. Considering these challenges, a staged implementation pathway is proposed. In the short term (1–3 years), simple single-modal approaches using commercial sensors are ready for broader deployment [ 17 , 37 , 43 ]. In the medium term (3–5 years), integrated multimodal systems with machine learning can be piloted [ 26 , 44 , 58 ]. In the long term (>5 years), advanced AI-driven approaches require further development before clinical adoption [ 57 , 61 , 67 ]. 4.4. Ethical, Privacy, and User Acceptance Considerations The qualitative and review studies included describe concerns related to privacy, autonomy, and user acceptance. The main concerns and conditions for the acceptance of passive monitoring technologies in older adults are summarized in Table 7 . Table 7. Main Ethical and Privacy Concerns and Conditions for Acceptance of Passive Monitoring Technologies in Older Adults. Topic Key Findings References Privacy and data misuse Older adults often fear misuse by third parties, surveillance, and data leaks. [ 32 , 81 , 82 , 83 , 84 , 85 , 86 ] Ethics of emotional monitoring Skepticism about sensors’ ability to “read” emotions like loneliness; concerns about stigma or misinterpretation. [ 21 , 27 , 32 , 81 ] Conditional acceptance Many value potential benefits (safety, loneliness detection, aging in place) but only accept systems that are discreet, transparent, and controllable. [ 32 , 81 , 84 , 85 , 87 , 88 ] Autonomy and control Passive monitoring may threaten perceived autonomy; residents may resist or stop using systems that interfere with their routine or values. [ 89 , 90 ] Design preferences Non-invasive/ambient sensors preferred over video, integrated into familiar objects, with customizable alerts and clear data-sharing rules. [ 82 , 88 , 90 , 91 ] Trust and aesthetics Trust in information handling and system reliability, along with non-stigmatizing and aesthetically pleasing design, promote acceptance. [ 50 , 85 , 88 ] Open in a new tab These findings underscore that any implementation of passive sensors must prioritize trust in the system, privacy, and user control. Trust is at the core of acceptability. The design must incorporate an ethical framework based on transparency and a person-centered approach, as these technologies can be perceived as surveillance, eliminating the autonomy they are intended to protect. The adoption of edge computing architectures in IoT environments can be a solution to strengthen privacy. The processing of more sensitive data is carried out in the nodes or sensors themselves, without that information leaving the device. The results that are transmitted are anonymized, reducing the risk of exposure of personal data and reinforcing user trust and security. Notably, audio and video sensors—despite their rich informational value—are consistently perceived by older adults as more intrusive than ambient environmental sensors. This aligns with the design preferences summarized in Table 7 , where non-invasive and discreet solutions are prioritized over camera-based monitoring [ 82 , 88 , 90 , 91 ]. The adoption of edge computing architectures in IoT environments can be a solution to strengthen privacy. From a technical perspective, edge computing can be implemented so that raw signals (e.g., accelerometer data, passive infrared motion events) are processed locally on the sensor node. Only aggregated, anonymized features—such as hourly activity counts or time spent outside the home—are transmitted to external servers. Complementary privacy-preserving strategies, such as federated learning, enable model training across distributed devices without centralizing sensitive data. However, it is important to acknowledge that anonymization techniques (including aggregation, pseudonymization, and differential privacy) are not infallible; residual risks such as model inversion attacks or metadata leakage persist and must be addressed through continuous technical and procedural safeguards. 4.5. Future Research Directions To advance the field from promising prototypes to impactful real-world applications, future research must move beyond technical validation towards a structured, multi-dimensional agenda. Based on the gaps identified in this review, we organize the key priorities into five interconnected research directions, summarized in Table 8 . This framework distinguishes between short-term goals (1–3 years), which are immediately actionable, and long-term objectives (>5 years), which require sustained, multi-stakeholder effort. Table 8. Structured Research Agenda for Sensor-Based Detection of Loneliness and Social Isolation in Older Adults. Research Direction Key Priorities and Short-Term vs. Long-Term Goals Longitudinal and Diverse Cohorts Short-term: Conduct multi-site studies with larger, more diverse samples (including rural, low-income, and ethnic minority populations). Long-term: