AI Article Synopsis

  • * Over a period of three weeks, the research involved 30 inpatients whose speech samples and self-reported depression levels were analyzed to extract various speech features, aiming for more personalized predictions.
  • * Results showed that, despite testing over 500 machine learning models with different data-splitting methods, none outperformed random chance in predicting depression severity, indicating the need for further research in this area.

Article Abstract

Background: Mobile 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.

Objective: We 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.

Methods: To 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.

Results: In 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 (R²=0.339, mean absolute error=0.38; both P<.001), indicating that 33.9% of the variance in depression severity could be predicted by the speech features.

Conclusions: Overall, 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.

Download full-text PDF

Source
http://dx.doi.org/10.2196/64578DOI Listing

Publication Analysis

Top Keywords

depression severity
12
momentary depression
8
sleep deprivation
8
deprivation therapy
8
machine learning
8
speech samples
8
extreme gradient
8
gradient boosting
8
models tested
8
chronological split
8

Similar Publications

Importance: Recreational use of drug-soaked paper strips (hereafter, strips) in correctional facilities poses a major public health risk owing to the diverse and potentially severe toxic effects of the substances they contain. Understanding the clinical manifestations and outcomes of exposure to these strips is important for developing effective management and prevention strategies.

Objective: To characterize the clinical manifestations, management, and outcomes of intoxication from strips in a correctional facility population, and to identify the specific substances present in these strips.

View Article and Find Full Text PDF

The COVID-19 pandemic has a profound and lasting impact on the mental health of recovered individuals. To investigate the clinical risk factors associated with long-term post-traumatic stress symptoms (PTSS), anxiety, and depression in COVID-19 survivors, demographic information and medical records were collected during February 19 and March 20, 2020. Assessments of PTSS, anxiety, and depressive symptoms were conducted at two months (April to May 2020, Session 1) and two years (April to May 2022, Session 2) post-discharge.

View Article and Find Full Text PDF

Purpose: Attitudes toward schizophrenia and depression have evolved differently over the last decades, exposing people with schizophrenia to growing stigma. Classic descriptions of schizophrenia symptoms as being particularly unrelatable might offer an explanation for this gap in attitudes that has not yet been tested. We examine to what extent relatability explains the difference in social distance toward people with depression or schizophrenia.

View Article and Find Full Text PDF

Electroconvulsive therapy (ECT) effectively treats severe psychiatric disorders such as depression, mania, catatonia, and schizophrenia. Although its exact mechanism remains unclear, ECT is thought to induce neurochemical and neuroendocrine changes. Positron emission tomography (PET) and single-photon emission computed tomography (SPECT) have provided vital insights into ECT's neurobiological effects.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!