À retenir
In 37 adolescents with depression symptoms whose phone and Fitbit data were analyzed, personalized models predicted weekly depression scores with a root mean squared error of 2.83 and weekly change with 3.21. This is a very small exploratory study, and the abstract gives no comparison against simple baselines, so it does not show a usable screening tool.
Study design
Article de recherche
Reading basis
Published abstract
Reading focus
Design & applicability
Cette explication éditoriale existe en anglais, chinois simplifié et chinois traditionnel. Cette interface affiche la lecture en anglais.
Question de recherche
Depression is rising among adolescents, and many do not get treatment, so spotting worsening symptoms earlier could help. Most sensor-based prediction work has been done in adults, so the authors tested how well two machine learning approaches could predict adolescents' depression scores and changes, and which sensor features mattered most.
Population et méthode
Exploratory study of 55 adolescents aged 12 to 17 with depression symptoms, monitored for 24 weeks with smartphone sensors and Fitbit devices that collected call, conversation, location and heart rate data daily. After preprocessing, 67% (37/55) of the participants were analyzed. Weekly Patient Health Questionnaire-9 (PHQ-9) surveys served as the ground truth. Regression models, both linear and nonlinear, predicted PHQ-9 score and change in severity using universal strategies (Leave One Participant Out, Leave Week X Out) and personalized strategies (Accumulated Weeks, Leave One Week One User Instance Out).
Résultats
The authors’ reported results. Read each estimate with its population and uncertainty.
- 1Personalized approaches predicted adolescent depression better than universal approaches.
- 2The best models, using the Accumulated Weeks personalized strategy, predicted the depression score with a root mean squared error of 2.83 and the weekly change in depression level with a root mean squared error of 3.21.
- 3Screen-, call- and location-based features contributed most to the best models and were predictive of depression. The abstract does not report effects of individual heart rate features.
Limites et incertitude
- 1Only 37 of 55 participants (67%) had usable data after preprocessing, which is a small sample for machine learning and may be a non-random subset.
- 2Personalized models learn from each teen's own earlier weeks, so they need weeks of data first and may not work for a new user.
- 3Errors of 2.83 and 3.21 are reported without a baseline model such as predicting the average score, so how much better than guessing they are is not clear from the abstract.
- 4The ground truth was self-reported PHQ-9, not a clinical diagnosis, and the work is described by the authors as exploratory.
Mise en perspective
This work suggests passive phone and wearable data may someday help flag changes in teenagers' mood, but it is not a tool to diagnose or monitor depression now. If a young person seems low or their mood is changing, a conversation with them and their clinician or school counselor counts for far more than any app or tracker.
Vos données
Heart rate and activity from a wearable are not depression measures and shouldn't be read as one. If you track mood, a brief regular questionnaire alongside screen time, sleep and activity trends from the same devices may show your own patterns over weeks.
Comprendre ce type de preuve
The index does not establish a precise design. Read the methods before making causal claims; check population, comparator, outcomes and uncertainty in the source.
Résumé structuré des auteurs English
Le résumé original en anglais est reproduit avec attribution sous la licence ouverte indiquée. Il s’agit d’une source, pas d’une interprétation MoveVitals. L’analyse préparée figure ci-dessus ; vérifiez les conclusions importantes dans la source.
Background
Depression levels in adolescents have trended upward over the past several years. According to a 2020 survey by the National Survey on Drug Use and Health, 4.1 million US adolescents have experienced at least one major depressive episode. This number constitutes approximately 16% of adolescents aged 12 to 17 years. However, only 32.3% of adolescents received some form of specialized or nonspecialized treatment. Identifying worsening symptoms earlier using mobile and wearable sensors may lead to earlier intervention. Most studies on predicting depression using sensor-based data are geared toward the adult population. Very few studies look into predicting depression in adolescents.
Objective
The aim of our work was to study passively sensed data from adolescents with depression and investigate the predictive capabilities of 2 machine learning approaches to predict depression scores and change in depression levels in adolescents. This work also provided an in-depth analysis of sensor features that serve as key indicators of change in depressive symptoms and the effect of variation of data samples on model accuracy levels.
