Daniel Emden

Background: The benefits of physical activity (PA) for both physical and mental health, including major depressive disorder (MDD), are well established. Mobile devices, such as smartphones, offer a scalable way to monitor PA and its relationship with depressive symptoms in daily life. Objective: This study aimed to investigate the association between passive smartphone-recorded step counts and current depressive symptoms in individuals with and without a lifetime diagnosis of MDD, using a naturalistic bring-your-own-device approach. Methods: We used the Remote Monitoring Application in Psychiatry (ReMAP) to collect passive step count data from participants’ personal smartphones. The sample included 181 individuals with a lifetime MDD diagnosis, assessed via the structured clinical interview for the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition; DSM-IV ), and 195 healthy controls (HCs). Current depressive symptoms were assessed using the Beck Depression Inventory. PA was operationalized as daily and weekly step counts, passively recorded via smartphone sensors. Hierarchical models were applied to examine the association between PA and depression severity. Results: Patients with MDD exhibited significantly lower daily step counts (mean 3454, SD 2683) compared to HCs (mean 4699, SD 3175; P <.001) and showed reduced diurnal variability (β=–0.36; P =.003). Higher daily step counts were associated with lower Beck Depression Inventory scores across the full sample (β=–0.06, 95% CI –0.09 to –0.02; P =.002), with similar trends in both MDD and HC groups. Weekly step counts also significantly predicted lower concurrent depressive symptoms (β=–0.29, 95% CI –0.43 to –0.14; P <.001), while patients with MDD displayed less variability in weekly activity levels than HCs (β=–0.75; P =.001). Conclusions: These findings underscore the potential of mobile devices to be used as effective tools for monitoring PA in patients with MDD, supporting more customized and adaptive approaches to prevention and treatment. They also emphasize the importance of incorporating PA into the clinical management of depression.

We investigated whether the brain age gap (BAG)—the difference between chronological age and age estimated from structural MRI scans—is associated with long-term disease course in affective disorders, using a prospective nine-year follow-up design. T1-weighted MRI data were collected at two time points (mean interval = 8.98 ± 2.20 years) from patients with Major Depressive Disorder (MDD; N = 32), Bipolar Disorder (BD; N = 6), and healthy controls (HC; N = 37) across two sites. Using a brain age prediction model trained on a sample of over 10,000 subjects of the German National Cohort (GNC), we estimated individual BAG at baseline and follow-up using gray matter segments derived from MRI images. Employing linear-mixed-effects models, we tested main effects of diagnosis and hospitalizations as well as their interaction with time on BAG. In an exploratory analysis, we tested if BAG at baseline was predictive of hospitalizations during the nine-year follow-up using logistic regression and 10-fold nested cross-validation. MDD patients showed significantly higher BAG compared to HC (2.27 ± 5.68 vs. 1.00 ± 5.12 years, d = –0.23), while BAG in BD patients was descriptively elevated (4.71 ± 5.40 years). In the Münster subsample (N = 52), patients with at least one hospitalization had higher BAG than those without (4.16 ± 5.74 vs. 1.65 ± 5.41 years, d = –0.45). No group-by-time interaction was observed. Higher BAG at baseline predicted hospitalization during follow-up ( p = 0.035), although cross-validated prediction accuracy (64.3%) did not reach significance ( p = 0.071). BAG remained stable over time and was not influenced by future recurrence, supporting its role as a potential trait-like marker of vulnerability to illness recurrence. While exploratory, these findings suggest that BAG may capture individual risk for future hospitalization in affective disorders.

Objective, scalable biomarkers are needed for continuous monitoring of major depressive disorder (MDD). Smartphone-collected speech is promising, yet extracting clinically useful signals remains difficult. We analysed 3 151 weekly voice diaries from 284 German-speaking adults (128 MDD, 156 controls) and regressed Beck Depression Inventory (BDI) scores. Sentence embeddings from the open-source 8-billion-parameter Qwen3-8B model predicted scores with MAE = 4.45 and R ² = 0.35, explaining 16 more points of variance than the best traditional feature set (TF-IDF). Adding lexical–prosodic or TF-IDF features provided only marginal improvement (best MAE = 4.39). To interpret the embeddings we applied BERTopic and uncovered ten coherent themes; BDI scores peaked for “Persistent Low Mood” and “Pain Distress”, confirming clinical relevance. Large-language-model embeddings therefore capture the dominant signal of depression severity in everyday speech and, paired with interpretable topic analysis, offer a privacy-preserving, scalable route to digital mental-health phenotyping.

