Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data
Background: Individuals with major depressive disorder (MDD) vary in their response to antidepressants. However, identifying objective biomarkers, prior to or early in the course of treatment that can predict antidepressant efficacy, remains a challenge.Methods: Individuals with MDD participated in...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2019-01-01
|
| Series: | Frontiers in Psychiatry |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fpsyt.2018.00768/full |
| _version_ | 1818136073527623680 |
|---|---|
| author | Natalia Jaworska Natalia Jaworska Natalia Jaworska Sara de la Salle Mohamed-Hamza Ibrahim Pierre Blier Pierre Blier Pierre Blier Verner Knott Verner Knott Verner Knott |
| author_facet | Natalia Jaworska Natalia Jaworska Natalia Jaworska Sara de la Salle Mohamed-Hamza Ibrahim Pierre Blier Pierre Blier Pierre Blier Verner Knott Verner Knott Verner Knott |
| author_sort | Natalia Jaworska |
| collection | DOAJ |
| description | Background: Individuals with major depressive disorder (MDD) vary in their response to antidepressants. However, identifying objective biomarkers, prior to or early in the course of treatment that can predict antidepressant efficacy, remains a challenge.Methods: Individuals with MDD participated in a 12-week antidepressant pharmacotherapy trial. Electroencephalographic (EEG) data was collected before and 1 week post-treatment initiation in 51 patients. Response status at week 12 was established with the Montgomery-Asberg Depression Scale (MADRS), with a ≥50% decrease characterizing responders (N = 27/24 responders/non-responders). We used a machine learning (ML)-approach for predicting response status. We focused on Random Forests, though other ML methods were compared. First, we used a tree-based estimator to select a relatively small number of significant features from: (a) demographic/clinical data (age, sex, individual item/total MADRS scores at baseline, week 1, change scores); (b) scalp-level EEG power; (c) source-localized current density (via exact low-resolution electromagnetic tomography [eLORETA] software). Second, we applied kernel principal component analysis to reduce and map important features. Third, a set of ML models were constructed to classify response outcome based on mapped features. For each dataset, predictive features were extracted, followed by a model of all predictive features, and finally by a model of the most predictive features.Results: Fifty eLORETA features were predictive of response (across bands, both time-points); alpha1/theta eLORETA features showed the highest predictive value. Eighty-eight scalp EEG features were predictive of response (across bands, both time-points), with theta/alpha2 being most predictive. Clinical/demographic data consisted of 31 features, with the most important being week 1 “concentration difficulty” scores. When all features were included into one model, its predictive utility was high (88% accuracy). When the most important features were extracted in the final model, 12 predictive features emerged (78% accuracy), including baseline scalp-EEG frontopolar theta, parietal alpha2 and frontopolar alpha1.Conclusions: These findings suggest that ML models of pre- and early treatment-emergent EEG profiles and clinical features can serve as tools for predicting antidepressant response. While this must be replicated using large independent samples, it lays the groundwork for research on personalized, “biomarker”-based treatment approaches. |
| first_indexed | 2024-12-11T09:34:37Z |
| format | Article |
| id | doaj.art-e00d7b0e411e4bb185f4fa54b25bba76 |
| institution | Directory Open Access Journal |
| issn | 1664-0640 |
| language | English |
| last_indexed | 2024-12-11T09:34:37Z |
| publishDate | 2019-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Psychiatry |
| spelling | doaj.art-e00d7b0e411e4bb185f4fa54b25bba762022-12-22T01:12:56ZengFrontiers Media S.A.Frontiers in Psychiatry1664-06402019-01-01910.3389/fpsyt.2018.00768431031Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical DataNatalia Jaworska0Natalia Jaworska1Natalia Jaworska2Sara de la Salle3Mohamed-Hamza Ibrahim4Pierre Blier5Pierre Blier6Pierre Blier7Verner Knott8Verner Knott9Verner Knott10Institute of Mental Health Research, University of Ottawa, Ottawa, ON, CanadaCellular & Molecular Medicine, Faculty of Medicine, University of Ottawa, Ottawa, ON, CanadaBrain and Mind Research Institute, University of Ottawa, Ottawa, ON, CanadaInstitute of Mental Health Research, University of Ottawa, Ottawa, ON, CanadaDepartment of Mathematics, Faculty of Science, Zagazig University, Zagazig, EgyptInstitute of Mental Health Research, University of Ottawa, Ottawa, ON, CanadaCellular & Molecular Medicine, Faculty of Medicine, University of Ottawa, Ottawa, ON, CanadaBrain and Mind Research Institute, University of Ottawa, Ottawa, ON, CanadaInstitute of Mental Health Research, University of Ottawa, Ottawa, ON, CanadaCellular & Molecular Medicine, Faculty of Medicine, University of Ottawa, Ottawa, ON, CanadaBrain and Mind Research Institute, University of Ottawa, Ottawa, ON, CanadaBackground: Individuals with major depressive disorder (MDD) vary in their response to antidepressants. However, identifying objective biomarkers, prior to or early in the course of treatment that can predict antidepressant efficacy, remains a challenge.Methods: Individuals with MDD participated in a 12-week antidepressant pharmacotherapy trial. Electroencephalographic (EEG) data was collected before and 1 week post-treatment initiation in 51 patients. Response status at week 12 was established with the Montgomery-Asberg Depression Scale (MADRS), with a ≥50% decrease characterizing responders (N = 27/24 responders/non-responders). We used a machine learning (ML)-approach for predicting response status. We focused on Random Forests, though other ML methods were compared. First, we used a tree-based estimator to select a relatively small number of significant features from: (a) demographic/clinical data (age, sex, individual item/total MADRS scores at baseline, week 1, change scores); (b) scalp-level EEG power; (c) source-localized current density (via exact low-resolution electromagnetic tomography [eLORETA] software). Second, we applied kernel principal component analysis to reduce and map important features. Third, a set of ML models were constructed to classify response outcome based on mapped features. For each dataset, predictive features were extracted, followed by a model of all predictive features, and finally by a model of the most predictive features.Results: Fifty eLORETA features were predictive of response (across bands, both time-points); alpha1/theta eLORETA features showed the highest predictive value. Eighty-eight scalp EEG features were predictive of response (across bands, both time-points), with theta/alpha2 being most predictive. Clinical/demographic data consisted of 31 features, with the most important being week 1 “concentration difficulty” scores. When all features were included into one model, its predictive utility was high (88% accuracy). When the most important features were extracted in the final model, 12 predictive features emerged (78% accuracy), including baseline scalp-EEG frontopolar theta, parietal alpha2 and frontopolar alpha1.Conclusions: These findings suggest that ML models of pre- and early treatment-emergent EEG profiles and clinical features can serve as tools for predicting antidepressant response. While this must be replicated using large independent samples, it lays the groundwork for research on personalized, “biomarker”-based treatment approaches.https://www.frontiersin.org/article/10.3389/fpsyt.2018.00768/fullmajor depressive disorder (MDD)antidepressantsbiomarkerquantitative EEGmachine learning (ML)classification and regression trees |
| spellingShingle | Natalia Jaworska Natalia Jaworska Natalia Jaworska Sara de la Salle Mohamed-Hamza Ibrahim Pierre Blier Pierre Blier Pierre Blier Verner Knott Verner Knott Verner Knott Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data Frontiers in Psychiatry major depressive disorder (MDD) antidepressants biomarker quantitative EEG machine learning (ML) classification and regression trees |
| title | Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data |
| title_full | Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data |
| title_fullStr | Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data |
| title_full_unstemmed | Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data |
| title_short | Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data |
| title_sort | leveraging machine learning approaches for predicting antidepressant treatment response using electroencephalography eeg and clinical data |
| topic | major depressive disorder (MDD) antidepressants biomarker quantitative EEG machine learning (ML) classification and regression trees |
| url | https://www.frontiersin.org/article/10.3389/fpsyt.2018.00768/full |
| work_keys_str_mv | AT nataliajaworska leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT nataliajaworska leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT nataliajaworska leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT saradelasalle leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT mohamedhamzaibrahim leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT pierreblier leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT pierreblier leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT pierreblier leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT vernerknott leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT vernerknott leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata AT vernerknott leveragingmachinelearningapproachesforpredictingantidepressanttreatmentresponseusingelectroencephalographyeegandclinicaldata |