Digital Communication Biomarkers of Mood and Diagnosis in Borderline Personality Disorder, Bipolar Disorder, and Healthy Control Populations.

Background: Remote monitoring and digital phenotyping harbor potential to aid clinical diagnosis, predict episode course and recognize early signs of mental health crises. Digital communication metrics, such as phone call and short message service (SMS) use may represent novel biomarkers of mood and diagnosis in Bipolar Disorder (BD) and Borderline Personality Disorder (BPD). Materials and Methods: BD (n = 17), BPD (n = 17) and Healthy Control (HC, n = 21) participants used a smartphone application which monitored phone calls and SMS messaging, alongside self-reported mood. Linear mixed-effects regression models were used to assess the association between digital communications and mood symptoms, mood state, trait-impulsivity, diagnosis and the interaction effect between mood and diagnosis. Results: Transdiagnostically, self-rated manic symptoms and manic state were positively associated with total and outgoing call frequency and cumulative total, incoming and outgoing call duration. Manic symptoms were also associated with total and outgoing SMS frequency. Transdiagnostic depressive symptoms were associated with increased mean incoming call duration. For the different diagnostic groups, BD was associated with increased total call frequency and BPD with increased total and outgoing SMS frequency and length compared to HC. Depression in BD, but not BPD, was associated with decreased total and outgoing call frequency, mean total and outgoing call duration and total and outgoing SMS frequency. Finally, trait-impulsivity was positively associated with total call frequency, total and outgoing SMS frequency and cumulative total and outgoing SMS length. Conclusion: These results identify a general increase in phone call and SMS communications associated with self-reported manic symptoms and a diagnosis-moderated decrease in communications associated with depression in BD, but not BPD, participants. These findings may inform the development of clinical tools to aid diagnosis and remote symptom monitoring, as well as informing understanding of differential psychopathologies in BD and BPD.

Machine learning for nocturnal mass diagnosis of atrial fibrillation in a population at risk of sleep-disordered breathing.

OBJECTIVE: In this research, we introduce a new methodology for atrial fibrillation (AF) diagnosis during sleep in a large population sample at risk of sleep-disordered breathing. APPROACH: The approach leverages digital biomarkers and recent advances in machine learning (ML) for mass AF diagnosis from overnight-hours of single-channel electrocardiogram (ECG) recording. Four databases, totaling n = 3088 patients and p = 26 913 h of continuous single-channel electrocardiogram raw data were used. Three of the databases (n = 125, p = 2513) were used for training a ML model in recognizing AF events from beat-to-beat time series. Visit 1 of the sleep heart health study database (SHHS1, n = 2963, p = 24 400) was used as the test set to evaluate the feasibility of identifying prominent AF from polysomnographic recordings. By combining AF diagnosis history and a cardiologist's visual inspection of individuals suspected of having AF (n = 118), a total of 70 patients were diagnosed with prominent AF in SHHS1. MAIN RESULTS: Model prediction on SHHS1 showed an overall [Formula: see text]and [Formula: see text] in classifying individuals with or without prominent AF. [Formula: see text] was non-inferior (p = 0.03) for individuals with an apnea-hypopnea index (AHI) ≥15 versus AHI < [Formula: see text]. Over 22% of correctly identified prominent AF rhythm cases were not previously documented as AF in SHHS1. SIGNIFICANCE: Individuals with prominent AF can be automatically diagnosed from an overnight single-channel ECG recording, with an accuracy unaffected by the presence of moderate-to-severe obstructive sleep apnea. This approach enables identifying a large proportion of AF individuals that were otherwise missed by regular care.

Single-channel oximetry monitor versus in-lab polysomnography oximetry analysis: does it make a difference?

OBJECTIVE: Portable oximetry has been shown to be a promising candidate for large-scale obstructive sleep apnea screening. In polysomnography (PSG), the gold standard OSA diagnosis test, the oxygen desaturation index (ODI) is usually computed from desaturation events occurring during sleep periods only, i.e. overnight desaturations occurring during or overlapping with a wake state are excluded. However, for unattended home oximetry, all desaturations are taken into account since no reference electroencephalogram is available for sleep staging. We aim to evaluate the hypothesis that the predictive power of oximetry for OSA screening is not impaired when reference sleep stages are not available. APPROACH: We used a PSG clinical database of 887 individuals from a representative São Paulo (Brazil) population sample. Using features derived from the oxygen saturation time series and demographic information, OxyDOSA, a published machine learning model, was trained to distinguish between non-OSA and OSA individuals using the ODI computed while including versus excluding overnight desaturations overlapping with a wake period, thus mimicking portable and PSG oximetry analyses, respectively. MAIN RESULTS: When excluding wake desaturations, the OxyDOSA model had an AUROC = 94.9 ± 1.6, Se = 85.9 ± 2.8, Sp = 90.1 ± 2.6 and F1 = 86.4 ± 2.7. When considering wake desaturations, the OxyDOSA model had an AUROC = 94.4 ± 1.6, Se = 88.0 ± 2.0, Sp = 87.7 ± 2.9 and F1 = 86.2 ± 2.4. Non-inferiority was demonstrated (p = 0.049) at a tolerance level of 3%. In addition, analysis of the desaturations excluded by PSG oximetry analysis suggests that up to 21% of the total number of desaturations might actually be related to apneas or hypopneas. SIGNIFICANCE: This analysis of a large representative population sample provided strong evidence that the predictive power of oximetry for OSA screening using the OxyDOSA model is not impaired when reference sleep stages are not available. This finding motivates the usage of portable oximetry for OSA screening.

