Computer-aided autism diagnosis using visual attention models and eye-tracking: replication and improvement proposal
Nenhuma Miniatura disponível
Citações na Scopus
0
Tipo de produção
article
Data de publicação
2023
Título da Revista
ISSN da Revista
Título do Volume
Editora
BMC
Autores
OLIVEIRA, Jessica S.
MACHADO-LIMA, Ariane
NUNES, Fatima L. S.
Citação
BMC MEDICAL INFORMATICS AND DECISION MAKING, v.23, n.1, article ID 285, 9p, 2023
Resumo
BackgroundAutism Spectrum Disorder (ASD) diagnosis can be aided by approaches based on eye-tracking signals. Recently, the feasibility of building Visual Attention Models (VAMs) from features extracted from visual stimuli and their use for classifying cases and controls has been demonstrated using Neural Networks and Support Vector Machines. The present work has three aims: 1) to evaluate whether the trained classifier from the previous study was generalist enough to classify new samples with a new stimulus; 2) to replicate the previously approach to train a new classifier with a new dataset; 3) to evaluate the performance of classifiers obtained by a new classification algorithm (Random Forest) using the previous and the current datasets.MethodsThe previously approach was replicated with a new stimulus and new sample, 44 from the Typical Development group and 33 from the ASD group. After the replication, Random Forest classifier was tested to substitute Neural Networks algorithm.ResultsThe test with the trained classifier reached an AUC of 0.56, suggesting that the trained classifier requires retraining of the VAMs when changing the stimulus. The replication results reached an AUC of 0.71, indicating the potential of generalization of the approach for aiding ASD diagnosis, as long as the stimulus is similar to the originally proposed. The results achieved with Random Forest were superior to those achieved with the original approach, with an average AUC of 0.95 for the previous dataset and 0.74 for the new dataset.ConclusionIn summary, the results of the replication experiment were satisfactory, which suggests the robustness of the approach and the VAM-based approaches feasibility to aid in ASD diagnosis. The proposed method change improved the classification performance. Some limitations are discussed and additional studies are encouraged to test other conditions and scenarios.
Palavras-chave
Autism spectrum disorder, Eye-tracking, Machine learning, Classifier, Replicability
Referências
- Alharam AK, 2020, IEEE INT SYMP CIRC S, P1, DOI [10.1109/ISCAS45731.2020.9181004, DOI 10.1109/iscas45731.2020.9181004]
- American Psychiatric Association (APA), 2013, Diagnostic and statistical manual of mental disorders: DSM-5, V5th, DOI 10.1176/appi.books.9780890425596
- [Anonymous], 2014, Free SoftwareFoundation I. XPaint
- Bylinskii Z, 2019, IEEE T PATTERN ANAL, V41, P740, DOI 10.1109/TPAMI.2018.2815601
- Chong EJ, 2018, LECT NOTES COMPUT SC, V11209, P397, DOI 10.1007/978-3-030-01228-1_24
- Duan HZ, 2023, SIGNAL PROCESS-IMAGE, V115, DOI 10.1016/j.image.2023.116968
- Fan RE, 2008, J MACH LEARN RES, V9, P1871
- Fang YM, 2021, SIGNAL PROCESS-IMAGE, V93, DOI 10.1016/j.image.2021.116186
- Fang YM, 2019, IEEE INT CONF MULTI, P625, DOI 10.1109/ICMEW.2019.00120
- Guy J, 2016, AUTISM RES, V9, P866, DOI 10.1002/aur.1579
- Holmqvist K, 2023, BEHAV RES METHODS, V55, P364, DOI 10.3758/s13428-021-01762-8
- Hu SJ, 2003, J MECH DESIGN, V125, P262, DOI 10.1115/1.1563633
- Itti L, 1998, IEEE T PATTERN ANAL, V20, P1254, DOI 10.1109/34.730558
- Jesus GRd, 2009, Normatizacao e validacao do teste nao-verbal de inteligencia son-r 2 1/2-7 [a] para o Brasil
- Jia S, 2020, P IEEECVF C COMPUTER, P2667
- Jiang M, 2017, IEEE I CONF COMP VIS, P3287, DOI 10.1109/ICCV.2017.354
- Judd T, 2009, IEEE I CONF COMP VIS, P2106, DOI 10.1109/ICCV.2009.5459462
- Kienzle W, 2006, Adv Neural Inf Process Syst., V19, P1
- Kou J, 2019, AUTISM RES, V12, P1529, DOI 10.1002/aur.2174
- Li Q, 2022, JAMA PEDIATR, V176, P943, DOI 10.1001/jamapediatrics.2022.1846
- Liu GH, 2015, PATTERN RECOGN, V48, P2554, DOI 10.1016/j.patcog.2015.02.005
- Ma XE, 2021, AUTISM RES, V14, P2625, DOI 10.1002/aur.2607
- Mastergeorge AM, 2021, J AUTISM DEV DISORD, V51, P2578, DOI 10.1007/s10803-020-04731-w
- Moore A, 2018, MOL AUTISM, V9, DOI 10.1186/s13229-018-0202-z
- Nosek BA, 2022, ANNU REV PSYCHOL, V73, P719, DOI 10.1146/annurev-psych-020821-114157
- Oliveira JS, 2021, SCI REP-UK, V11, DOI 10.1038/s41598-021-89023-8
- Patil P, 2019, NAT HUM BEHAV, V3, P650, DOI 10.1038/s41562-019-0629-z
- Pepperdine CR, 2018, CAN J SCH PSYCHOL, V33, P157, DOI 10.1177/0829573517733845
- Pierce K, 2016, BIOL PSYCHIAT, V79, P657, DOI 10.1016/j.biopsych.2015.03.032
- Rahman S, 2021, INT C PATT RECOG, P9288, DOI 10.1109/ICPR48806.2021.9412308
- RStudio Team, 2020, RStudio: IDE for R
- Startsev M, 2019, IEEE INT CONF MULTI, P633, DOI 10.1109/ICMEW.2019.00122
- Tatler BW, 2007, J VISION, V7, DOI 10.1167/7.14.4
- The MathWorks Inc, 2019, MATLAB
- The MathWorks Inc, 2022, TreeBagger
- Tobii Technology, 2012, The Tobii I-VT Fixation Filter: Algorithm description
- Tobii Technology, 2022, Tobii
- Tobii Technology, 2022, Tobii Pro Lab
- Tobii Technology, 2021, Tobii Pro Studio-Discontinued
- Tseng PH, 2009, J VISION, V9, DOI 10.1167/9.7.4
- Venker CE, 2022, AUTISM DEV LANG IMP, V7, DOI 10.1177/23969415221085476
- Wang S, 2015, NEURON, V88, P604, DOI 10.1016/j.neuron.2015.09.042
- Wechsler D, 2003, Wechsler intelligence scale for children-Fourth Edition (WISC-IV), P3
- Xiao Yang, 2020, Journal of Physics: Conference Series, V1693, DOI 10.1088/1742-6596/1693/1/012173
- Xu J, 2014, J VISION, V14, DOI 10.1167/14.1.28
- Yaneva V, 2020, IEEE T NEUR SYS REH, V28, P1254, DOI 10.1109/TNSRE.2020.2991675
- Zablotsky B, 2019, PEDIATRICS, V144, DOI 10.1542/peds.2019-0811
- Zachi EC, 2017, FRONT PSYCHOL, V8, DOI 10.3389/fpsyg.2017.01127