Methods
This study included 55 adolescents with symptoms of depression aged 12 to 17 years. Each participant was passively monitored through smartphone sensors and Fitbit wearable devices for 24 weeks. Passive sensors collected call, conversation, location, and heart rate information daily. Following data preprocessing, 67% (37/55) of the participants in the aggregated data set were analyzed. Weekly Patient Health Questionnaire-9 surveys answered by participants served as the ground truth. We applied regression-based approaches to predict the Patient Health Questionnaire-9 depression score and change in depression severity. These approaches were consolidated using universal and personalized modeling strategies. The universal strategies consisted of Leave One Participant Out and Leave Week X Out. The personalized strategy models were based on Accumulated Weeks and Leave One Week One User Instance Out. Linear and nonlinear machine learning algorithms were trained to model the data.
Results
We observed that personalized approaches performed better on adolescent depression prediction compared with universal approaches. The best models were able to predict depression score and weekly change in depression level with root mean squared errors of 2.83 and 3.21, respectively, following the Accumulated Weeks personalized modeling strategy. Our feature importance investigation showed that the contribution of screen-, call-, and location-based features influenced optimal models and were predictive of adolescent depression.
Conclusions
This study provides insight into the feasibility of using passively sensed data for predicting adolescent depression. We demonstrated prediction capabilities in terms of depression score and change in depression level. The prediction results revealed that personalized models performed better on adolescents than universal approaches. Feature importance provided a better understanding of depression and sensor data. Our findings can help in the development of advanced adolescent depression predictions.
Sources et attribution
Predicting Depression in Adolescents Using Mobile and Wearable Sensors: Multimodal Machine Learning-Based Exploratory Study.
Mullick T, Radovic A, Shaaban S, Doryab A.
JMIR formative research · 2022-06-24
DOI: 10.2196/35807 ↗PubMed · 35749157 ↗Europe PMC · Licence indiquée: cc by ↗Abstract © the original authors, reproduced under the recorded open licence. Translations support reading; use the English record to check terminology and numerical results. Our methodological interpretation is separate from the authors’ conclusions.
Interprétation éditoriale MoveVitals · contenu éducatif · aucune évaluation clinique individuelle enregistrée
Sources vérifiées 2026-11-01
Guides pratiques
À lire ensemble
Activity–stroke and heart attack curves differ by accelerometer processing method
Among 90,237 UK Biobank adults free of cardiovascular disease at baseline, followed until December 2022 for 1,298 strokes and 2,031 myocardial infarctions, the shape of the link between moderate-vigorous physical activity (MVPA) and each outcome depended on how the accelerometer data were processed. For stroke, hazard fell in a straight line only with the machine-learning metric; for myocardial infarction, two metrics gave a curved decline and activity counts a straight one. The abstract gives no hazard ratios, so it shows that methods disagree, not which one is right, and the findings are observational associations.
Preventive Medicine
Smartwatch data as forensic evidence: heart rate is best validated, the rest uneven
A systematic review and meta-analysis of studies on smartwatches from four major brands, framed around digital forensics, found that heart rate was both the most studied and the most accurate measure. Step count, distance, energy expenditure, activity classification and sleep were less consistently validated, and results varied by brand, protocol and reporting. The abstract gives no pooled numbers, and most studies were run in healthy people in laboratories, so it cannot say how a particular watch performs for a particular person in daily life.
Forensic Science International
Activity trackers: about 1,800 more steps a day, but small effects on other outcomes
An umbrella review of 39 systematic reviews and meta-analyses, covering 163,992 participants, found that activity trackers were associated with more physical activity (SMD 0.3–0.6), equating to about 1,800 extra steps and 40 minutes more walking per day, and with a bodyweight reduction of about 1 kg. Effects on blood pressure, cholesterol, glycosylated haemoglobin, quality of life and pain were typically small and often non-significant. The abstract does not say what else came with the trackers in these programs or how much the reviews overlap, so the gains cannot be credited to the device alone.
The Lancet Digital Health
Seven sleep trackers: good at spotting sleep, weak on wake and sleep stages
In a sleep laboratory, 34 healthy young adults wore a subset of seven consumer sleep trackers for three nights. Against polysomnography every device detected sleep very well (sensitivity 0.93 or higher) but caught wake poorly (specificity 0.18–0.54), and five of the seven matched or beat research actigraphy (specificity 0.39). Sleep-stage estimates were inconsistent, and the participants were healthy adults aged 18 to 35 in a fixed 8-hour bed window, with each device tested in only 11 to 20 of them, so the results say little about people with sleep problems.
Sleep