Feasibility of Single Channel Oximetry for Mass Screening of Obstructive Sleep Apnea.

BACKGROUND: The growing awareness for the high prevalence of obstructive sleep apnea (OSA) coupled with the dramatic proportion of undiagnosed individuals motivates the elaboration of a simple but accurate screening test. This study assesses, for the first time, the performance of oximetry combined with demographic information as a screening tool for identifying OSA in a representative (i.e. non-referred) population sample. METHODS: A polysomnography (PSG) clinical database of 887 individuals from a representative population sample of São Paulo's city (Brazil) was used. Using features derived from the oxygen saturation signal during sleep periods and demographic information, a logistic regression model (termed OxyDOSA) was trained to distinguish between non-OSA and OSA individuals (mild, moderate, and severe). The OxyDOSA model performance was assessed against the PSG-based diagnosis of OSA (AASM 2017) and compared to the NoSAS and STOP-BANG questionnaires. FINDINGS: The OxyDOSA model had mean AUROC = 0.94 ±â€¯0.02, Se = 0.87 ±â€¯0.04 and Sp = 0.85 ±â€¯0.03. In particular, it did not miss any of the 75 severe OSA individuals. In comparison, the NoSAS questionnaire had AUROC = 0.83 ±â€¯0.03, and missed 23/75 severe OSA individuals. The STOP-BANG had AUROC = 0.77 ±â€¯0.04 and missed 14/75 severe OSA individuals. INTERPRETATION: We provide strong evidence on a representative population sample that oximetry biomarkers combined with few demographic information, the OxyDOSA model, is an effective screening tool for OSA. Our results suggest that sleep questionnaires should be used with caution for OSA screening as they fail to identify many moderate and even some severe cases. The OxyDOSA model will need to be further validated on data recorded using overnight portable oximetry.

Group-Personalized Regression Models for Predicting Mental Health Scores From Objective Mobile Phone Data Streams: Observational Study.

BACKGROUND: Objective behavioral markers of mental illness, often recorded through smartphones or wearable devices, have the potential to transform how mental health services are delivered and to help users monitor their own health. Linking objective markers to illness is commonly performed using population-level models, which assume that everyone is the same. The reality is that there are large levels of natural interindividual variability, both in terms of response to illness and in usual behavioral patterns, as well as intraindividual variability that these models do not consider. OBJECTIVE: The objective of this study was to demonstrate the utility of splitting the population into subsets of individuals that exhibit similar relationships between their objective markers and their mental states. Using these subsets, "group-personalized" models can be built for individuals based on other individuals to whom they are most similar. METHODS: We collected geolocation data from 59 participants who were part of the Automated Monitoring of Symptom Severity study at the University of Oxford. This was an observational data collection study. Participants were diagnosed with bipolar disorder (n=20); borderline personality disorder (n=17); or were healthy controls (n=22). Geolocation data were collected using a custom Android app installed on participants' smartphones, and participants weekly reported their symptoms of depression using the 16-item quick inventory of depressive symptomatology questionnaire. Population-level models were built to estimate levels of depression using features derived from the geolocation data recorded from participants, and it was hypothesized that results could be improved by splitting individuals into subgroups with similar relationships between their behavioral features and depressive symptoms. We developed a new model using a Dirichlet process prior for splitting individuals into groups, with a Bayesian Lasso model in each group to link behavioral features with mental illness. The result is a model for each individual that incorporates information from other similar individuals to augment the limited training data available. RESULTS: The new group-personalized regression model showed a significant improvement over population-level models in predicting mental health severity (P<.001). Analysis of subgroups showed that different groups were characterized by different features derived from raw geolocation data. CONCLUSIONS: This study demonstrates the importance of handling interindividual variability when developing models of mental illness. Population-level models do not capture nuances in how different individuals respond to illness, and the group-personalized model demonstrates a potential way to overcome these limitations when estimating mental state from objective behavioral features.

Desynchronization of diurnal rhythms in bipolar disorder and borderline personality disorder.

It has long been proposed that diurnal rhythms are disturbed in bipolar disorder (BD). Such changes are obvious in episodes of mania or depression. However, detailed study of patients between episodes has been rare and comparison with other psychiatric disorders rarer still. Our hypothesis was that evidence for desynchronization of diurnal rhythms would be evident in BD and that we could test the specificity of any effect by studying borderline personality disorder (BPD). Individuals with BD (n = 36), BPD (n = 22) and healthy volunteers (HC, n = 25) wore a portable heart rate and actigraphy device and used a smart-phone to record self-assessed mood scores 10 times per day for 1 week. Average diurnal patterns of heart rate (HR), activity and sleep were compared within and across groups. Desynchronization in the phase of diurnal rhythms of HR compared with activity were found in BPD (+3 h) and BD (+1 h), but not in HC. A clear diurnal pattern for positive mood was found in all subject groups. The coherence between negative and irritable mood and HR showed a four-cycle per day component in BD and BPD, which was not present in HC. The findings highlight marked de-synchronisation of measured diurnal function in both BD but particularly BPD and suggest an increased association with negative and irritable mood at ultradian frequencies. These findings enhance our understanding of the underlying physiological changes associated with BPD and BD, and suggest objective markers for monitoring and potential treatment targets. Improved mood stabilisation is a translational objective for management of both patient groups.

Clinical Insight Into Latent Variables of Psychiatric Questionnaires for Mood Symptom Self-Assessment.

BACKGROUND: We recently described a new questionnaire to monitor mood called mood zoom (MZ). MZ comprises 6 items assessing mood symptoms on a 7-point Likert scale; we had previously used standard principal component analysis (PCA) to tentatively understand its properties, but the presence of multiple nonzero loadings obstructed the interpretation of its latent variables. OBJECTIVE: The aim of this study was to rigorously investigate the internal properties and latent variables of MZ using an algorithmic approach which may lead to more interpretable results than PCA. Additionally, we explored three other widely used psychiatric questionnaires to investigate latent variable structure similarities with MZ: (1) Altman self-rating mania scale (ASRM), assessing mania; (2) quick inventory of depressive symptomatology (QIDS) self-report, assessing depression; and (3) generalized anxiety disorder (7-item) (GAD-7), assessing anxiety. METHODS: We elicited responses from 131 participants: 48 bipolar disorder (BD), 32 borderline personality disorder (BPD), and 51 healthy controls (HC), collected longitudinally (median [interquartile range, IQR]: 363 [276] days). Participants were requested to complete ASRM, QIDS, and GAD-7 weekly (all 3 questionnaires were completed on the Web) and MZ daily (using a custom-based smartphone app). We applied sparse PCA (SPCA) to determine the latent variables for the four questionnaires, where a small subset of the original items contributes toward each latent variable. RESULTS: We found that MZ had great consistency across the three cohorts studied. Three main principal components were derived using SPCA, which can be tentatively interpreted as (1) anxiety and sadness, (2) positive affect, and (3) irritability. The MZ principal component comprising anxiety and sadness explains most of the variance in BD and BPD, whereas the positive affect of MZ explains most of the variance in HC. The latent variables in ASRM were identical for the patient groups but different for HC; nevertheless, the latent variables shared common items across both the patient group and HC. On the contrary, QIDS had overall very different principal components across groups; sleep was a key element in HC and BD but was absent in BPD. In GAD-7, nervousness was the principal component explaining most of the variance in BD and HC. CONCLUSIONS: This study has important implications for understanding self-reported mood. MZ has a consistent, intuitively interpretable latent variable structure and hence may be a good instrument for generic mood assessment. Irritability appears to be the key distinguishing latent variable between BD and BPD and might be useful for differential diagnosis. Anxiety and sadness are closely interlinked, a finding that might inform treatment effects to jointly address these covarying symptoms. Anxiety and nervousness appear to be amongst the cardinal latent variable symptoms in BD and merit close attention in clinical practice.

SleepAp: an automated obstructive sleep apnoea screening application for smartphones.

Obstructive sleep apnoea (OSA) is a sleep disorder with long-term consequences. Long-term effects include sleep-related issues and cardiovascular diseases. OSA is often diagnosed with an overnight sleep test called a polysomnogram. Monitoring can be costly with long wait times for diagnosis. In this paper, a novel OSA screening framework and prototype phone application are introduced. A database of 856 patients that underwent at-home polygraphy was collected. Features were derived from audio, actigraphy, photoplethysmography (PPG), and demographics, and used as the inputs of a support vector machine (SVM) classifier. The SVM was trained on 735 patients and tested on 121 patients. Classification on the test set had an accuracy of up to 92.2% when classifying subjects as having moderate or severe OSA versus being healthy or a snorer based on the clinicians' diagnoses. The signal processing and machine learning algorithms were ported to Java and integrated into the phone application-SleepAp. SleepAp records the body position, audio, actigraphy and PPG signals, and implements the clinically validated STOP-BANG questionnaire. It derives features from the signals and classifies the user as having OSA or not using the SVM trained on the clinical database. The resulting software could provide a new, easy-to-use, low-cost, and widely available modality for OSA screening.