Spectral Analysis of Light-Adapted Electroretinograms in Neurodevelopmental Disorders: Classification with Machine Learning
Electroretinograms (ERGs) show differences between typically developing populations and those with a diagnosis of autism spectrum disorder (ASD) or attention deficit/hyperactivity disorder (ADHD). In a series of ERGs collected in ASD (<i>n</i> = 77), ADHD (<i>n</i> = 43), ASD...
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/12/1/15 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589088315670528 |
|---|---|
| author | Paul A. Constable Javier O. Pinzon-Arenas Luis Roberto Mercado Diaz Irene O. Lee Fernando Marmolejo-Ramos Lynne Loh Aleksei Zhdanov Mikhail Kulyabin Marek Brabec David H. Skuse Dorothy A. Thompson Hugo Posada-Quintero |
| author_facet | Paul A. Constable Javier O. Pinzon-Arenas Luis Roberto Mercado Diaz Irene O. Lee Fernando Marmolejo-Ramos Lynne Loh Aleksei Zhdanov Mikhail Kulyabin Marek Brabec David H. Skuse Dorothy A. Thompson Hugo Posada-Quintero |
| author_sort | Paul A. Constable |
| collection | DOAJ |
| description | Electroretinograms (ERGs) show differences between typically developing populations and those with a diagnosis of autism spectrum disorder (ASD) or attention deficit/hyperactivity disorder (ADHD). In a series of ERGs collected in ASD (<i>n</i> = 77), ADHD (<i>n</i> = 43), ASD + ADHD (<i>n</i> = 21), and control (<i>n</i> = 137) groups, this analysis explores the use of machine learning and feature selection techniques to improve the classification between these clinically defined groups. Standard time domain and signal analysis features were evaluated in different machine learning models. For ASD classification, a balanced accuracy (BA) of 0.87 was achieved for male participants. For ADHD, a BA of 0.84 was achieved for female participants. When a three-group model (ASD, ADHD, and control) the BA was lower, at 0.70, and fell further to 0.53 when all groups were included (ASD, ADHD, ASD + ADHD, and control). The findings support a role for the ERG in establishing a broad two-group classification of ASD or ADHD, but the model’s performance depends upon sex and is limited when multiple classes are included in machine learning modeling. |
| format | Article |
| id | doaj-art-f92efd5decea40f4a6946f7bb22ef834 |
| institution | Kabale University |
| issn | 2306-5354 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Bioengineering |
| spelling | doaj-art-f92efd5decea40f4a6946f7bb22ef8342025-01-24T13:22:58ZengMDPI AGBioengineering2306-53542024-12-011211510.3390/bioengineering12010015Spectral Analysis of Light-Adapted Electroretinograms in Neurodevelopmental Disorders: Classification with Machine LearningPaul A. Constable0Javier O. Pinzon-Arenas1Luis Roberto Mercado Diaz2Irene O. Lee3Fernando Marmolejo-Ramos4Lynne Loh5Aleksei Zhdanov6Mikhail Kulyabin7Marek Brabec8David H. Skuse9Dorothy A. Thompson10Hugo Posada-Quintero11Caring Futures Institute, College of Nursing and Health Sciences, Flinders University, Adelaide 5000, SA, AustraliaBiomedical Engineering Department, University of Connecticut, Storrs, CT 06269, USABiomedical Engineering Department, University of Connecticut, Storrs, CT 06269, USABehavioural and Brain Sciences Unit, Population Policy and Practice Programme, UCL Great Ormond Street Institute of Child Health, University College London, London WC1N 1EH, UKCollege of Psychology and Education, Flinders University, Adelaide 5000, SA, AustraliaCaring Futures Institute, College of Nursing and Health Sciences, Flinders University, Adelaide 5000, SA, Australia“VisioMed.AI”, Golovinskoe Highway, 8/2A, 125212 Moscow, RussiaPattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, 91058 Erlangen, GermanyInstitute of Computer Science of the Czech Academy of Sciences, Pod Vodarenskou Vezi 2, 182 00 Prague, Czech RepublicBehavioural and Brain Sciences Unit, Population Policy and Practice Programme, UCL Great Ormond Street Institute of Child Health, University College London, London WC1N 1EH, UKThe Tony Kriss Visual Electrophysiology Unit, Clinical and Academic Department of Ophthalmology, Great Ormond Street Hospital for Children NHS Trust, London WC1N 3BH, UKBiomedical Engineering Department, University of Connecticut, Storrs, CT 06269, USAElectroretinograms (ERGs) show differences between typically developing populations and those with a diagnosis of autism spectrum disorder (ASD) or attention deficit/hyperactivity disorder (ADHD). In a series of ERGs collected in ASD (<i>n</i> = 77), ADHD (<i>n</i> = 43), ASD + ADHD (<i>n</i> = 21), and control (<i>n</i> = 137) groups, this analysis explores the use of machine learning and feature selection techniques to improve the classification between these clinically defined groups. Standard time domain and signal analysis features were evaluated in different machine learning models. For ASD classification, a balanced accuracy (BA) of 0.87 was achieved for male participants. For ADHD, a BA of 0.84 was achieved for female participants. When a three-group model (ASD, ADHD, and control) the BA was lower, at 0.70, and fell further to 0.53 when all groups were included (ASD, ADHD, ASD + ADHD, and control). The findings support a role for the ERG in establishing a broad two-group classification of ASD or ADHD, but the model’s performance depends upon sex and is limited when multiple classes are included in machine learning modeling.https://www.mdpi.com/2306-5354/12/1/15biomarkerretinaautismattention deficit hyperactivity disordersexmedication |
| spellingShingle | Paul A. Constable Javier O. Pinzon-Arenas Luis Roberto Mercado Diaz Irene O. Lee Fernando Marmolejo-Ramos Lynne Loh Aleksei Zhdanov Mikhail Kulyabin Marek Brabec David H. Skuse Dorothy A. Thompson Hugo Posada-Quintero Spectral Analysis of Light-Adapted Electroretinograms in Neurodevelopmental Disorders: Classification with Machine Learning Bioengineering biomarker retina autism attention deficit hyperactivity disorder sex medication |
| title | Spectral Analysis of Light-Adapted Electroretinograms in Neurodevelopmental Disorders: Classification with Machine Learning |
| title_full | Spectral Analysis of Light-Adapted Electroretinograms in Neurodevelopmental Disorders: Classification with Machine Learning |
| title_fullStr | Spectral Analysis of Light-Adapted Electroretinograms in Neurodevelopmental Disorders: Classification with Machine Learning |
| title_full_unstemmed | Spectral Analysis of Light-Adapted Electroretinograms in Neurodevelopmental Disorders: Classification with Machine Learning |
| title_short | Spectral Analysis of Light-Adapted Electroretinograms in Neurodevelopmental Disorders: Classification with Machine Learning |
| title_sort | spectral analysis of light adapted electroretinograms in neurodevelopmental disorders classification with machine learning |
| topic | biomarker retina autism attention deficit hyperactivity disorder sex medication |
| url | https://www.mdpi.com/2306-5354/12/1/15 |
| work_keys_str_mv | AT paulaconstable spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT javieropinzonarenas spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT luisrobertomercadodiaz spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT ireneolee spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT fernandomarmolejoramos spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT lynneloh spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT alekseizhdanov spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT mikhailkulyabin spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT marekbrabec spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT davidhskuse spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT dorothyathompson spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning AT hugoposadaquintero spectralanalysisoflightadaptedelectroretinogramsinneurodevelopmentaldisordersclassificationwithmachinelearning |