kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Potential för maskininlärning som preliminärt diagnostiseringsverktyg för psykiatriska diagnoser
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2024 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
POTENTIAL OF MACHINE LEARNING IN PRELIMINARY DIAGNOSTICS OF PSYCHIATRIC DISORDERS (English)
Abstract [en]

This study explores the potential of machine learning algorithms as a tool to indicate psychiatric disorders by using audio data from podcasts. The research also investigates if podcasts can be used as valuable data for this purpose, identifies important attributes for classifying psychiatric disorders with machine learning and raises ethical implications of further developing and implementing this technology in healthcare. The implementation of such technology in healthcare could significantly improve patient care by enabling early detection and intervention, potentially reducing the need for costly and prolonged treatments. To investigate these questions a Random Forest Model was trained on both text attributes as well as acustical attributes and the most important attributes for different psychiatric disorders were gathered. The outcome of the research indicates that there is potential for this technology to be developed further and potentially be implemented as a preliminary diagnostics tool. The paper also identifies several key aspects that need to be considered when developing this technology in order for it to comply with the United Nations ethical guidelines for artificial intelligence.

Abstract [sv]

Arbetet undersöker potentialen hos maskininlärningsalgoritmer som verktyg för att indikera psykiatriska diagnoser med hjälp av ljuddata från podcasts. Arbetet undersöker också huruvida podcasts kan användas som värdefull data för detta, identifierar viktiga attribut för att klassificera psykiatriska diagnoser och belyser etiska implikationer av att vidareutveckla och implementera denna teknologi inom sjukvården. Implementeringen av denna teknologi har potential att förbättra patientvården avsevärt genom att möjliggöra tidig upptäckt och intervention, vilket kan minska behovet av kostsamma och långvariga behandlingar. För att undersöka dessa frågor tränades en Random Forest modell på både textattribut samt akustiska attribut och de viktigaste attributen för olika psykiatriska störningar sammanställdes. Resultatet indikerar att det finns potential för att vidareutveckla denna teknologi och potentiellt implementera den som ett verktyg för preliminär diagnostisering. Studien identifierar också flera nyckelaspekter som behöver beaktas vid utveckling av denna teknologi för att den ska gå i linje med FN etiska riktlinjer för artificiell intelligens. 

Place, publisher, year, edition, pages
2024. , p. 12
Series
TRITA-EECS-EX ; 2024:257
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-356560OAI: oai:DiVA.org:kth-356560DiVA, id: diva2:1914150
Supervisors
Examiners
Available from: 2024-12-05 Created: 2024-11-18 Last updated: 2024-12-05Bibliographically approved

Open Access in DiVA

fulltext(460 kB)62 downloads
File information
File name FULLTEXT01.pdfFile size 460 kBChecksum SHA-512
9d1cb29e796b5edb8e6ba411bf9099941a823d9b0d111c727c72846263fbc6c4ef12efbc17bfa4f9597c23142075def2532cefc0268b6426dfb32c00e9bb500d
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 62 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 268 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf