Correction: Replication of machine learning methods to predict treatment outcome with antidepressant...
Correction: Replication of machine learning methods to predict treatment outcome with antidepressant medications in patients with major depressive disorder from STARD and CAN-BIND-1
About this item
Full title
Author / Creator
Publisher
United States: Public Library of Science
Journal title
Language
English
Formats
Publication information
Publisher
United States: Public Library of Science
More information
Scope and Contents
Contents
The models were used to predict antidepressant response by eight weeks in the first treatment level, as defined as a 50% or greater reduction in their last QIDS-SR score in this period. For the STAR*D datasets, replicating the subject selection from Nie et al [15] for TRD prediction as defined by QIDS-C criteria results in 218 subjects, with 571 (26.2%) labelled as TRD. Resulting from replicating a prior study’s cross-validation, predicting treatment-resistant depression according to the Quick Inventory of Depressive Symptomatology, Clinician version (QID-C) scale, using data from Sequenced Treatment Alternatives to Relieve Depression. GBDT: gradient boosting decision tree, AUC: area-under-curve. https://doi.org/10.1371/journal.pone.0315844.t002 thumbnail Download: * PPT PowerPoint slide * PNG larger image * TIFF original image Table 5....
Alternative Titles
Full title
Correction: Replication of machine learning methods to predict treatment outcome with antidepressant medications in patients with major depressive disorder from STARD and CAN-BIND-1
Authors, Artists and Contributors
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_plos_journals_3143786189
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_plos_journals_3143786189
Other Identifiers
ISSN
1932-6203
E-ISSN
1932-6203
DOI
10.1371/journal.pone.0315844
