SYNTHETIC ELECTRORETINOGRAM SIGNALS FOR ENHANCING CLASSIFICATION OF AUTISM SPECTRUM DISORDER

Citation Author(s):
Mikhail
Kulyabin
Pattern Recognition Lab, University of Erlangen-Nuremberg, Germany
Paul A.
Constable
College of Nursing and Health Sciences, Caring Futures Institute, Flinders University, Australia
Aleksei
Zhdanov
Siemens Healthineers, Erlangen, Germany
Irene O.
Lee
Behavioural and Brain Sciences Unit, Population Policy and Practice Programme, UCL Great Ormond Street Institute of Child Health, University College London, London, UK
David H.
Skuse
Behavioural and Brain Sciences Unit, Population Policy and Practice Programme, UCL Great Ormond Street Institute of Child Health, University College London, London, UK
Dorothy A.
Thompson
The Tony Kriss Visual Electrophysiology Unit, Clinical and Academic Department of Ophthalmology, Great Ormond Street Hospital for Children NHS Trust, London, UK
Andreas
Maier
Pattern Recognition Lab, University of Erlangen-Nuremberg, Germany
Submitted by:
Mikhail Kulyabin
Last updated:
Sat, 03/16/2024 - 07:06
DOI:
10.21227/npv7-8063
Data Format:
License:
139 Views
Categories:
Keywords:
0
0 ratings - Please login to submit your rating.

Abstract 

The electroretinogram (ERG) is a clinical test that records the retina's electrical response to light. The ERG is a promising way to study different neurodevelopmental and neurodegenerative disorders, including Autism Spectrum Disorder (ASD) - a neurodevelopmental condition that impacts language, communication, and social interactions. However, privacy issues and a lack of data complicate Artificial Intelligence applications in this domain. Synthetic ERG signals generated from real ERG recordings should carry similar information and could be used as an extension for natural data. The synthetic dataset consists of ASD and Control with flash strengths of 1.204, 1.114, 0.949, and 0.799 (log cd.s.m^−2). Synthetic reference signals can enhance medical operational efficiency by offering a feasible alternative to natural ones. Synthesizing facilitates dataset expansion within specialized domains, enabling training resource-intensive networks such as transformers.

Instructions: 

Dataset consists of 8 CSV files of ASD and Control synthetic ERG signals of 4 flash strength.