Utility of Consumer-Grade Wearable Devices for Inferring Physical and Mental Health Outcomes in Severe Mental Illness: Systematic Review

JMIR Mental Health, Journal Year: 2025, Volume and Issue: 12, P. e65143 - e65143

Published: Jan. 7, 2025

Background Digital wearable devices, worn on or close to the body, have potential for passively detecting mental and physical health symptoms among people with severe illness (SMI); however, roles of consumer-grade devices are not well understood. Objective This study aims examine utility data from consumer-grade, digital, (including smartphones wrist-worn devices) remotely monitoring predicting changes in adults schizophrenia bipolar disorder. Studies were included that collected physiological sleep duration, heart rate, wake patterns, activity) at least 3 days. Research-grade actigraphy methods physically obtrusive excluded. Methods We conducted a systematic review following databases: Cochrane Central Register Controlled Trials, Technology Assessment, AMED (Allied Complementary Medicine), APA PsycINFO, Embase, MEDLINE(R), IEEE XPlore. Searches completed May 2024. Results synthesized narratively due heterogeneity divided into phenotypes: activity, circadian rhythm, rate. Overall, 23 studies reported 12 distinct studies, mostly using centered relapse prevention. Only 1 explicitly aimed address outcomes SMI. In total, over 500 participants SMI, predominantly high-income countries. Most commonly, papers presented activity (n=18), followed by rhythm (n=14) rate (n=6). The use smartwatches support collection 8 papers; rest used only smartphones. There was some evidence lower levels higher rates, later irregular onset times associated psychiatric diagnoses poorer symptoms. However, measures, sampling statistical approaches complicated interpretation. Conclusions Consumer-grade wearables show ability detect digital markers indicative status but few currently these inequalities. phenotyping field psychiatry would benefit moving toward agreed standards regarding descriptions outcome measures ensuring valuable temporal provided fully exploited. Trial Registration PROSPERO CRD42022382267; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=382267

Language: Английский

---

Artificial Intelligence−Powered Electrochemical Sensor: Recent Advances, Challenges, and Prospects

Heliyon, Journal Year: 2024, Volume and Issue: 10(18), P. e37964 - e37964

Published: Sept. 1, 2024

Language: Английский

---

Multimodal Mental Health Digital Biomarker Analysis From Remote Interviews Using Facial, Vocal, Linguistic, and Cardiovascular Patterns

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2024, Volume and Issue: 28(3), P. 1680 - 1691

Published: Jan. 10, 2024

Psychiatric evaluation suffers from subjectivity and bias, is hard to scale due intensive professional training requirements. In this work, we investigated whether behavioral physiological signals, extracted tele-video interviews, differ in individuals with psychiatric disorders.

Language: Английский

---

Benefits and Harms of Large Language Models in Digital Mental Health

Published: Nov. 15, 2023

The past decade has been transformative for mental health research and practice. ability to harness large repositories of data, whether from electronic records (EHR), mobile devices, or social media, revealed a potential valuable insights into patient experiences, promising early, proactive interventions, as well personalized treatment plans. Recent developments in generative artificial intelligence, particularly language models (LLMs), show promise leading digital uncharted territory. Patients are arriving at doctors' appointments with information sourced chatbots, state-of-the-art LLMs being incorporated medical software EHR systems, chatbots an ever-increasing number startups serve AI companions, friends, partners. This article presents contemporary perspectives on the opportunities risks posed by design, development, implementation tools. We adopt ecological framework draw affordances offered discuss four application areas---care-seeking behaviors individuals need care, community care provision, institutional larger ecologies societal level. engage thoughtful consideration how LLM-based technologies could should be employed enhancing health. benefits harms our surfaces help shape future research, advocacy, regulatory efforts focused creating more responsible, user-friendly, equitable, secure tools intervention.

Language: Английский

---

Behind the Screen: A Narrative Review on the Translational Capacity of Passive Sensing for Mental Health Assessment

Deleted Journal, Journal Year: 2024, Volume and Issue: 2(2), P. 778 - 810

Published: Feb. 22, 2024

Mental health disorders—including depression, anxiety, trauma-related, and psychotic conditions—are pervasive impairing, representing considerable challenges for both individual well-being public health. Often the first to treatment include financial, geographic, stigmatic barriers, which limit accessibility of traditional assessment measures. Further, compounded by frequent misdiagnosis or delayed detection, there is a need effective, accessible, scalable approaches identification management. Considering advances in computing ubiquitous nature personal mobile wearable technology, this narrative review examines utilization passive sensor data as screening diagnostic tool mental disorders. As an alternative measures, sensing offers overcome barriers that prevent many from seeking services. We critically assess literature up September 2023, exploring use data—such heart rate variability, movement patterns, geolocation—to predict outcomes across spectrum From translational perspective, our explores state science, with special emphasis on capacity science be implemented real world clinical general populations, novelty specific best knowledge. Toward aim, we consider multiple study factors, including participant demographics, collection methods, modalities, outcome analytic modeling approaches. find features, such GPS, rate, actigraphy offer promise enhancing early detection improving process Despite promise, however, findings highlight important limitations research (1) trend toward smaller, specialized samples, (2) predominance apps built Android operating system, (3) reliance self-reported measures proxies outcomes. These ultimately stymie efforts implement scale larger more heterogeneous populations. With future mind, emphasize importance validating larger, diverse samples ensuring tools can deployed device types systems. where possible, robust, objectively validated clinician assessment. conclude careful consideration factors design will aid impact studies, broad scale.

