Predicting individual clinical trajectories of depression with generative embedding
Patients with major depressive disorder (MDD) show heterogeneous treatment response and highly variable clinical trajectories: while some patients experience swift recovery, others show relapsing-remitting or chronic courses. Predicting individual clinical trajectories at an early stage is a key cha...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2020-01-01
|
| Series: | NeuroImage: Clinical |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2213158220300504 |
| _version_ | 1811204861518151680 |
|---|---|
| author | Stefan Frässle Andre F. Marquand Lianne Schmaal Richard Dinga Dick J. Veltman Nic J.A. van der Wee Marie-José van Tol Dario Schöbi Brenda W.J.H. Penninx Klaas E. Stephan |
| author_facet | Stefan Frässle Andre F. Marquand Lianne Schmaal Richard Dinga Dick J. Veltman Nic J.A. van der Wee Marie-José van Tol Dario Schöbi Brenda W.J.H. Penninx Klaas E. Stephan |
| author_sort | Stefan Frässle |
| collection | DOAJ |
| description | Patients with major depressive disorder (MDD) show heterogeneous treatment response and highly variable clinical trajectories: while some patients experience swift recovery, others show relapsing-remitting or chronic courses. Predicting individual clinical trajectories at an early stage is a key challenge for psychiatry and might facilitate individually tailored interventions. So far, however, reliable predictors at the single-patient level are absent. Here, we evaluated the utility of a machine learning strategy – generative embedding (GE) – which combines interpretable generative models with discriminative classifiers. Specifically, we used functional magnetic resonance imaging (fMRI) data of emotional face perception in 85 MDD patients from the NEtherlands Study of Depression and Anxiety (NESDA) who had been followed up over two years and classified into three subgroups with distinct clinical trajectories. Combining a generative model of effective (directed) connectivity with support vector machines (SVMs), we could predict whether a given patient would experience chronic depression vs. fast remission with a balanced accuracy of 79%. Gradual improvement vs. fast remission could still be predicted above-chance, but less convincingly, with a balanced accuracy of 61%. Generative embedding outperformed classification based on conventional (descriptive) features, such as functional connectivity or local activation estimates, which were obtained from the same data and did not allow for above-chance classification accuracy. Furthermore, predictive performance of GE could be assigned to a specific network property: the trial-by-trial modulation of connections by emotional content. Given the limited sample size of our study, the present results are preliminary but may serve as proof-of-concept, illustrating the potential of GE for obtaining clinical predictions that are interpretable in terms of network mechanisms. Our findings suggest that abnormal dynamic changes of connections involved in emotional face processing might be associated with higher risk of developing a less favorable clinical course. |
| first_indexed | 2024-04-12T03:20:46Z |
| format | Article |
| id | doaj.art-700a5c474eda4dac9589211d864f45cc |
| institution | Directory Open Access Journal |
| issn | 2213-1582 |
| language | English |
| last_indexed | 2024-04-12T03:20:46Z |
| publishDate | 2020-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | NeuroImage: Clinical |
| spelling | doaj.art-700a5c474eda4dac9589211d864f45cc2022-12-22T03:49:54ZengElsevierNeuroImage: Clinical2213-15822020-01-0126Predicting individual clinical trajectories of depression with generative embeddingStefan Frässle0Andre F. Marquand1Lianne Schmaal2Richard Dinga3Dick J. Veltman4Nic J.A. van der Wee5Marie-José van Tol6Dario Schöbi7Brenda W.J.H. Penninx8Klaas E. Stephan9Translational Neuromodeling Unit (TNU), Institute for Biomedical Engineering, University of Zurich & ETH Zurich, Zurich 8032, Switzerland; Corresponding author.Donders Institute for Brain, Cognition and Behaviour, Radbound University, Nijmegen, The Netherlands; Department of Neuroimaging, Institute of Psychiatry, King's College London, London, United KingdomOrygen, The National Centre of Excellence in Youth Mental Health, Parkville, Australia; Centre for Youth Mental Health, University of Melbourne, Melbourne, AustraliaDepartment of Psychiatry and Neuroscience Campus Amsterdam, VU University Medical Center Amsterdam, Amsterdam, The NetherlandsDepartment of Psychiatry and Neuroscience Campus Amsterdam, VU University Medical Center Amsterdam, Amsterdam, The NetherlandsDepartment of Psychiatry, Leiden University Medical Center, Leiden University, Leiden, The