RESUMO
BACKGROUND: Bipolar disorder is highly prevalent and consists of biphasic recurrent mood episodes of mania and depression, which translate into altered mood, sleep and activity alongside their physiological expressions. AIMS: The IdenTifying dIgital bioMarkers of illnEss activity and treatment response in BipolAr diSordEr with a novel wearable device (TIMEBASE) project aims to identify digital biomarkers of illness activity and treatment response in bipolar disorder. METHOD: We designed a longitudinal observational study including 84 individuals. Group A comprises people with acute episode of mania (n = 12), depression (n = 12 with bipolar disorder and n = 12 with major depressive disorder (MDD)) and bipolar disorder with mixed features (n = 12). Physiological data will be recorded during 48 h with a research-grade wearable (Empatica E4) across four consecutive time points (acute, response, remission and episode recovery). Group B comprises 12 people with euthymic bipolar disorder and 12 with MDD, and group C comprises 12 healthy controls who will be recorded cross-sectionally. Psychopathological symptoms, disease severity, functioning and physical activity will be assessed with standardised psychometric scales. Physiological data will include acceleration, temperature, blood volume pulse, heart rate and electrodermal activity. Machine learning models will be developed to link physiological data to illness activity and treatment response. Generalisation performance will be tested in data from unseen patients. RESULTS: Recruitment is ongoing. CONCLUSIONS: This project should contribute to understanding the pathophysiology of affective disorders. The potential digital biomarkers of illness activity and treatment response in bipolar disorder could be implemented in a real-world clinical setting for clinical monitoring and identification of prodromal symptoms. This would allow early intervention and prevention of affective relapses, as well as personalisation of treatment.
RESUMO
BACKGROUND: Lifestyle factors are being increasingly studied in bipolar disorder (BD) due to their possible effects on both course of disease and physical health. The aim of this study was to jointly describe and explore the interrelations between diet patterns, exercise, pharmacological treatment with course of disease and metabolic profile in BD. METHODS: The sample consisted of 66 euthymic or mild depressive individuals with BD. Clinical and metabolic outcomes were assessed, as well as pharmacological treatment or lifestyle habits (diet and exercise). Correlations were explored for different interrelations and a factor analysis of dietary patterns was performed. RESULTS: Adherence to the Mediterranean diet was low, seen in 37.9% of the patients and was positively associated with perceived quality of life. The amount of exercise was negatively associated with cholesterol levels, with 32.8% of participants rated as low active by International Physical Activity Questionnaire. There was a high prevalence of obesity (40.6%) and metabolic syndrome (29.7%). Users of lithium showed the best metabolic profile. Interestingly, three dietary patterns were identified: "vegetarian," "omnivore" and "Western." The key finding was the overall positive impact of the "vegetarian" pattern in BD, which was associated with reduced depression scores, better psychosocial functioning, and perceived quality of life, decreased body mass index, cholesterol, LDL and diastolic blood pressure. Nuts consumption was associated with a better metabolic profile. CONCLUSIONS: A vegetarian diet pattern was associated with both, better clinical and metabolic parameters, in patients with BD. Future studies should prioritize prospective and randomized designs to determine causal relationships, and potentially inform clinical recommendations.
