Mobile Monitoring of Mood (MoMo-Mood): a Multimodal Digital Phenotyping Study with Major Depressive Patients and Healthy Controls (Preprint) DOI Creative Commons
Talayeh Aledavood, Nguyen Luong, Ilya Baryshnikov

и другие.

JMIR Mental Health, Год журнала: 2024, Номер 12, С. e63622 - e63622

Опубликована: Дек. 4, 2024

Mood disorders are among the most common mental health conditions worldwide. Wearables and consumer-grade personal digital devices create traces that can be collected, processed, analyzed, offering a unique opportunity to quantify monitor individuals with in their natural living environments. This study comprised (1) 3 subcohorts of patients major depressive episode, either disorder, bipolar or concurrent borderline personality (2) healthy control group. We investigated whether differences behavioral patterns could observed at group level, is, versus controls. studied volume temporal smartphone screen app use, communication, sleep, mobility, physical activity. controls exhibited more homogenous activity when compared other same examined which variables were associated severity depression. In total, 188 participants recruited complete 2-phase study. first 2 weeks, data from bed sensors, actigraphy, smartphones, 5 sets daily questions collected. second phase, lasted up 1 year, only passive biweekly 9-item Patient Health Questionnaire Survival analysis, statistical tests, linear mixed models performed. analysis showed no statistically significant difference adherence. Most did not stay for year. Weekday location variance lower values (control: mean -10.04, SD 2.73; patient: -11.91, 2.50; Mann-Whitney U [MWU] test P=.004). Normalized entropy was 2.10, 1.38; 1.57, 1.10; MWU P=.05). The communication diverse those (MWU P<.001). contrast, varied use found duration incoming calls (β=-0.08, 95% CI -0.12 -0.04; P<.001) magnitude (β=-2.05, -4.18 -0.20; P=.02) over 14 days before records negatively depression severity. Conversely, outgoing positive association (β=0.05, 0.00-0.09; P=.02). Our work shows important features future analyses markers mood disorders. However, outpatients mild moderate disorders, group-level any single modality remain relatively modest. Therefore, studies need combine multiple modalities detect subtle identify individualized signatures. high dropout rates longer periods challenge limit generalizability.

Язык: Английский

Digital phenotyping using smartphones could help steer mental health treatment DOI Creative Commons

David Adam

Proceedings of the National Academy of Sciences, Год журнала: 2025, Номер 122(14)

Опубликована: Апрель 2, 2025

Язык: Английский

Процитировано

0
Development and validation of a predictive model for depression in patients with advanced stage of cardiovascular-kidney-metabolic syndrome DOI
Bowen Zha,

Angshu Cai,

Hao Yu

и другие.

Journal of Affective Disorders, Год журнала: 2025, Номер unknown

Опубликована: Апрель 1, 2025

Язык: Английский

Процитировано

0
Multi-Modal Sleep Measurement and Alignment Analysis in Outpatients with Major Depressive Episode DOI Creative Commons

Afrooz Mahir,

Nguyen Luong, Ilya Baryshnikov

и другие.

medRxiv (Cold Spring Harbor Laboratory), Год журнала: 2025, Номер unknown

Опубликована: Апрель 30, 2025

Abstract Study Objectives Sleep plays a crucial role for mental health. This study examines sleep tracking in naturalistic settings patients with major depressive episodes (MDE) using actigraphy, smartphone data, bed sensors, and the ecological momentary assessment (EMA) assesses discrepancies between these modalities. Methods We measured onset, offset, total time (TST) over two weeks 172 participants, including healthy controls three MDE subgroups (borderline personality disorder, bipolar disorder). Agreement measurement modalities was assessed Bland-Altman plots Pearson correlation. Predictors of alignment were analyzed mixed-effects models, accounting demographics, daylight length, participant subgroup. Results Patients showed greater variability than controls. Actigraphy overestimated TST compared to sensors (0.48 min) smartphones (0.99 min), while underestimated other Older age improved actigraphy as well sensor offset. (smartphone vs. sensor) worse females bipolar/borderline patients. Longer duration offset across Conclusions Our highlights biases, seasonal effects, demographic factors associated objective measures. While show potential offer several advantages assessing longer periods, misalignment should be considered future studies or clinical settings. Statement Significance Tracking psychiatric is challenging due frequent disturbances, making accurate diagnosis care. Traditional methods are limited lab settings, restricting long-term monitoring. evaluates self-reports both individuals disorders. findings demonstrate feasibility non-invasive monitor episodes. uncover systematic biases estimates modalities, reveal environmental that influence agreement, populations exhibit more patterns. work addresses critical gap validating consumer-grade technologies contexts.

Язык: Английский

Процитировано

0
Mobile Monitoring of Mood (MoMo-Mood): a Multimodal Digital Phenotyping Study with Major Depressive Patients and Healthy Controls (Preprint) DOI Creative Commons
Talayeh Aledavood, Nguyen Luong, Ilya Baryshnikov

и другие.

JMIR Mental Health, Год журнала: 2024, Номер 12, С. e63622 - e63622

Опубликована: Дек. 4, 2024

Mood disorders are among the most common mental health conditions worldwide. Wearables and consumer-grade personal digital devices create traces that can be collected, processed, analyzed, offering a unique opportunity to quantify monitor individuals with in their natural living environments. This study comprised (1) 3 subcohorts of patients major depressive episode, either disorder, bipolar or concurrent borderline personality (2) healthy control group. We investigated whether differences behavioral patterns could observed at group level, is, versus controls. studied volume temporal smartphone screen app use, communication, sleep, mobility, physical activity. controls exhibited more homogenous activity when compared other same examined which variables were associated severity depression. In total, 188 participants recruited complete 2-phase study. first 2 weeks, data from bed sensors, actigraphy, smartphones, 5 sets daily questions collected. second phase, lasted up 1 year, only passive biweekly 9-item Patient Health Questionnaire Survival analysis, statistical tests, linear mixed models performed. analysis showed no statistically significant difference adherence. Most did not stay for year. Weekday location variance lower values (control: mean -10.04, SD 2.73; patient: -11.91, 2.50; Mann-Whitney U [MWU] test P=.004). Normalized entropy was 2.10, 1.38; 1.57, 1.10; MWU P=.05). The communication diverse those (MWU P<.001). contrast, varied use found duration incoming calls (β=-0.08, 95% CI -0.12 -0.04; P<.001) magnitude (β=-2.05, -4.18 -0.20; P=.02) over 14 days before records negatively depression severity. Conversely, outgoing positive association (β=0.05, 0.00-0.09; P=.02). Our work shows important features future analyses markers mood disorders. However, outpatients mild moderate disorders, group-level any single modality remain relatively modest. Therefore, studies need combine multiple modalities detect subtle identify individualized signatures. high dropout rates longer periods challenge limit generalizability.

Язык: Английский

Процитировано

0