NetherlandsCognitive Neuroscience Center, University Medical Center Groningen, University of Groningen, Groningen, The NetherlandsTranslational Neuromodeling Unit (TNU), Institute for Biomedical Engineering, University of Zurich & ETH Zurich, Zurich 8032, SwitzerlandDepartment of Psychiatry and Neuroscience Campus Amsterdam, VU University Medical Center Amsterdam, Amsterdam, The Netherlands; Department of Psychiatry, Amsterdam UMC, VU University, and Amsterdam Neuroscience, Amsterdam, The NetherlandsTranslational Neuromodeling Unit (TNU), Institute for Biomedical Engineering, University of Zurich & ETH Zurich, Zurich 8032, Switzerland; Wellcome Centre for Human Neuroimaging, University College London, London WC1N 3BG, United Kingdom; Max Planck Institute for Metabolism Research, Cologne, GermanyPatients with major depressive disorder (MDD) show heterogeneous treatment response and highly variable clinical trajectories: while some patients experience swift recovery, others show relapsing-remitting or chronic courses. Predicting individual clinical trajectories at an early stage is a key challenge for psychiatry and might facilitate individually tailored interventions. So far, however, reliable predictors at the single-patient level are absent. Here, we evaluated the utility of a machine learning strategy – generative embedding (GE) – which combines interpretable generative models with discriminative classifiers. Specifically, we used functional magnetic resonance imaging (fMRI) data of emotional face perception in 85 MDD patients from the NEtherlands Study of Depression and Anxiety (NESDA) who had been followed up over two years and classified into three subgroups with distinct clinical trajectories. Combining a generative model of effective (directed) connectivity with support vector machines (SVMs), we could predict whether a given patient would experience chronic depression vs. fast remission with a balanced accuracy of 79%. Gradual improvement vs. fast remission could still be predicted above-chance, but less convincingly, with a balanced accuracy of 61%. Generative embedding outperformed classification based on conventional (descriptive) features, such as functional connectivity or local activation estimates, which were obtained from the same data and did not allow for above-chance classification accuracy. Furthermore, predictive performance of GE could be assigned to a specific network property: the trial-by-trial modulation of connections by emotional content. Given the limited sample size of our study, the present results are preliminary but may serve as proof-of-concept, illustrating the potential of GE for obtaining clinical predictions that are interpretable in terms of network mechanisms. Our findings suggest that abnormal dynamic changes of connections involved in emotional face processing might be associated with higher risk of developing a less favorable clinical course.http://www.sciencedirect.com/science/article/pii/S2213158220300504 |
| spellingShingle | Stefan Frässle Andre F. Marquand Lianne Schmaal Richard Dinga Dick J. Veltman Nic J.A. van der Wee Marie-José van Tol Dario Schöbi Brenda W.J.H. Penninx Klaas E. Stephan Predicting individual clinical trajectories of depression with generative embedding NeuroImage: Clinical |
| title | Predicting individual clinical trajectories of depression with generative embedding |
| title_full | Predicting individual clinical trajectories of depression with generative embedding |
| title_fullStr | Predicting individual clinical trajectories of depression with generative embedding |
| title_full_unstemmed | Predicting individual clinical trajectories of depression with generative embedding |
| title_short | Predicting individual clinical trajectories of depression with generative embedding |
| title_sort | predicting individual clinical trajectories of depression with generative embedding |
| url | http://www.sciencedirect.com/science/article/pii/S2213158220300504 |
| work_keys_str_mv | AT stefanfrassle predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT andrefmarquand predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT lianneschmaal predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT richarddinga predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT dickjveltman predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT nicjavanderwee predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT mariejosevantol predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT darioschobi predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT brendawjhpenninx predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding AT klaasestephan predictingindividualclinicaltrajectoriesofdepressionwithgenerativeembedding |