Momentary Depression Severity Prediction in Patients With Acute Depression Who Undergo Sleep Deprivation Therapy: Speech-Based Machine Learning Approach
Abstract BackgroundMobile devices for remote monitoring are inevitable tools to support treatment and patient care, especially in recurrent diseases such as major depressive disorder. The aim of this study was to learn if machine learning (ML) models based on longitudinal spee...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
JMIR Publications
2024-12-01
|
| Series: | JMIR Mental Health |
| Online Access: | https://mental.jmir.org/2024/1/e64578 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846100170021797888 |
|---|---|
| author | Lisa-Marie Hartnagel Daniel Emden Jerome C Foo Fabian Streit Stephanie H Witt Josef Frank Matthias F Limberger Sara E Schmitz Maria Gilles Marcella Rietschel Tim Hahn Ulrich W Ebner-Priemer Lea Sirignano |
| author_facet | Lisa-Marie Hartnagel Daniel Emden Jerome C Foo Fabian Streit Stephanie H Witt Josef Frank Matthias F Limberger Sara E Schmitz Maria Gilles Marcella Rietschel Tim Hahn Ulrich W Ebner-Priemer Lea Sirignano |
| author_sort | Lisa-Marie Hartnagel |
| collection | DOAJ |
| description |
Abstract
BackgroundMobile devices for remote monitoring are inevitable tools to support treatment and patient care, especially in recurrent diseases such as major depressive disorder. The aim of this study was to learn if machine learning (ML) models based on longitudinal speech data are helpful in predicting momentary depression severity. Data analyses were based on a dataset including 30 inpatients during an acute depressive episode receiving sleep deprivation therapy in stationary care, an intervention inducing a rapid change in depressive symptoms in a relatively short period of time. Using an ambulatory assessment approach, we captured speech samples and assessed concomitant depression severity via self-report questionnaire over the course of 3 weeks (before, during, and after therapy). We extracted 89 speech features from the speech samples using the Extended Geneva Minimalistic Acoustic Parameter Set from the Open-Source Speech and Music Interpretation by Large-Space Extraction (audEERING) toolkit and the additional parameter speech rate.
ObjectiveWe aimed to understand if a multiparameter ML approach would significantly improve the prediction compared to previous statistical analyses, and, in addition, which mechanism for splitting training and test data was most successful, especially focusing on the idea of personalized prediction.
MethodsTo do so, we trained and evaluated a set of >500 ML pipelines including random forest, linear regression, support vector regression, and Extreme Gradient Boosting regression models and tested them on 5 different train-test split scenarios: a group 5-fold nested cross-validation at the subject level, a leave-one-subject-out approach, a chronological split, an odd-even split, and a random split.
ResultsIn the 5-fold cross-validation, the leave-one-subject-out, and the chronological split approaches, none of the models were statistically different from random chance. The other two approaches produced significant results for at least one of the models tested, with similar performance. In total, the superior model was an Extreme Gradient Boosting in the odd-even split approach (RP
ConclusionsOverall, our analyses highlight that ML fails to predict depression scores of unseen patients, but prediction performance increased strongly compared to our previous analyses with multilevel models. We conclude that future personalized ML models might improve prediction performance even more, leading to better patient management and care. |
| format | Article |
| id | doaj-art-62acb32aa59a4460abccdf370dae423f |
| institution | Kabale University |
| issn | 2368-7959 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | JMIR Publications |
| record_format | Article |
| series | JMIR Mental Health |
| spelling | doaj-art-62acb32aa59a4460abccdf370dae423f2024-12-30T15:01:47ZengJMIR PublicationsJMIR Mental Health2368-79592024-12-0111e64578e6457810.2196/64578Momentary Depression Severity Prediction in Patients With Acute Depression Who Undergo Sleep Deprivation Therapy: Speech-Based Machine Learning ApproachLisa-Marie Hartnagelhttp://orcid.org/0000-0001-8449-4017Daniel Emdenhttp://orcid.org/0000-0001-7459-6634Jerome C Foohttp://orcid.org/0000-0003-1067-5725Fabian Streithttp://orcid.org/0000-0003-1080-4339Stephanie H Witthttp://orcid.org/0000-0002-1571-1468Josef Frankhttp://orcid.org/0000-0003-4867-9465Matthias F Limbergerhttp://orcid.org/0000-0002-6332-397XSara E Schmitzhttp://orcid.org/0000-0003-4496-7398Maria Gilleshttp://orcid.org/0000-0002-9604-4459Marcella Rietschelhttp://orcid.org/0000-0002-5236-6149Tim Hahnhttp://orcid.org/0000-0001-6541-3795Ulrich W Ebner-Priemerhttp://orcid.org/0000-0002-2769-5944Lea Sirignanohttp://orcid.org/0000-0002-7989-5833 Abstract BackgroundMobile devices for remote monitoring are inevitable tools to support treatment and patient care, especially in recurrent diseases such as major depressive disorder. The aim of this study was to learn if machine learning (ML) models based on longitudinal speech data are helpful in predicting momentary depression severity. Data analyses were based on a dataset including 30 inpatients during an acute depressive episode receiving sleep deprivation therapy in stationary care, an intervention inducing a rapid change in depressive symptoms in a relatively short period of time. Using an ambulatory assessment approach, we captured speech samples and assessed concomitant depression severity via self-report questionnaire over the course of 3 weeks (before, during, and after therapy). We extracted 89 speech features from the speech samples using the Extended Geneva Minimalistic Acoustic Parameter Set from the Open-Source Speech and Music Interpretation by Large-Space Extraction (audEERING) toolkit and the additional parameter speech rate. ObjectiveWe aimed to understand if a multiparameter ML approach would significantly improve the prediction compared to previous statistical analyses, and, in addition, which mechanism for splitting training and test data was most successful, especially focusing on the idea of personalized prediction. MethodsTo do so, we trained and evaluated a set of >500 ML pipelines including random forest, linear regression, support vector regression, and Extreme Gradient Boosting regression models and tested them on 5 different train-test split scenarios: a group 5-fold nested cross-validation at the subject level, a leave-one-subject-out approach, a chronological split, an odd-even split, and a random split. ResultsIn the 5-fold cross-validation, the leave-one-subject-out, and the chronological split approaches, none of the models were statistically different from random chance. The other two approaches produced significant results for at least one of the models tested, with similar performance. In total, the superior model was an Extreme Gradient Boosting in the odd-even split approach (RP ConclusionsOverall, our analyses highlight that ML fails to predict depression scores of unseen patients, but prediction performance increased strongly compared to our previous analyses with multilevel models. We conclude that future personalized ML models might improve prediction performance even more, leading to better patient management and care.https://mental.jmir.org/2024/1/e64578 |
| spellingShingle | Lisa-Marie Hartnagel Daniel Emden Jerome C Foo Fabian Streit Stephanie H Witt Josef Frank Matthias F Limberger Sara E Schmitz Maria Gilles Marcella Rietschel Tim Hahn Ulrich W Ebner-Priemer Lea Sirignano Momentary Depression Severity Prediction in Patients With Acute Depression Who Undergo Sleep Deprivation Therapy: Speech-Based Machine Learning Approach JMIR Mental Health |
| title | Momentary Depression Severity Prediction in Patients With Acute Depression Who Undergo Sleep Deprivation Therapy: Speech-Based Machine Learning Approach |
| title_full | Momentary Depression Severity Prediction in Patients With Acute Depression Who Undergo Sleep Deprivation Therapy: Speech-Based Machine Learning Approach |
| title_fullStr | Momentary Depression Severity Prediction in Patients With Acute Depression Who Undergo Sleep Deprivation Therapy: Speech-Based Machine Learning Approach |
| title_full_unstemmed | Momentary Depression Severity Prediction in Patients With Acute Depression Who Undergo Sleep Deprivation Therapy: Speech-Based Machine Learning Approach |
| title_short | Momentary Depression Severity Prediction in Patients With Acute Depression Who Undergo Sleep Deprivation Therapy: Speech-Based Machine Learning Approach |
| title_sort | momentary depression severity prediction in patients with acute depression who undergo sleep deprivation therapy speech based machine learning approach |
| url | https://mental.jmir.org/2024/1/e64578 |
| work_keys_str_mv | AT lisamariehartnagel momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT danielemden momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT jeromecfoo momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT fabianstreit momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT stephaniehwitt momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT joseffrank momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT matthiasflimberger momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT saraeschmitz momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT mariagilles momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT marcellarietschel momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT timhahn momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT ulrichwebnerpriemer momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach AT leasirignano momentarydepressionseveritypredictioninpatientswithacutedepressionwhoundergosleepdeprivationtherapyspeechbasedmachinelearningapproach |