Language: Английский

---

Objective monitoring of loneliness levels using smart devices: A multi-device approach for mental health applications

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(6), P. e0298949 - e0298949

Published: June 20, 2024

Loneliness is linked to wide ranging physical and mental health problems, including increased rates of mortality. Understanding how loneliness manifests important for targeted public treatment intervention. With advances in mobile sending wearable technologies, it possible collect data on human phenomena a continuous uninterrupted way. In doing so, such approaches can be used monitor physiological behavioral aspects relevant an individual’s loneliness. this study, we proposed method detection using fully objective from smart devices passive sensing. We also investigated whether features differed their importance predicting across individuals. Finally, examined informative each device tasks. assessed subjective feelings while monitoring patterns 30 college students over 2-month period. smartphones (e.g., location changes, type notifications, in-coming out-going calls/text messages) watches rings physiology sleep heart-rate, heart-rate variability, duration). Participants reported feeling multiple times day through questionnaire app phone. Using the collected devices, trained random forest machine learning based model detect levels. found support prediction multi-device fully-objective approach. Furthermore, by generally were most all participants. The study provides promising results indicators, which could provide source information healthcare applications.

Language: Английский

---

Machine learning applied to digital phenotyping: A systematic literature review and taxonomy

Computers in Human Behavior, Journal Year: 2024, Volume and Issue: 161, P. 108422 - 108422

Published: Aug. 24, 2024

Language: Английский

---

Neuroscience meets behavior: A systematic literature review on magnetic resonance imaging of the brain combined with real‐world digital phenotyping

Human Brain Mapping, Journal Year: 2024, Volume and Issue: 45(4)

Published: March 1, 2024

Abstract A primary goal of neuroscience is to understand the relationship between brain and behavior. While magnetic resonance imaging (MRI) examines structure function under controlled conditions, digital phenotyping via portable automatic devices (PAD) quantifies behavior in real‐world settings. Combining these two technologies may bridge gap imaging, physiology, real‐time behavior, enhancing generalizability laboratory clinical findings. However, use MRI data from PADs outside scanner remains underexplored. Herein, we present a Preferred Reporting Items for Systematic Reviews Meta‐Analysis systematic literature review that identifies analyzes current state research on integration PADs. PubMed Scopus were automatically searched using keywords covering various techniques Abstracts screened only include articles collected PAD environment. Full‐text screening was then conducted ensure included combined quantitative with PADs, yielding 94 selected papers total N = 14,778 subjects. Results reported as cross‐frequency tables sampling methods patterns identified through network analysis. Furthermore, maps studies synthesized according measurement modalities used. demonstrate feasibility integrating across study designs, patient control populations, age groups. The majority published combines functional, T1‐weighted, diffusion weighted physical activity sensors, ecological momentary assessment sleep. further highlights specific regions frequently correlated distinct MRI‐PAD combinations. These combinations enable in‐depth how influence each other. Our potential constructing brain–behavior models extend beyond into contexts.

Language: Английский

---

Optimizing personalized psychological well-being interventions through digital phenotyping: results from a randomized non-clinical trial

Frontiers in Psychology, Journal Year: 2025, Volume and Issue: 15

Published: Jan. 6, 2025

Digital technologies, including smartphones, hold great promise for expanding mental health services and improving access to care. phenotyping, which involves the collection of behavioral physiological data using offers a novel way understand monitor health. This study examines feasibility psychological well-being program telegram-integrated chatbot digital phenotyping. A one-month randomized non-clinical trial was conducted with 81 young adults aged 18-35 from Italy canton Ticino, region in southern Switzerland. Participants were an experimental group that interacted chatbot, or control received general information on well-being. The collected real-time participants' such as user-chatbot interactions, responses exercises, emotional metrics. clustering algorithm created user profile content recommendation system provide personalized exercises based users' responses. Four distinct clusters participants emerged, factors online alerts, social media use, insomnia, attention energy levels. reported improvements found recommended by useful. demonstrates phenotyping-based chatbot. Despite limitations small sample size short duration, findings suggest phenotyping systems could improve Future research should include larger samples longer follow-up periods validate these explore clinical applications.

Language: Английский

---

Psychological well-being, gender, and age-specific difference on objectively recorded smartphone screen time in Japanese adults: A regression and clustering analysis

Computers in Human Behavior Reports, Journal Year: 2025, Volume and Issue: unknown, P. 100612 - 100612

Published: Feb. 1, 2025

Language: Английский

---