Establish long-term cohorts (>5 years) to assess the stability, predictive validity, and causal relationships of digital behavioral markers with loneliness and health outcomes. Standardization and Benchmarking Short-term: Develop community-agreed reporting standards for sensor-based loneliness studies (sample, sensors, features, models, validation). Long-term: Create open-source benchmark datasets to allow direct, fair comparison between unimodal and multimodal approaches, and between different AI models. Implementation Science and Real-World Integration Short-term: Conduct pragmatic trials to evaluate the integration of simple sensor systems into existing social care workflows. Assess cost-effectiveness, user burden, and technical reliability. Long-term: Develop interoperable platforms that can feed data into electronic health records and trigger timely, ethical interventions. Ethics, Privacy, and User-Centered Design Short-term: Operationalize concepts like “functional privacy” and “perceived invasiveness” through co-design studies with older adults, caregivers, and practitioners. Long-term: Develop and validate privacy-preserving technologies (e.g., edge computing, federated learning) that are transparent and give users meaningful control over their data. Clinical Translation and Risk Management Short-term: Quantify the clinical risks of false positives (unnecessary anxiety, intervention) and false negatives (missed support) in pilot implementation studies. Long-term: Establish clear clinical guidelines on how to interpret and act upon alerts generated by these systems, ensuring they augment, not replace, human care. Open in a new tab Crucially, an overarching priority that must permeate all the above directions is equity. Future research must proactively ensure that these technologies are accessible and valid across diverse populations, including people with cognitive impairment, low digital literacy, ethnic minorities, and those living in socioeconomically disadvantaged or rural settings with limited access to technological infrastructure. Without this focus, there is a significant risk of exacerbating existing health disparities in both loneliness and social isolation. 4.6. Limitations of This Review This review has several limitations. First, although we followed PRISMA guidelines and conducted a systematic search, the review was not registered in a database such as PROSPERO. Second, the substantial heterogeneity across studies—in terms of design, sample size, sensors used, and outcome measures—prevented us from conducting a meta-analysis and limited our synthesis to a narrative approach. Third, the exclusion of non-English publications and grey literature may have introduced language or publication bias. Fourth, the overall quality of the included studies remains limited, with most relying on small samples and lacking external validation. Finally, the rapid evolution of technology means that recent innovations may not be fully captured, despite our focus on the 2017–2025 period. Abbreviations The following abbreviations are used in this manuscript: AAL Ambient Assisted Living AI Artificial Intelligence AUC Area Under the Curve BLE Bluetooth Low Energy ECG Electrocardiogram EDA Electrodermal Activity EEG Electroencephalography EMA Ecological Momentary Assessment GBM Gradient Boosting Machine GNSS Global Navigation Satellite System HR Heart Rate HRV Heart Rate Variability IoT Internet of Things ML Machine Learning MLP Multilayer Perceptron MCI Mild Cognitive Impairment NLP Natural Language Processing NMAE Normalized Mean Absolute Error NRMSE Normalized Root Mean Square Error PA Physical Activity PIR Passive Infrared PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses RF Random Forest RFID Radio-Frequency Identification SESLA Social and Emotional Loneliness Scale for Adults UCLA University of California, Los Ángeles Loneliness Scale XAI Explainable Artificial Intelligence Open in a new tab Supplementary Materials The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/s26072028/s1 , Table S1: PRISMA 2020 for Abstracts Checklist; Table S2: PRISMA 2020 Checklist. sensors-26-02028-s001.zip (80.8KB, zip) Appendix A Table A1 presents the complete search strings used for each database, following the PRISMA 2020 guidelines [ 29 ]. The searches were conducted between October and December 2025, as specified in Section 2.3 . Table A1. Search strings for each database. Database Search String PubMed (loneliness OR “social isolation” OR “social behavior” OR “social interaction”) AND (“older adults” OR elderly OR aging OR aged) AND (sensor* OR wearable* OR “smart home” OR “ambient assisted living” OR “passive sensing” OR monitoring) Scopus TITLE-ABS-KEY ((loneliness OR “social isolation”) AND (“older adults” OR elderly) AND (sensor* OR wearable* OR “smart home”)) Web of Science TS = (loneliness OR “social isolation”) AND TS = (“older adults” OR elderly) AND TS = (sensor* OR wearable* OR “smart home”) IEE Xplore (“All Metadata”:loneliness OR “All Metadata”:”social isolation”) AND (“All Metadata”:”older adults” OR “All Metadata”:elderly) AND (“All Metadata”:sensor* OR “All Metadata”:wearable*) Open in a new tab Author Contributions Conceptualization, M.M.P.V., M.E.E. and J.C.C.M.; Methodology, M.M.P.V. and J.M.M.M.; Investigation, M.M.P.V.; Formal Analysis, M.M.P.V. and J.F.G.