Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Nio hinder för klinisk diagnostik av autismspektrumtillstånd med artificiell intelligens: en systematisk litteraturstudie

Kanakura, Max LU (2022) PSPR14 20212
Department of Psychology
Abstract
This systematic literature review has tried to answer how eye-tracking and adjacent technologies in combination with machine learning (n = 13 studies) or deep learning (n = 6 studies) respectively can be used for the purpose of autism spectrum diagnostics (ASD). Included articles have been published in peer-reviewed journals. Most studies have sample sizes below 100 participants (n = 14 studies), 101-160 participants (n = 3 studies), 161-1000 participants (n = 2 studies). Studies include toddlers (16 months) to adults. Machine learning tends to be less accurate (59-93%) than deep learning (81-95%) in sorting individuals with or without ASD. The highest accuracy, precision (positive predicative value) and specificity (reliability) in the... (More)
This systematic literature review has tried to answer how eye-tracking and adjacent technologies in combination with machine learning (n = 13 studies) or deep learning (n = 6 studies) respectively can be used for the purpose of autism spectrum diagnostics (ASD). Included articles have been published in peer-reviewed journals. Most studies have sample sizes below 100 participants (n = 14 studies), 101-160 participants (n = 3 studies), 161-1000 participants (n = 2 studies). Studies include toddlers (16 months) to adults. Machine learning tends to be less accurate (59-93%) than deep learning (81-95%) in sorting individuals with or without ASD. The highest accuracy, precision (positive predicative value) and specificity (reliability) in the included studies were achieved by comparing eye-scanning and EEG data (95%, 95% and 95% respectively). A list of 9 obstacles that scientific studies would need to address before these AI technologies could be implemented in clinical practice as a tool for identifying ASD are reviewed. (Less)
Abstract (Swedish)
Det här är en systematisk litteraturöversikt som har prövat att besvara hur ögonskanning och angränsande teknologier i kombination med maskin- (n = 13 studier) och djupinlärning (n = 6 studier), var för sig, kan användas för att diagnostisera autismspektrumstörning (AST). Inkluderade artiklar har varit publicerade i vetenskapligt granskade journaler. De flesta studierna har haft under 100 deltagare (n = 14 studier), 101-160 deltagare (n = 3 studier), 161-1000 deltagare (n = 2 studier). Studierna har inkluderat småbarn (16 månader) till vuxna personer. Maskininlärning tenderar att vara mindre träffsäkert (59-93%) än djupinlärning (81-95%) i att särskilja individer med och utan AST. Den högsta träffsäkerheten, precisionen (positivt... (More)
Det här är en systematisk litteraturöversikt som har prövat att besvara hur ögonskanning och angränsande teknologier i kombination med maskin- (n = 13 studier) och djupinlärning (n = 6 studier), var för sig, kan användas för att diagnostisera autismspektrumstörning (AST). Inkluderade artiklar har varit publicerade i vetenskapligt granskade journaler. De flesta studierna har haft under 100 deltagare (n = 14 studier), 101-160 deltagare (n = 3 studier), 161-1000 deltagare (n = 2 studier). Studierna har inkluderat småbarn (16 månader) till vuxna personer. Maskininlärning tenderar att vara mindre träffsäkert (59-93%) än djupinlärning (81-95%) i att särskilja individer med och utan AST. Den högsta träffsäkerheten, precisionen (positivt prediktionsvärde) och specificiteten (tillförlitlighet) i en inkluderad studie uppnåddes med ögonskanning och EEG data (95%, 95% och 95%). En lista med 9 hinder som vetenskapliga studier skulle behöva övervinna innan AI teknologi kan införas inom klinisk praktik med syfte att diagnostisera patienter med AST har sammanställts. (Less)
Please use this url to cite or link to this publication:
author
Kanakura, Max LU
supervisor
organization
course
PSPR14 20212
year
type
H3 - Professional qualifications (4 Years - )
subject
keywords
Maskininlärning, djupinlärning, autismspektrumstörning, artificiell intelligens, diagnostik.
language
Swedish
id
9076523
date added to LUP
2022-03-16 10:35:17
date last changed
2022-03-16 10:35:17
@misc{9076523,
  abstract     = {{This systematic literature review has tried to answer how eye-tracking and adjacent technologies in combination with machine learning (n = 13 studies) or deep learning (n = 6 studies) respectively can be used for the purpose of autism spectrum diagnostics (ASD). Included articles have been published in peer-reviewed journals. Most studies have sample sizes below 100 participants (n = 14 studies), 101-160 participants (n = 3 studies), 161-1000 participants (n = 2 studies). Studies include toddlers (16 months) to adults. Machine learning tends to be less accurate (59-93%) than deep learning (81-95%) in sorting individuals with or without ASD. The highest accuracy, precision (positive predicative value) and specificity (reliability) in the included studies were achieved by comparing eye-scanning and EEG data (95%, 95% and 95% respectively). A list of 9 obstacles that scientific studies would need to address before these AI technologies could be implemented in clinical practice as a tool for identifying ASD are reviewed.}},
  author       = {{Kanakura, Max}},
  language     = {{swe}},
  note         = {{Student Paper}},
  title        = {{Nio hinder för klinisk diagnostik av autismspektrumtillstånd med artificiell intelligens: en systematisk litteraturstudie}},
  year         = {{2022}},
}