RESUMO
BACKGROUND: Personal sensing, leveraging data passively and near-continuously collected with wearables from patients in their ecological environment, is a promising paradigm to monitor mood disorders (MDs), a major determinant of the worldwide disease burden. However, collecting and annotating wearable data is resource intensive. Studies of this kind can thus typically afford to recruit only a few dozen patients. This constitutes one of the major obstacles to applying modern supervised machine learning techniques to MD detection. OBJECTIVE: In this paper, we overcame this data bottleneck and advanced the detection of acute MD episodes from wearables' data on the back of recent advances in self-supervised learning (SSL). This approach leverages unlabeled data to learn representations during pretraining, subsequently exploited for a supervised task. METHODS: We collected open access data sets recording with the Empatica E4 wristband spanning different, unrelated to MD monitoring, personal sensing tasks-from emotion recognition in Super Mario players to stress detection in undergraduates-and devised a preprocessing pipeline performing on-/off-body detection, sleep/wake detection, segmentation, and (optionally) feature extraction. With 161 E4-recorded subjects, we introduced E4SelfLearning, the largest-to-date open access collection, and its preprocessing pipeline. We developed a novel E4-tailored transformer (E4mer) architecture, serving as the blueprint for both SSL and fully supervised learning; we assessed whether and under which conditions self-supervised pretraining led to an improvement over fully supervised baselines (ie, the fully supervised E4mer and pre-deep learning algorithms) in detecting acute MD episodes from recording segments taken in 64 (n=32, 50%, acute, n=32, 50%, stable) patients. RESULTS: SSL significantly outperformed fully supervised pipelines using either our novel E4mer or extreme gradient boosting (XGBoost): n=3353 (81.23%) against n=3110 (75.35%; E4mer) and n=2973 (72.02%; XGBoost) correctly classified recording segments from a total of 4128 segments. SSL performance was strongly associated with the specific surrogate task used for pretraining, as well as with unlabeled data availability. CONCLUSIONS: We showed that SSL, a paradigm where a model is pretrained on unlabeled data with no need for human annotations before deployment on the supervised target task of interest, helps overcome the annotation bottleneck; the choice of the pretraining surrogate task and the size of unlabeled data for pretraining are key determinants of SSL success. We introduced E4mer, which can be used for SSL, and shared the E4SelfLearning collection, along with its preprocessing pipeline, which can foster and expedite future research into SSL for personal sensing.
Assuntos
Transtornos do Humor , Aprendizado de Máquina Supervisionado , Dispositivos Eletrônicos Vestíveis , Humanos , Estudos Prospectivos , Dispositivos Eletrônicos Vestíveis/estatística & dados numéricos , Dispositivos Eletrônicos Vestíveis/normas , Masculino , Feminino , Transtornos do Humor/diagnóstico , Transtornos do Humor/psicologia , Adulto , Exercício Físico/psicologia , Exercício Físico/fisiologia , Universidades/estatística & dados numéricos , Universidades/organização & administraçãoRESUMO
BACKGROUND: Affective states influence the sympathetic nervous system, inducing variations in electrodermal activity (EDA), however, EDA association with bipolar disorder (BD) remains uncertain in real-world settings due to confounders like physical activity and temperature. We analysed EDA separately during sleep and wakefulness due to varying confounders and potential differences in mood state discrimination capacities. METHODS: We monitored EDA from 102 participants with BD including 35 manic, 29 depressive, 38 euthymic patients, and 38 healthy controls (HC), for 48 h. Fifteen EDA features were inferred by mixed-effect models for repeated measures considering sleep state, group and covariates. RESULTS: Thirteen EDA feature models were significantly influenced by sleep state, notably including phasic peaks (p < 0.001). During wakefulness, phasic peaks showed different values for mania (M [SD] = 6.49 [5.74, 7.23]), euthymia (5.89 [4.83, 6.94]), HC (3.04 [1.65, 4.42]), and depression (3.00 [2.07, 3.92]). Four phasic features during wakefulness better discriminated between HC and mania or euthymia, and between depression and euthymia or mania, compared to sleep. Mixed symptoms, average skin temperature, and anticholinergic medication affected the models, while sex and age did not. CONCLUSION: EDA measured from awake recordings better distinguished between BD states than sleep recordings, when controlled by confounders.
RESUMO
Background: Aerobic capacity has shown to predict physical and mental health-related quality of life in bipolar disorder (BD). However, the correlation between exercise respiratory capacity and mitochondrial function remains understudied. We aimed to assess longitudinally intra-individual differences in these factors during mood episodes and remission in BD. Methods: This study included eight BD patients admitted to an acute psychiatric unit. Incremental cardiopulmonary exercise test (CPET) was conducted during acute episodes (T0), followed by constant work rate cycle ergometry (CWRCE) to evaluate endurance time, oxygen uptake at peak exercise (VO2peak) and at the anaerobic threshold. The second test was repeated during remission (T1). Mitochondrial respiration rates were assessed at T0 and T1 in peripheral blood mononuclear cells. Results: Endurance time, VO2peak, and anaerobic threshold oxygen consumption showed no significant variations between T0 and T1. Basal oxygen consumption at T1 tended to inversely correlate with maximal mitochondrial respiratory capacity (r=-0.690, p=0.058), and VO2peak during exercise at T1 inversely correlated with basal and minimum mitochondrial respiration (r=-0.810, p=0.015; r=-0.786, p=0.021, respectively). Conclusions: Our preliminary data showed that lower basal oxygen consumption may be linked to greater mitochondrial respiratory capacity, and maximum oxygen uptake during the exercise task was associated with lower basal mitochondrial respiration, suggesting that lower oxygen requirements could be associated with greater mitochondrial capacity. These findings should be replicated in larger samples stratified for manic and depressive states.