-G.; Data Curation, M.M.P.V.; Writing—Original Draft Preparation, M.M.P.V. and J.M.M.M.; Writing—Review & Editing, M.M.P.V., J.M.M.M., J.L.H.B., M.E.E. and J.C.C.M.; Visualization, M.M.P.V.; Supervision, M.M.P.V., J.L.H.B. and M.E.E.; Project Administration, M.E.E.; Funding Acquisition, J.C.C.M.; Resources, J.C.C.M. All authors have read and agreed to the published version of the manuscript. Institutional Review Board Statement Not applicable. Informed Consent Statement Not applicable. Data Availability Statement No new data were created or analyzed in this study. Data sharing is not applicable to this article. Conflicts of Interest The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding Statement This work has been partially supported by grant PID2024-156412OB-I00 funded by MICIU/AEI/10.13039/501100011033 and by ERDF/EU; grant M.2 PDC_000756 funded by Consejería de Universidad, Investigación e Innovación and by ERDF Andalusia Program 2021-2027; and project PDC2023-145863-I00, funded by MCIN/AEI/10.13039/501100011033 and by the European Union NextGenerationEU/PRTR. Footnotes Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. References 1. Puyané M., Chabrera C., Camón E., Cabrera E. Uncovering the Impact of Loneliness in Ageing Populations: A Comprehensive Scoping Review. BMC Geriatr. 2025;25:244. doi: 10.1186/s12877-025-05846-4. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Lapane K.L., Lim E., McPhillips E., Barooah A., Yuan Y., Dube C.E. Health Effects of Loneliness and Social Isolation in Older Adults Living in Congregate Long Term Care Settings: A Systematic Review of Quantitative and Qualitative Evidence. Arch. Gerontol. Geriatr. 2022;102:104728. doi: 10.1016/j.archger.2022.104728. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Courtin E., Knapp M. Social Isolation, Loneliness and Health in Old Age: A Scoping Review. Health Soc. Care Community. 2017;25:799–812. doi: 10.1111/hsc.12311. [ DOI ] [ PubMed ] [ Google Scholar ] 4. Perissinotto C., Holt-Lunstad J., Periyakoil V.S., Covinsky K. A Practical Approach to Assessing and Mitigating Loneliness and Isolation in Older Adults. J. Am. Geriatr. Soc. 2019;67:657–662. doi: 10.1111/jgs.15746. [ DOI ] [ PubMed ] [ Google Scholar ] 5. Berg-Weger M., Morley J.E. Loneliness in Old Age: An Unaddressed Health Problem. J. Nutr. Health Aging. 2020;24:243–245. doi: 10.1007/s12603-020-1323-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Wu B. Social Isolation and Loneliness among Older Adults in the Context of COVID-19: A Global Challenge. Glob. Health Res. Policy. 2020;5:27. doi: 10.1186/s41256-020-00154-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Kobayashi L.C., Steptoe A. Social Isolation, Loneliness, and Health Behaviors at Older Ages: Longitudinal Cohort Study. Ann. Behav. Med. 2018;52:582–593. doi: 10.1093/abm/kax033. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Sepúlveda-Loyola W., Rodríguez-Sánchez I., Pérez-Rodríguez P., Ganz F., Torralba R., Oliveira D.V., Rodríguez-Mañas L. Impact of Social Isolation Due to COVID-19 on Health in Older People: Mental and Physical Effects and Recommendations. J. Nutr. Health Aging. 2020;24:938–947. doi: 10.1007/s12603-020-1500-7. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Nakagomi A., Tsuji T., Saito M., Ide K., Kondo K., Shiba K. Social Isolation and Subsequent Health and Well-Being in Older Adults: A Longitudinal Outcome-Wide Analysis. Soc. Sci. Med. 2023;327:115937. doi: 10.1016/j.socscimed.2023.115937. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Gerst-Emerson K., Jayawardhana J. Loneliness as a Public Health Issue: The Impact of Loneliness on Health Care Utilization Among Older Adults. Am. J. Public Health. 2015;105:1013–1019. doi: 10.2105/AJPH.2014.302427. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Czaja S.J., Moxley J.H., Rogers W.A. Social Support, Isolation, Loneliness, and Health Among Older Adults in the PRISM Randomized Controlled Trial. Front. Psychol. 2021;12:728658. doi: 10.3389/fpsyg.2021.728658. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Maes M., Qualter P., Lodder G.M.A., Mund M. How (Not) to Measure Loneliness: A Review of the Eight Most Commonly Used Scales. Int. J. Environ. Res. Public Health. 2022;19:10816. doi: 10.3390/ijerph191710816. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Mund M., Maes M., Drewke P.M., Gutzeit A., Jaki I., Qualter P. Would the Real Loneliness Please Stand Up? The Validity of Loneliness Scores and the Reliability of Single-Item Scores. Assessment. 