Complexity and variability analyses of motor activity distinguish mood states in bipolar disorder

Jakobsen, Petter; Stautland, Andrea; Riegler, Michael; Côté-Allard, Ulysse; Sepasdar, Zahra; Nordgreen, Tine; Tørresen, Jim; Fasmer, Ole Bernt; Oedegaard, Ketil J.

doi:10.1371/journal.pone.0262232

Cited by 15 publications

(22 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, the significant increased use of antipsychotic medications (see Figure ? ?) in the manic group probably inhibited the elevated energy level commonly associated with mania [16]. Therefore, the data employed in this work probably had an artificially altered gap between manic and euthymic state.…”

Section: Limitationsmentioning

confidence: 87%

See 1 more Smart Citation

Long-Short Ensemble Network for Bipolar Manic-Euthymic State Recognition Based on Wrist-worn Sensors

Côté-Allard,

Jakobsen,

Stautland

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Manic episodes of bipolar disorder can lead to uncritical behaviour and delusional psychosis, often with destructive consequences for those affected and their surroundings. Early detection and intervention of a manic episode are crucial to prevent escalation, hospital admission and premature death. However, people with bipolar disorder may not recognize that they are experiencing a manic episode and symptoms such as euphoria and increased productivity can also deter affected individuals from seeking help. This work proposes to perform user-independent, automatic mood-state detection based on actigraphy and electrodermal activity acquired from a wrist-worn device during mania and after recovery (euthymia). This paper proposes a new deep learning-based ensemble method leveraging long (20h) and short (5 minutes) time-intervals to discriminate between the mood-states. When tested on 47 bipolar patients, the proposed classification scheme achieves an average accuracy of 91.59% in euthymic/manic mood-state recognition.BIPOLAR DISORDER is a severe mental disorder characterized by intense periodic mood fluctuations, lifelong disability and a high disease burden which affects more than 1% of the global population [1], [2].

show abstract

Section: Limitationsmentioning

confidence: 87%

“…The Bipolar Complexity Variability (BCV) feature set is derived from [16] and is defined as follows:…”

Section: Bipolar Complexity Variability Features Setmentioning

confidence: 99%

Long-Short Ensemble Network for Bipolar Manic-Euthymic State Recognition Based on Wrist-worn Sensors

Côté-Allard,

Jakobsen,

Stautland

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…One could argue that the higher HR during mania could be due to increased motor activity, and consequentially higher HRV scores. However, we did not discover a significant difference in the amount of movement, as measured by actigraphy in this sample [40].…”

Section: Discussionmentioning

confidence: 99%

Heart rate variability as biomarker for bipolar disorder

Stautland

Jakobsen

Fasmer

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Bipolar disorder (BD) is characterized by alterations in mood, energy levels and the ability to function. Accordingly, it is also associated with dysfunction of the autonomic nervous system (ANS), indexed by heart rate variability (HRV). Literature concerning differences in ANS functioning between mood states is still sparse. The main aim of the study was to investigate within-individual changes in HRV from manic to euthymic states in bipolar disorder (BD). This is the first study to do so using wrist-worn sensors. Seventeen patients with BD were equipped with photoplethysmography (PPG) sensor wristbands and provided 24-hour recordings both during a manic state and a euthymic state. We calculated mean heart rate and the commonly used HRV measures SDNN, RMSSD, HF, LF and Sample Entropy in 5-minute segments during rest at night. We compared HRV by mood state within individuals using paired t-tests and linear regression to control for age and sex. Recordings from 15 BD patients were analyzed. There were statistically significant increases in HRV measures SDNN, RMSSD, LF and Sample Entropy from mania to euthymia. Effect sizes were predominately large. Our findings reveal lower HRV in the manic state compared to the euthymic state. This indicates that HRV collected by wrist-worn PPG sensors is a possible biomarker for bipolar mood states. Movement artifacts were problematic and sampling during rest or in combination with actigraphy is recommended. Our findings can be further implemented to develop a monitoring device for bipolar patients.

show abstract

“…Another study from the same group aimed to distinguish mania and remission in BD patients intra-individually using motor activity wearable digital data. The motor activity of 16 manic BD patients was characterised by altered complexity and variability when compared within-subject to euthymia [22]. Regarding depression versus healthy controls (HCs) identification, one study reached an accuracy of 84.00% on a sample of 23 patients with depression (both unipolar and bipolar) and 32 HCs, using motor activity data and leveraging deep neural networks and SMOTE during training for class imbalance [21].…”

Section: Related Workmentioning

confidence: 99%

Can machine learning with data from wearable devices distinguish disease severity levels and generalise across patients? A pilot study in Mania and Depression

Corponi

Anmella

et al. 2022

Preprint

View full text Add to dashboard Cite

Mood disorders are severe and chronic mental conditions exacting high costs from society. The lack of reliable biomarkers to aid clinicians in tailoring pharmacotherapy based on distinguishable patient-specific traits means that the current prescribing paradigm is largely one of trial and error. Previous studies showed that different biological signatures, such as patterns of heart rate variability or electro-dermal reactivity, are associated with clinically meaningful outcomes. Against this backdrop, the advances in machine learning and the spread of wearable devices capable of providing continuous and ecological monitoring of patients may unlock great opportunities in mental healthcare. We herewith present a pilot study on mania and depression where we moved beyond the simple disease state binary classification but pursued the more informative and clinically meaningful task of differentiating between levels of disease severity. While most previous similar endeavours used recording segments extracted from the same subjects for both training and testing, we explicitly carried out model development and evaluation on segments from different groups of patients, in order to have a fair assessment of the model out-of-sample generalisation. This illustrated how individuals heterogeneity and non-disease-related dimensions of variations (e.g. sex, age, physical fitness) may dominate the signal so that in low sample size regimes a model might learn and overfit subject-specific patterns rather than capturing disease-relevant traits generalisable across disorders. Lastly, we developed a viable baseline for pre-processing raw data from wristband recordings and compared three classical and two deep-learning models to identify levels of disease severity.

show abstract

Complexity and variability analyses of motor activity distinguish mood states in bipolar disorder

Cited by 15 publications

References 54 publications

Long-Short Ensemble Network for Bipolar Manic-Euthymic State Recognition Based on Wrist-worn Sensors

Long-Short Ensemble Network for Bipolar Manic-Euthymic State Recognition Based on Wrist-worn Sensors

Heart rate variability as biomarker for bipolar disorder

Can machine learning with data from wearable devices distinguish disease severity levels and generalise across patients? A pilot study in Mania and Depression

Contact Info

Product

Resources

About