RESUMO
Mood disorders (MDs) are among the leading causes of disease burden worldwide. Limited specialized care availability remains a major bottleneck thus hindering pre-emptive interventions. MDs manifest with changes in mood, sleep, and motor activity, observable in ecological physiological recordings thanks to recent advances in wearable technology. Therefore, near-continuous and passive collection of physiological data from wearables in daily life, analyzable with machine learning (ML), could mitigate this problem, bringing MDs monitoring outside the clinician's office. Previous works predict a single label, either the disease state or a psychometric scale total score. However, clinical practice suggests that the same label may underlie different symptom profiles, requiring specific treatments. Here we bridge this gap by proposing a new task: inferring all items in HDRS and YMRS, the two most widely used standardized scales for assessing MDs symptoms, using physiological data from wearables. To that end, we develop a deep learning pipeline to score the symptoms of a large cohort of MD patients and show that agreement between predictions and assessments by an expert clinician is clinically significant (quadratic Cohen's κ and macro-average F1 score both of 0.609). While doing so, we investigate several solutions to the ML challenges associated with this task, including multi-task learning, class imbalance, ordinal target variables, and subject-invariant representations. Lastly, we illustrate the importance of testing on out-of-distribution samples.
Assuntos
Afeto , Transtornos do Humor , Humanos , Transtornos do Humor/diagnóstico , Aprendizado de Máquina , SonoRESUMO
BACKGROUND: Bipolar disorder (BD) is a chronic and recurrent disease characterized by acute mood episodes and periods of euthymia. The available literature postulates that a biphasic dysregulation of mitochondrial bioenergetics might underpin the neurobiology of BD. However, most studies focused on inter-subject differences rather than intra-subject variations between different mood states. To test this hypothesis, in this preliminary proof-of-concept study, we measured in vivo mitochondrial respiration in patients with BD during a mood episode and investigated differences compared to healthy controls (HC) and to the same patients upon clinical remission. METHODS: This longitudinal study recruited 20 patients with BD admitted to our acute psychiatric ward with a manic (n = 15) or depressive (n = 5) episode, and 10 matched HC. We assessed manic and depressive symptoms using standardized psychometric scales. Different mitochondrial oxygen consumption rates (OCRs: Routine, Leak, electron transport chain [ETC], Rox) were assessed during the acute episode (T0) and after clinical remission (T1) using high-resolution respirometry at 37°C by polarographic oxygen sensors in a two-chamber Oxygraph-2k system in one million of peripheral blood mononuclear cells (PMBC). Specific OCRs were expressed as mean ± SD in picomoles of oxygen per million cells. Significant results were adjusted for age, sex, and body mass index. RESULTS: The longitudinal analysis showed a significant increase in the maximal oxygen consumption capacity (ETC) in clinical remission (25.7 ± 16.7) compared to the acute episodes (19.1 ± 11.8, p = 0.025), and was observed separately for patients admitted with a manic episode (29.2 ± 18.9 in T1, 22.3 ± 11.9 in T0, p = 0.076), and at a trend-level for patients admitted with a depressive episode (15.4 ± 3.9 in T1 compared to 9.4 ± 3.2 in T0, p = 0.107). Compared to HC, significant differences were observed in ETC in patients with a bipolar mood episode (H = 11.7; p = 0.003). Individuals with bipolar depression showed lower ETC than those with a manic episode (t = -3.7, p = 0.001). Also, significant differences were observed in ETC rates between HC and bipolar depression (Z = 1.000, p = 0.005). CONCLUSIONS: Bioenergetic and mitochondrial dysregulation could be present in both manic and depressive phases in BD and, importantly, they may restore after clinical remission. These preliminary results suggest that mitochondrial respiratory capacity could be a biomarker of illness activity and clinical response in BD. Further studies with larger samples and similar approaches are needed to confirm these results and identify potential biomarkers in different phases of the disease.