2023;30:1226–1248. doi: 10.1177/10731911221077227. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Reinwarth A.C., Ernst M., Krakau L., Brähler E., Beutel M.E. Screening for Loneliness in Representative Population Samples: Validation of a Single-Item Measure. PLoS ONE. 2023;18:e0279701. doi: 10.1371/journal.pone.0279701. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Gosling C.J., Colle R., Cartigny A., Jollant F., Corruble E., Frajerman A. Measuring Loneliness: A Head-to-Head Psychometric Comparison of the 3- and 20-Item UCLA Loneliness Scales. Psychol. Med. 2024;54:3821–3827. doi: 10.1017/S0033291724002083. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Deol E.S., Yamashita K., Elliott S., Malmstorm T.K., Morley J.E. Validation of the ALONE Scale: A Clinical Measure of Loneliness. J. Nutr. Health Aging. 2022;26:421–424. doi: 10.1007/s12603-022-1794-8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Austin J., Dodge H.H., Riley T., Jacobs P.G., Thielke S., Kaye J. A Smart-Home System to Unobtrusively and Continuously Assess Loneliness in Older Adults. IEEE J. Transl. Eng. Health Med. 2016;4:2800311. doi: 10.1109/JTEHM.2016.2579638. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Bugallo-Carrera C., Dosil-Díaz C., Anido-Rifón L., Pacheco-Lorenzo M., Fernández-Iglesias M.J., Gandoy-Crego M. A Systematic Review Evaluating Loneliness Assessment Instruments in Older Adults. Front. Psychol. 2023;14:1101462. doi: 10.3389/fpsyg.2023.1101462. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Hughes M.E., Waite L.J., Hawkley L.C., Cacioppo J.T. A Short Scale for Measuring Loneliness in Large Surveys: Results from Two Population-Based Studies. Res. Aging. 2004;26:655–672. doi: 10.1177/0164027504268574. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. De Jong Gierveld J., Van Tilburg T. The De Jong Gierveld Short Scales for Emotional and Social Loneliness: Tested on Data from 7 Countries in the UN Generations and Gender Surveys. Eur. J. Ageing. 2010;7:121–130. doi: 10.1007/s10433-010-0144-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Qirtas M.M., Zafeiridi E., Pesch D., White E.B. Loneliness and Social Isolation Detection Using Passive Sensing Techniques: Scoping Review. JMIR Mhealth Uhealth. 2022;10:e34638. doi: 10.2196/34638. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Zhou Y., Ratcliffe J., Molteni E., Patel A., Liu J., Mexia N., Rees J., Matcham F., Antonelli M., Tinker A., et al. Smart Textile Systems for Loneliness Monitoring in Older People Care: A Review of Sensing and Design Innovations. Adv. Electron. Mater. 2025;11:e00300. doi: 10.1002/aelm.202500300. [ DOI ] [ Google Scholar ] 23. Mohr D.C., Zhang M., Schueller S.M. Personal Sensing: Understanding Mental Health Using Ubiquitous Sensors and Machine Learning. Annu. Rev. Clin. Psychol. 2017;13:23–47. doi: 10.1146/annurev-clinpsy-032816-044949. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Garcia-Ceja E., Riegler M., Nordgreen T., Jakobsen P., Oedegaard K.J., Tørresen J. Mental Health Monitoring with Multimodal Sensing and Machine Learning: A Survey. Pervasive Mob. Comput. 2018;51:1–26. doi: 10.1016/j.pmcj.2018.09.003. [ DOI ] [ Google Scholar ] 25. Muzammal M., Talat R., Sodhro A.H., Pirbhulal S. A Multi-Sensor Data Fusion Enabled Ensemble Approach for Medical Data from Body Sensor Networks. Inf. Fusion. 2020;53:155–164. doi: 10.1016/j.inffus.2019.06.021. [ DOI ] [ Google Scholar ] 26. Cook D.J., Schmitter-Edgecombe M. Fusing Ambient and Mobile Sensor Features into a Behaviorome for Predicting Clinical Health Scores. IEEE Access. 2021;9:65033–65043. doi: 10.1109/ACCESS.2021.3076362. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Prabhu D., Kholghi M., Sandhu M., Lu W., Packer K., Higgins L., Silvera-Tawil D. Sensor-Based Assessment of Social Isolation and Loneliness in Older Adults: A Survey. Sensors. 2022;22:9944. doi: 10.3390/s22249944. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Khan S.S., Gu T., Spinelli L., Wang R.H. Sensor-Based Assessment of Social Isolation in Community-Dwelling Older Adults: A Scoping Review. BioMed. Eng. OnLine. 2023;22:18. doi: 10.1186/s12938-023-01080-4. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Page M.J., Moher D., Bossuyt P.M., Boutron I., Hoffmann T.C., Mulrow C.D., Shamseer L., Tetzlaff J.M., Akl E.A., Brennan S.E., et al. PRISMA 2020 Explanation and Elaboration: Updated Guidance and Exemplars for Reporting Systematic Reviews. BMJ. 2021;372:n160. doi: 10.1136/bmj.n160. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Goonawardene N., Toh X., Tan H.-P. Sensor-Driven Detection of Social Isolation in Community-Dwelling Elderly. In: Zhou J., Salvendy G., editors. Human Aspects of IT for the Aged Population. Applications, Services and Contexts. Springer International Publishing; Cham, Switzerland: 2017. pp. 378–392. Lecture Notes in Computer Science, Vol. 10298. [ Google Scholar ] 31. Probst F., Ratcliffe J., Molteni E., Mexia N., Rees J., Matcham F., Antonelli M., Tinker A., Shi Y., Ourselin S., et al. A Scoping Review on Human-Centered Design Approaches and Considerations in the Design of Technologies for Loneliness and Social Isolation in Older Adults. Des. Sci. 2024;10:e39. doi: 10.1017/dsj.2024.22. [ DOI ] [ Google Scholar ] 32. Cho E., Cho H., Demiris G., Harrison S., Ji X., Yuh A., Sokolsky O., Lee I. Older Adults’ Perceptions and Attitudes Toward Passive Sensors to Measure Loneliness: A Qualitative Study. Sage Open Aging. 2025;11:30495334251325607. doi: 10.1177/30495334251325607. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Yu K., Wu C., Barnes L.L., Silbert L.C., Beattie Z., Croff R., Miller L., Dodge H.H., Kaye J.A. Life-Space Mobility Is Related to Loneliness Among Living-Alone Older Adults: Longitudinal Analysis with Motion Sensor Data. J. Am. Geriatr. Soc. 2025;73:1125–1134. doi: 10.1111/jgs.19331. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Bouaziz G., Brulin D., Campo E. Technological Solutions for Social Isolation Monitoring of the Elderly: A Survey of Selected Projects from Academia and Industry. Sensors. 2022;22:8802. doi: 10.3390/s22228802. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. Ji X., Yuh A., Erdélyi V., Mizumoto T., Choi H., Harrison S.L., Cho E., Weimer J., Nagahara H., Higashino T., et al. Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems. ACM; New York, NY, USA: 2024. Exploring Effective Sensing Indicators of Loneliness for Elderly Community in US and Japan; pp. 1–9. [ Google Scholar ] 36. Dayyani F., Chu C.H., Abedi A., Khan S.S. Correlations between Social Isolation and Functional Decline in Older Adults after Lower Limb Fractures Using Multimodal Sensors: A Pilot Study. Algorithms. 2024;17:383. doi: 10.3390/a17090383. [ DOI ] [ Google Scholar ] 37. Site A., Lohan E.S., Jolanki O., Valkama O., Hernandez R.R., Latikka R., Alekseeva D., Vasudevan S., Afolaranmi S., Ometov A., et al. Managing Perceived Loneliness and Social-Isolation Levels for Older Adults: A Survey with Focus on Wearables-Based Solutions. Sensors. 2022;22:1108. doi: 10.3390/s22031108. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Rees J., Ratcliffe J., Liu W., Zhou Y., Ourselin S., Antonelli M., Patel A., Shi Y., Liu J., Tinker A., et al. Understanding the Thoughts and Preferences for Technologies Designed to Detect Feelings of Loneliness: Interview Study Among Older Adults. JMIR Hum. Factors. 2025;12:e73694. doi: 10.2196/73694. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Kang B., Park M.K., Kim J.I., Yoon S., Heo S.-J., Kang C., Lee S., Choi Y., Hong D. Exploring Factors Related to Social Isolation Among Older Adults in the Predementia Stage Using Ecological Momentary Assessments and Actigraphy: Machine Learning Approach. J. Med. Internet Res. 2025;27:e69379. doi: 10.2196/69379. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Schrempft S., Jackowska M., Hamer M., Steptoe A. Associations between Social Isolation, Loneliness, and Objective Physical Activity in Older Men and Women. BMC Public Health. 2019;19:74. doi: 10.1186/s12889-019-6424-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Osa Sanchez A., Jossa-Bastidas O., Mendez-Zorrilla A., Oleagordia-Ruiz I., Garcia-Zapirain B. Multisource Data for Monitoring and Intervention of Loneliness in the Elderly. Int. J. Integr. Care. 2025;25:08. doi: 10.5334/ijic.ICIC24003. [ DOI ] [ PubMed ] [ Google Scholar ] 42. Jossa O., Osa A., Miranda A., Oleagordia I., Gomez C., Mendez A., Garcia-Zapirain Soto B. Big Data and IoT System Using a Serverless Architecture for Monitoring and Intervention of Loneliness in the Elderly. Int. J. Integr. Care. 2023;23:698. doi: 10.5334/ijic.ICIC23603. [ DOI ] [ Google Scholar ] 43. Gao H., Wang X., Jiang Y., Chen F., Moriyama M., Zhou X. Psychometric Properties of the Social Isolation Scale for Older Adults in Geriatric Long-Term Care Facilities. Geriatr. Nurs. 2025;65:103486. doi: 10.1016/j.gerinurse.2025.103486. [ DOI ] [ PubMed ] [ Google Scholar ] 44. Abedi A., Chu C.H., Khan S.S. Multimodal Sensor Dataset for Monitoring Older Adults Post Lower Limb Fractures in Community Settings. Sci. Data. 2025;12:733. doi: 10.1038/s41597-025-05069-7. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. Lee K., Marchal N., Robinson E.L., Powell K.R. Behavioral Markers in Older Adults During COVID-19 Confinement: Secondary Analysis of In-Home Sensor Data. JMIR Mhealth Uhealth. 2025;13:e56678. doi: 10.2196/56678. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Jesús-Azabal M., Mariano L., Galán-Jiménez J., García-Alonso J. Evaluating Technical Viability of a Loneliness Detector System for Older Adults from Rural Areas. Gerontechnology. 