Automatic Depression Assessment using Deep Learning Techniques

dc.contributor.authorLau, Clintonen
dc.contributor.departmentElectrical and Computer Engineeringen
dc.contributor.supervisorChan, Wai-Yip
dc.contributor.supervisorZhu, Xiaodan
dc.date.accessioned2023-04-24T21:38:54Z
dc.date.available2023-04-24T21:38:54Z
dc.degree.grantorQueen's University at Kingstonen
dc.description.abstractEarly recognition and treatment of depression can avert escalation of the mental disorder and alleviate suffering for the patients and their families. To assist mental health care providers with the early recognition and assessment of depression, research to develop a standardized, accessible, and non-invasive technique has garnered considerable attention. This work investigates the use of deep learning techniques for the task of automatic depression assessment. We propose two novel solutions to address current limitations faced by machine learning-powered depression assessment systems. Our first solution utilizes multi-task learning to improve upon calibration of a prediction model. Calibration is an important trait in developing trustworthy models for high-stake applications. Our second solution leverages recent advances in pretrained large language models and parameter-efficient tuning techniques to effectively learn with limited data. Using speech and spoken text extracted from clinical interviews, our resultant models achieve new state-of-the-art performance results on a benchmark depression dataset against strong baselines and previously published methods. Results from this work show promise in applying deep learning models to assist care providers with depression assessment.en
dc.description.degreeM.A.Sc.en
dc.embargo.liftdate2028-04-24T01:40:10Z
dc.embargo.termsOne of the thesis chapters contains a manuscript which is currently under review for a journal publication. We would like it to be restricted until its publication.en
dc.identifier.urihttp://hdl.handle.net/1974/31559
dc.language.isoengen
dc.relation.ispartofseriesCanadian thesesen
dc.rightsQueen's University's Thesis/Dissertation Non-Exclusive License for Deposit to QSpace and Library and Archives Canada*
dc.rightsProQuest PhD and Master's Theses International Dissemination Agreement*
dc.rightsIntellectual Property Guidelines at Queen's University*
dc.rightsCopying and Preserving Your Thesis*
dc.rightsThis publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.*
dc.rightsAttribution-NonCommercial-ShareAlike 3.0 United States*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/3.0/us/*
dc.subjectDepression assessmenten
dc.subjectTransfer learningen
dc.subjectNatural language processingen
dc.subjectDeep learningen
dc.subjectClinical decision supporten
dc.subjectMulti-task learningen
dc.subjectSpeech processingen
dc.subjectModel calibrationen
dc.titleAutomatic Depression Assessment using Deep Learning Techniquesen
dc.typethesisen
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
Lau_Clinton_202304_MASc.pdf
Size:
5.79 MB
Format:
Adobe Portable Document Format
Description:
thesis document
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.67 KB
Format:
Item-specific license agreed upon to submission
Description: