Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single number
Abstract 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 reco...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Publishing Group
2024-03-01
|
| Series: | Translational Psychiatry |
| Online Access: | https://doi.org/10.1038/s41398-024-02876-1 |
| _version_ | 1797233253653938176 |
|---|---|
| author | Filippo Corponi Bryan M. Li Gerard Anmella Ariadna Mas Isabella Pacchiarotti Marc Valentí Iria Grande Antoni Benabarre Marina Garriga Eduard Vieta Stephen M. Lawrie Heather C. Whalley Diego Hidalgo-Mazzei Antonio Vergari |
| author_facet | Filippo Corponi Bryan M. Li Gerard Anmella Ariadna Mas Isabella Pacchiarotti Marc Valentí Iria Grande Antoni Benabarre Marina Garriga Eduard Vieta Stephen M. Lawrie Heather C. Whalley Diego Hidalgo-Mazzei Antonio Vergari |
| author_sort | Filippo Corponi |
| collection | DOAJ |
| description | Abstract 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. |
| first_indexed | 2024-04-24T16:13:14Z |
| format | Article |
| id | doaj.art-60db65f5f8d5460bb40878054c16efb2 |
| institution | Directory Open Access Journal |
| issn | 2158-3188 |
| language | English |
| last_indexed | 2024-04-24T16:13:14Z |
| publishDate | 2024-03-01 |
| publisher | Nature Publishing Group |
| record_format | Article |
| series | Translational Psychiatry |
| spelling | doaj.art-60db65f5f8d5460bb40878054c16efb22024-03-31T11:36:18ZengNature Publishing GroupTranslational Psychiatry2158-31882024-03-011411910.1038/s41398-024-02876-1Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single numberFilippo Corponi0Bryan M. Li1Gerard Anmella2Ariadna Mas3Isabella Pacchiarotti4Marc Valentí5Iria Grande6Antoni Benabarre7Marina Garriga8Eduard Vieta9Stephen M. Lawrie10Heather C. Whalley11Diego Hidalgo-Mazzei12Antonio Vergari13School of Informatics, University of EdinburghSchool of Informatics, University of EdinburghBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaDivision of Psychiatry, Centre for Clinical Brain Sciences, University of EdinburghDivision of Psychiatry, Centre for Clinical Brain Sciences, University of EdinburghBipolar and Depressive Disorders Unit, Department of Psychiatry and Psychology, Hospital Clínic de BarcelonaSchool of Informatics, University of EdinburghAbstract 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.https://doi.org/10.1038/s41398-024-02876-1 |
| spellingShingle | Filippo Corponi Bryan M. Li Gerard Anmella Ariadna Mas Isabella Pacchiarotti Marc Valentí Iria Grande Antoni Benabarre Marina Garriga Eduard Vieta Stephen M. Lawrie Heather C. Whalley Diego Hidalgo-Mazzei Antonio Vergari Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single number Translational Psychiatry |
| title | Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single number |
| title_full | Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single number |
| title_fullStr | Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single number |
| title_full_unstemmed | Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single number |
| title_short | Automated mood disorder symptoms monitoring from multivariate time-series sensory data: getting the full picture beyond a single number |
| title_sort | automated mood disorder symptoms monitoring from multivariate time series sensory data getting the full picture beyond a single number |
| url | https://doi.org/10.1038/s41398-024-02876-1 |
| work_keys_str_mv | AT filippocorponi automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT bryanmli automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT gerardanmella automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT ariadnamas automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT isabellapacchiarotti automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT marcvalenti automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT iriagrande automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT antonibenabarre automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT marinagarriga automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT eduardvieta automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT stephenmlawrie automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT heathercwhalley automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT diegohidalgomazzei automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber AT antoniovergari automatedmooddisordersymptomsmonitoringfrommultivariatetimeseriessensorydatagettingthefullpicturebeyondasinglenumber |