Background: Mobile devices for remote monitoring are inevitable tools to support treatment and patient care, especially in recurrent diseases such as major depressive disorder. The aim of this study was to learn if machine learning (ML) models based on longitudinal speech data are helpful in predicting momentary depression severity. Data analyses were based on a dataset including 30 inpatients during an acute depressive episode receiving sleep deprivation therapy in stationary care, an intervention inducing a rapid change in depressive symptoms in a relatively short period of time. Using an ambulatory assessment approach, we captured speech samples and assessed concomitant depression severity via self-report questionnaire over the course of 3 weeks (before, during, and after therapy). We extracted 89 speech features from the speech samples using the Extended Geneva Minimalistic Acoustic Parameter Set from the Open-Source Speech and Music Interpretation by Large-Space Extraction (audEERING) toolkit and the additional parameter speech rate. Objective: We aimed to understand if a multiparameter ML approach would significantly improve the prediction compared to previous statistical analyses, and, in addition, which mechanism for splitting training and test data was most successful, especially focusing on the idea of personalized prediction. Methods: To do so, we trained and evaluated a set of >500 ML pipelines including random forest, linear regression, support vector regression, and Extreme Gradient Boosting regression models and tested them on 5 different train-test split scenarios: a group 5-fold nested cross-validation at the subject level, a leave-one-subject-out approach, a chronological split, an odd-even split, and a random split. Results: In the 5-fold cross-validation, the leave-one-subject-out, and the chronological split approaches, none of the models were statistically different from random chance. The other two approaches produced significant results for at least one of the models tested, with similar performance. In total, the superior model was an Extreme Gradient Boosting in the odd-even split approach (R²=0.339, mean absolute error=0.38; both P<.001), indicating that 33.9% of the variance in depression severity could be predicted by the speech features. Conclusions: Overall, our analyses highlight that ML fails to predict depression scores of unseen patients, but prediction performance increased strongly compared to our previous analyses with multilevel models. We conclude that future personalized ML models might improve prediction performance even more, leading to better patient management and care.

Depression is a widely prevalent psychiatric illness with variable levels of severity that necessitate different approaches to treatment. To enhance the management of this condition, there is a growing interest in utilizing mobile devices, especially smartphones, for remote monitoring of patients. This study aims to build prediction models for depression severity based on active and passive features collected from patients with major depressive disorder (MDD) and healthy controls to assess the feasibility of remote monitoring of depression severity. Using data from 142 participants (85 healthy controls, 67 MDD) we extracted features such as GPS-derived mobility markers, ecological momentary assessments (EMA), age, and sex to develop machine learning models of depression severity on the different diagnostic subgroups in this cohort. Our results indicate that the employed models outperformed baseline estimators in random split scenarios. However, the improvement was marginal in user-split scenarios, highlighting the need for larger and more diverse samples for clinical utility. Among the features, mood EMA emerged as the most influential predictor, followed by GPS-derived mobility features. Models also showed a significant association between depression severity and average reported mood, as well as GPS-derived mobility markers such as number of places visited and percent home. While predicting composite depression scores is important, future studies could explore predicting individual symptom items or symptom groups for a more comprehensive assessment of depression severity. Challenges for clinical utility include participant dropout, which could be addressed through more engaging app design to promote user adherence. Harmonization of phone-derived measures is also crucial to facilitate model transfer across studies. In conclusion, this study contributes valuable evidence supporting the potential utility of smartphone data for mood state monitoring and predicting depression severity. Future research should focus on predicting depression further ahead in time and addressing the challenges identified to create more robust and effective depression monitoring solutions using smartphone-based data.

Many therapeutic interventions in psychiatry can be viewed as attempts to influence the brain’s large-scale, dynamic network state transitions. Building on connectome-based graph analysis and control theory, Network Control Theory is emerging as a powerful tool to quantify network controllability—i.e., the influence of one brain region over others regarding dynamic network state transitions. If and how network controllability is related to mental health remains elusive. Here, from Diffusion Tensor Imaging data, we inferred structural connectivity and inferred calculated network controllability parameters to investigate their association with genetic and familial risk in patients diagnosed with major depressive disorder (MDD, n = 692) and healthy controls (n = 820). First, we establish that controllability measures differ between healthy controls and MDD patients while not varying with current symptom severity or remission status. Second, we show that controllability in MDD patients is associated with polygenic scores for MDD and psychiatric cross-disorder risk. Finally, we provide evidence that controllability varies with familial risk of MDD and bipolar disorder as well as with body mass index. In summary, we show that network controllability is related to genetic, individual, and familial risk in MDD patients. We discuss how these insights into individual variation of network controllability may inform mechanistic models of treatment response prediction and personalized intervention-design in mental health.

A network-based quantification of brain aging uncovers and fixes a fundamental problem of all previous approaches.