2022;21:1. doi: 10.4017/gt.2022.21.s.614.opp4. [ DOI ] [ Google Scholar ] 47. Probst F., Rees J., Aslam Z., Mexia N., Molteni E., Matcham F., Antonelli M., Tinker A., Shi Y., Ourselin S., et al. Evaluating a Smart Textile Loneliness Monitoring System for Older People: Co-Design and Qualitative Focus Group Study. JMIR Aging. 2024;7:e57622. doi: 10.2196/57622. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 48. Rees J., Matcham F., Probst F., Ourselin S., Shi Y., Antonelli M., Tinker A., Liu W. Wearables, Sensors and the Future of Technology to Detect and Infer Loneliness in Older Adults. Gerontechnology. 2023;22:1–4. doi: 10.4017/gt.2023.22.2.ree.08. [ DOI ] [ Google Scholar ] 49. Rees J., Liu W., Canson J., Crosby L., Tinker A., Probst F., Ourselin S., Antonelli M., Molteni E., Mexia N., et al. Qualitative Exploration of the Lived Experiences of Loneliness in Later Life to Inform Technology Development. Int. J. Qual. Stud. Health Well-Being. 2024;19:2398259. doi: 10.1080/17482631.2024.2398259. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 50. Wang Q., Zhang S., Wang Y., Zhao D., Zhou C. Dual Sensory Impairment as a Predictor of Loneliness and Isolation in Older Adults: National Cohort Study. JMIR Public Health Surveill. 2022;8:e39314. doi: 10.2196/39314. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Takemoto A., Iwamoto M., Yaegashi H., Yun S., Takashima R. Virtual Avatar Communication Task Eliciting Pseudo-Social Isolation and Detecting Social Isolation Using Non-Verbal Signal Monitoring in Older Adults. Front. Psychol. 2025;16:1507178. doi: 10.3389/fpsyg.2025.1507178. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Yamada Y., Shinkawa K., Nemoto M., Arai T. Automatic Assessment of Loneliness in Older Adults Using Speech Analysis on Responses to Daily Life Questions. Front. Psychiatry. 2021;12:712251. doi: 10.3389/fpsyt.2021.712251. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Badal V.D., Nebeker C., Shinkawa K., Yamada Y., Rentscher K.E., Kim H.-C., Lee E.E. Do Words Matter? Detecting Social Isolation and Loneliness in Older Adults Using Natural Language Processing. Front. Psychiatry. 2021;12:728732. doi: 10.3389/fpsyt.2021.728732. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 54. Ji X., Li X., Yuh A., Kendell C., Watson A., Weimer J., Nagahara H., Higashino T., Mizumoto T., Erdelyi V., et al. Proceedings of the 8th ACM/IEEE International Conference on Connected Health: Applications, Systems and Engineering Technologies. ACM; New York, NY, USA: 2023. Short: Integrated Sensing Platform for Detecting Social Isolation and Loneliness In the Elderly Community; pp. 148–152. [ Google Scholar ] 55. Lee K., Lee T.C., Yefimova M., Kumar S., Puga F., Azuero A., Kamal A., Bakitas M.A., Wright A.A., Demiris G., et al. Using Digital Phenotyping to Understand Health-Related Outcomes: A Scoping Review. Int. J. Med. Inform. 2023;174:105061. doi: 10.1016/j.ijmedinf.2023.105061. [ DOI ] [ PubMed ] [ Google Scholar ] 56. Schutz N., Botros A., Hassen S.B., Saner H., Buluschek P., Urwyler P., Pais B., Santschi V., Gatica-Perez D., Muri R.M., et al. A Sensor-Driven Visit Detection System in Older Adults’ Homes: Towards Digital Late-Life Depression Marker Extraction. IEEE J. Biomed. Health Inform. 2022;26:1560–1569. doi: 10.1109/JBHI.2021.3114595. [ DOI ] [ PubMed ] [ Google Scholar ] 57. Badal V.D., Graham S.A., Depp C.A., Shinkawa K., Yamada Y., Palinkas L.A., Kim H.-C., Jeste D.V., Lee E.E. Prediction of Loneliness in Older Adults Using Natural Language Processing: Exploring Sex Differences in Speech. Am. J. Geriatr. Psychiatry. 2021;29:853–866. doi: 10.1016/j.jagp.2020.09.009. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Wang N., Goel S., Ibrahim S., Badal V.D., Depp C., Bilal E., Subbalakshmi K., Lee E. Decoding Loneliness: Can Explainable AI Help in Understanding Language Differences in Lonely Older Adults? Psychiatry Res. 2024;339:116078. doi: 10.1016/j.psychres.2024.116078. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 59. Sanchez W., Martinez A., Campos W., Estrada H., Pelechano V. Inferring Loneliness Levels in Older Adults from Smartphones. J. Ambient Intell. Smart Environ. 2015;7:85–98. doi: 10.3233/AIS-140297. [ DOI ] [ Google Scholar ] 60. Lin Y., Li C., Wang X., Li H. Development of a Machine Learning-Based Risk Assessment Model for Loneliness among Elderly Chinese: A Cross-Sectional Study Based on Chinese Longitudinal Healthy Longevity Survey. BMC Geriatr. 2024;24:939. doi: 10.1186/s12877-024-05443-x. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 61. Carrasco P.M., Crespo D.P., García A.I.R., Ibáñez M.L., Rubio B.M., Montenegro-Peña M. Predictive Factors and Risk and Protection Groups for Loneliness in Older Adults: A Population-Based Study. BMC Psychol. 2024;12:238. doi: 10.1186/s40359-024-01708-7. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 62. Sheikh M., Qassem M., Kyriacou P.A. Wearable, Environmental, and Smartphone-Based Passive Sensing for Mental Health Monitoring. Front. Digit. Health. 2021;3:662811. doi: 10.3389/fdgth.2021.662811. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 63. Robie A.A., Seagraves K.M., Egnor S.E.R., Branson K. Machine Vision Methods for Analyzing Social Interactions. J. Exp. Biol. 2017;220:25–34. doi: 10.1242/jeb.142281. [ DOI ] [ PubMed ] [ Google Scholar ] 64. Hackett K., Xu S., McKniff M., Paglia L., Barnett I., Giovannetti T. Mobility-Based Smartphone Digital Phenotypes for Unobtrusively Capturing Everyday Cognition, Mood, and Community Life-Space in Older Adults: Feasibility, Acceptability, and Preliminary Validity Study. JMIR Hum. Factors. 2024;11:e59974. doi: 10.2196/59974. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 65. Virginia Anikwe C., Friday Nweke H., Chukwu Ikegwu A., Adolphus Egwuonwu C., Uchenna Onu F., Rita Alo U., Wah Teh Y. Mobile and Wearable Sensors for Data-Driven Health Monitoring System: State-of-the-Art and Future Prospect. Expert Syst. Appl. 2022;202:117362. doi: 10.1016/j.eswa.2022.117362. [ DOI ] [ Google Scholar ] 66. Torrado J.C., Husebo B.S., Allore H.G., Erdal A., Fæø S.E., Reithe H., Førsund E., Tzoulis C., Patrascu M. Digital Phenotyping by Wearable-Driven Artificial Intelligence in Older Adults and People with Parkinson’s Disease: Protocol of the Mixed Method, Cyclic ActiveAgeing Study. PLoS ONE. 2022;17:e0275747. doi: 10.1371/journal.pone.0275747. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 67. Rawtaer I., Jabbar K.B.A., Subagdja B., Saffari E., Tan A.H. 299. Sensors in-home for elder wellbeing (sinew): Digital phenotyping and artificial intelligence for early detection of dementia. Int. J. Neuropsychopharmacol. 2025;28:ii67–ii68. doi: 10.1093/ijnp/pyaf052.136. [ DOI ] [ Google Scholar ] 68. Daniels K., Vonck S., Robijns J., Spooren A., Hansen D., Bonnechère B. Characterising Physical Activity Patterns in Community-Dwelling Older Adults Using Digital Phenotyping: A 2-Week Observational Study Protocol. BMJ Open. 2025;15:e095769. doi: 10.1136/bmjopen-2024-095769. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 69. Daniels K., Quadflieg K., Robijns J., De Vry J., Van Alphen H., Van Beers R., Sourbron B., Vanbuel A., Meekers S., Mattheeussen M., et al. From Steps to Context: Optimizing Digital Phenotyping for Physical Activity Monitoring in Older Adults by Integrating Wearable Data and Ecological Momentary Assessment. Sensors. 2025;25:858. doi: 10.3390/s25030858. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 70. Hackett K., Xu S., McKniff M., Pinsky E., Holmqvist S., Vallecorsa G., Barnett I., Giovannetti T. 3 Smartphone Digital Phenotyping for Unobtrusive and Continuous Assessment of Everyday Cognition and Movement Trajectories in Older Adults. J. Int. Neuropsychol. Soc. 2023;29:207–208. doi: 10.1017/S1355617723003132. [ DOI ] [ Google Scholar ] 71. Hackett K., Giovannetti T. Capturing Cognitive Aging in Vivo: Application of a Neuropsychological Framework for Emerging Digital Tools. JMIR Aging. 2022;5:e38130. doi: 10.2196/38130. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 72. Fan L.-J., Wang F.-Y., Zhao J.-H., Zhang J.-J., Li Y.-A., Tang J., Lin T., Wei Q. From Physical Activity Patterns to Cognitive Status: Development and Validation of Novel Digital Biomarkers for Cognitive Assessment in Older Adults. Int. J. Behav. Nutr. Phys. Act. 2025;22:11. doi: 10.1186/s12966-025-01706-x. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 73. Marino F.R., Wu H.-T., Etzkorn L., Rooney M.R., Soliman E.Z., Deal J.A., Crainiceanu C., Spira A.P., Wanigatunga A.A., Schrack J.A., et al. Associations of Physical Activity and Heart Rate Variability from a Two-Week ECG Monitor with Cognitive Function and Dementia: The ARIC Neurocognitive Study. Sensors. 2024;24:4060. doi: 10.3390/s24134060. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 74. Aledavood T., Luong N., Baryshnikov I., Darst R., Heikkilä R., Holmén J., Ikäheimonen A., Martikkala A., Riihimäki K., Saleva O., et al. Multimodal Digital Phenotyping Study in Patients with Major Depressive Episodes and Healthy Controls (Mobile Monitoring of Mood): Observational Longitudinal Study. JMIR Ment. Health. 2025;12:e63622. doi: 10.2196/63622. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 75. Price G.D., Heinz M.V., Song S.H., Nemesure M.D., Jacobson N.C. Using Digital Phenotyping to Capture Depression Symptom Variability: Detecting Naturalistic Variability in Depression Symptoms across One Year Using Passively Collected Wearable Movement and Sleep Data. Transl. Psychiatry. 2023;13:381. doi: 10.1038/s41398-023-02669-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 76. Sun S., Folarin A.A., Zhang Y., Cummins N., Garcia-Dias R., Stewart C., Ranjan Y., Rashid Z., Conde P., Laiou P., et al. Challenges in Using mHealth Data from Smartphones and Wearable Devices to Predict Depression Symptom Severity: Retrospective Analysis. J. Med. Internet Res. 2023;25:e45233. doi: 10.2196/45233. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 77. Lekkas D., Gyorda J.A., Price G.D., Jacobson N.C. Depression Deconstructed: Wearables and Passive Digital Phenotyping for Analyzing Individual Symptoms. Behav. Res. Ther. 2023;168:104382. doi: 10.1016/j.brat.2023.104382. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 78. Baig M.M., Afifi S., GholamHosseini H., Mirza F. A Systematic Review of Wearable Sensors and IoT-Based Monitoring Applications for Older Adults—A Focus on Ageing Population and Independent Living. J. Med. Syst. 2019;43:233. doi: 10.1007/s10916-019-1365-7. [ DOI ] [ PubMed ] [ Google Scholar ] 79. Stavropoulos T.G., Papastergiou A., Mpaltadoros L., Nikolopoulos S., Kompatsiaris I. IoT Wearable Sensors and Devices in Elderly Care: A Literature Review. Sensors. 2020;20:2826. doi: 10.3390/s20102826. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 80. Orlandini R., Lušić Kalcina L., Antičević V. Understanding Loneliness in Older Adults During the Pandemic: Predictors and Questionnaire Validation. Diseases. 2025;13:45. doi: 10.3390/diseases13020045. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 81. Cho E., Demiris G., Cho H., Lee I., Yuh A., Ji X., Harrison S.L., Sokolsky O. Perceptions and attitudes toward loneliness detection using passive sensing technologies in older adults. Sage Open Aging. 2024;11:782. doi: 10.1093/geroni/igae098.2541. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 82. Wilkowska W., Offermann J., Spinsante S., Poli A., Ziefle M. Analyzing Technology Acceptance and Perception of Privacy in Ambient Assisted Living for Using Sensor-Based Technologies. PLoS ONE. 2022;17:e0269642. doi: 10.1371/journal.pone.0269642. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 83. Yusif S., Soar J., Hafeez-Baig A. Older People, Assistive Technologies, and the Barriers to Adoption: A Systematic Review. Int. J. Med. Inform. 2016;94:112–116. doi: 10.1016/j.ijmedinf.2016.07.004. [ DOI ] [ PubMed ] [ Google Scholar ] 84. Mujirishvili T., Maidhof C., Florez-Revuelta F., Ziefle M., Richart-Martinez M., Cabrero-García J. Acceptance and Privacy Perceptions Toward Video-Based Active and Assisted Living Technologies: Scoping Review. J. Med. Internet Res. 2023;25:e45297. doi: 10.2196/45297. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 85. Pirzada P., Wilde A., Doherty G.H., Harris-Birtill D. Ethics and Acceptance of Smart Homes for Older Adults. Inform. Health Soc. Care. 2022;47:10–37. doi: 10.1080/17538157.2021.1923500. [ DOI ] [ PubMed ] [ Google Scholar ] 86. Percy Campbell J., Buchan J., Chu C.H., Bianchi A., Hoey J., Khan S.S. User Perception of Smart Home Surveillance Among Adults Aged 50 Years and Older: Scoping Review. JMIR Mhealth Uhealth. 2024;12:e48526. doi: 10.2196/48526. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 87. Read E.A., Gagnon D.A., Donelle L., Ledoux K., Warner G., Hiebert B., Sharma R. Stakeholder Perspectives on In-Home Passive Remote Monitoring to Support Aging in Place in the Province of New Brunswick, Canada: Rapid Qualitative Investigation. JMIR Aging. 2022;5:e31486. doi: 10.2196/31486. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 88. Miller L.M., Kaye J., Lindauer A., Au-Yeung W.-T.M., Rodrigues N.K., Czaja S.J. Remote Passive Sensing of Older Adults’ Activities and Function: User-Centered Design Considerations for Behavioral Interventions Conducted in the Home Setting. J. Med. Internet Res. 2024;26:e54709. doi: 10.2196/54709. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 89. Berridge C. Active Subjects of Passive Monitoring: Responses to a Passive Monitoring System in Low-Income Independent Living. Ageing Soc. 2017;37:537–560. doi: 10.1017/S0144686X15001269. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 90. Sharma J., Gillani N., Saied I., Alzaabi A., Arslan T. Patient and Public Involvement in the Co-Design and Assessment of Unobtrusive Sensing Technologies for Care at Home: A User-Centric Design Approach. BMC Geriatr. 2025;25:48. doi: 10.1186/s12877-024-05674-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 91. Zieni B., Ritchie M.A., Mandalari A.M., Boem F. An Interdisciplinary Overview on Ambient Assisted Living Systems for Health Monitoring at Home: Trade-Offs and Challenges. Sensors. 2025;25:853. doi: 10.3390/s25030853. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials sensors-26-02028-s001.zip (80.8KB, zip) Data Availability Statement No new data were created or analyzed in this study. Data sharing is not applicable to this article. Articles from Sensors (Basel, Switzerland) are provided here courtesy of Multidisciplinary Digital Publishing Institute (MDPI) ACTIONS View on publisher site PDF (2.7 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top