izpis_h1_title_alt

Identifikacija otrok z avtizmom z uporabo strojnega učenja
ID Gaberšek, Eva (Author), ID Cugmas, Marjan (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (2,33 MB)
MD5: 478524B941253681519A48DDE399438A

Abstract
Avtizem je razvojna motnja, ki se začne izkazovati v obdobju malčkov (pri 14 mesecih), v povprečju pa je diagnosticirana pri 4 letih. Izražena je predvsem kot spremenjeno vedenje na področju socialne interackije, komunikacije in domišljije. Potreba po zgodnji diagnostiki avtizma ter potreba po časovni razbremenitvi diagnostičnega postopka sta glavna razloga za vedno pogostejšo implementacijo metod strojnega učenja na prvi stopnji diagnostičnega postopka, na kateri poteka identifikacija rizičnih enot. Cilja magistrskega dela sta identifikacija otrok z avtizmom in identifikacija najpomembnješih spremenljivk z uporabo napovednih klasifikacijskih modelov. V ta namen je bilo uporabljenih 6 različnih nadzorovanih metod strojnega učenja na 4 različnih starostnih skupinah. Vzorec zajema 48.050 otrok, starejših od 13 mesecev, od tega 60 % dečkov in 40 % deklic, pri čemer je diagnosticiranih s katerokoli diagnozo 20 % otrok. Pojavnost avtizma na vzorcu je 2 %. Uspešni smo bili pri identifikaciji avtizma v skupini otrok, starih od 37 do 48 mesecev, kjer je najuspešnejši model (metode naključnih gozdov) dosegel 72 % točnost, 59 % občutljivost za avtizem in 90 % specifičnost za avtizem. Model pri 10 % otrok brez avtizma napačno identificira znake avtizma. Pravilno identificira 59 % otrok z avtizmom, ostalim 41 % pa pripiše druge diagnoze. Za ocene metrik uspešnosti klasifikacije smo uporabili interno prečno preverjanje. Kot napomembnejši spremenljivki za klasifikacijo smo v skupini triletnikov identificirali spol in spremenljivki, ki merita otrokove slušne zaznave. Modeli so na splošno manj uspešni v mlajših starostnih skupinah, in sicer je v večini najmanj uspešen naiven Bayesov klasifikator. V najmlajših dveh skupinah smo bili najmanj uspešni pri identifikaciji otrok z avtizmom (9 % v najmlajši skupini in 18 % v skupini dvoletnikov). Uspešnost modelov v različnih starostnih skupinah se med seboj razlikuje, pri čemer ni videti jasnega trenda katerekoli metode, kar je lahko posledica različnih vprašalnikov in izraženosti avtizma v različnih starostnih skupinah. Povzamemo lahko, da navkljub veliki uporabnosti implementacije metod strojnega učenja pri iskanju rizičnih enot za avtizem nismo našli enovite rešitve, ki bi bila uspešna v vseh starostnih skupinah. Pri prepoznavanju diagnosticiranih otrok so modeli v vseh starostnih skupinah sicer zelo uspešni, a smo pri vseh kot največjo šibkost prepoznali slabo razločevanje avtizma od ostalih diagnoz.

Language:Slovenian
Keywords:avtizem, strojno učenje, klasifikacija, interno prečno preverjanje, presejalni testi za avtizem
Work type:Master's thesis/paper
Organization:FE - Faculty of Electrical Engineering
Year:2023
PID:20.500.12556/RUL-145048 This link opens in a new window
COBISS.SI-ID:149424899 This link opens in a new window
Publication date in RUL:31.03.2023
Views:1133
Downloads:280
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Identification of autistic children with machine learning
Abstract:
Autism is a developmental disorder with first signs in early childhood (at 14 months) and is typically diagnosed at the age of 4. It is primarily expressed as altered behavior in social interaction, communication and imagination. The need for early diagnosis of autism and the need for time-relieving the diagnostic process are the main reasons for the increasingly frequent implementation of machine learning methods in the first stage of the diagnostic process, which is the identification of children at risk . This study aims to identify children with autism using machine learning classification models and identify the most important variables for classification, using predictive classification models. For this purpose, 6 different supervised machine learning methods were used on 4 different age groups. The sample consisted of 48050 children over 13 months old, with 60 % boys and 40 % girls and 20 % diagnosed. The presence of autism in the sample was 2 %. The study was successful in identifying autism in a group of children aged 37-48 months where the most successful model (Random Forest method) achieved 72 % accuracy and 59 % sensitivity for autism and 90 % specificity for autism. The model misclassified 10 % of non-autistic cases and correctly identified 59 % of autistic cases and classified the remaining 41 % as having another diagnosis. Nested cross-validation was used for classification performance metrics estimation. In the group of three-year-olds, we identified gender and variables that measure children's auditory perception as the most important variables for classification. Models were generally less accurate in younger age groups, with the Naive Bayes classifier being the least accurate. The youngest two age groups however showed very low sensitivity for autism (9 % in the youngest group and 18 % in the two-year-old group). The evaluation of models in different age groups showed varying success and no clear trend for which method is the most successful across all groups, which could be a consequence of different screeners and expression of autism in different age groups. We can conclude that despite the growing use of machine learning methods in autism diagnosis, we were not successful with finding a definitive solution across different age groups. We can identify the main challenge: the models struggle to distinguish autism from other diagnoses and they successfully identify diagnosed cases but misclassify 40 % of autistic cases as having another diagnosis.

Keywords:autism, machine learning, classification, nested cross-validation, autism screening

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back