Publications
Publications and recent conference papers are available below.
Generating Synthetic Light-Adapted Electroretinogram Waveforms Using Artificial Intelligence to Improve Classification of Retinal Conditions in Under-Represented Populations
In this first work we have demonstrated the use of AI to generate realistic ERG waveforms from existing data. This first step may allow future studies to incorporate natural with AI geneearted data to enhance classification models of neurodiverse populations.
Attention to the Electroretinogram: Gated Multilayer Perceptron for ASD Classification
In this publication we used a new approach to classification with gated multilayer perceptrons to identify ASD from control ERG waveforms. A classification accuracy of 85% was achieved. This suggests that using a visual based (perceptron) interpretation of the waveform may further improve the classification accuracy of the ERG in ASD.
Detecting Autism Spectrum Disorder and Attention Deficit Hyperactivity Disorder Using Multimodal Time-Frequency Analysis with Machine Learning Using the Electroretinogram from Two Flash Strengths
In this publication we aimed to use a minimum number of flash strengths that were selected based on contributions from the OFF- and ON- retinal signalling pathways. By comparing the spectral composition of the waveforms in ASD, control and ADHD individuals and applying machine learning models a classification accuracy of 70% was achieved. This suggests that a greater number and or additional physiological or psychological markers may be required to enhance classification.
Autism spectrum disorder detection using variable frequency complex demodulation of the electroretinogram
We analyzed the raw ERG waveforms using signal analysis of the electroretinogram (ERG) and computed features to train ML models such as Random Forest, Gradient Boosting, Support Vector Machine to classify ASD from controls, We found that the ML models with VFCDM features outperformed models using the DWT, achieving an area under the receiver operating characteristics curve of 0.90 (accuracy = 0.81, sensitivity = 0.85, specificity = 0.78). We found that the higher frequency range (80–300 Hz) included more relevant information for classifying ASD compared to the lower frequencies. We also found that the stronger flash strength of 446 Td.s in the right eye provided the best classification result which supports VFCDM analysis of the ERG waveform as a potential tool to aid in the identification of the ASD phenotype.
Role of noninvasive ocular imaging as a biomarker in peripheral artery disease: A systematic review
As Peripheral artery disease (PAD)causes thinning and ischemic changes in retinal vessels, examination of the retinal vessels using retinal imaging techniques can provide useful information about early microvascular damage in PAD. Ocular imaging could potentially serve as a biomarker for PAD.
Prototype for Smartphone-based Electroretinogram
Spectral analysis of Electroretinography to differentiate autism spectrum disorder and attention deficit hyperactivity disorder
Detecting Autism Spectrum Disorder Using Spectral Analysis of Electroretinogram and Machine Learning: Preliminary results
Preliminary results first reported at the 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). We showed that time-domain features were able to detect ASD with a maximum 65% accuracy. However, the classification accuracy of our best ML model using time-domain and spectral features was 86%, with 98% sensitivity. Our preliminary results indicate that spectral analysis of ERG provides helpful information for the classification of ASD.
Optimal Combination of Mother Wavelet and AI Model for Precise Classification of Pediatric Electroretinogram Signals
Application of AI modelling for a large datset of paediatric ERG recordings. The study provides valuable insights into the effectiveness of different combinations of wavelets and models in classifying ERG wavelet scalograms.
Group comparisons of the individual electroretinogram time trajectories for the ascending limb of the b-wave using a raw and registered time series
The paper presents a statistical graphics-based analysis of the ascending limb of the b-wave (AL-b) of the electroretinogram in children with and without a diagnosis of autism spectrum disorder (ASD). A narrative explanation of the statistical approach is given to illustrate how different features of the waveform based on location and scale derived from raw and registered time series can reveal subtle differences between the groups.
Discrete Wavelet Transform Analysis of the Electroretinogram in Autism Spectrum Disorder and Attention Deficit Hyperactivity Disorder
Spectral analysis of the electroretinogram in ASD, ADHD and control children revealed a difference in the energies contained within the b-wave and the oscillatory potentials. This early finding suggested that the ERG waveform energy content within different frequncy bands differred between these groups.
The photopic negative response in autism spectrum disorder
An analysis of the ganglion cell component of the full field electroretinogram in ASD found no significant differences compared to the control population. This finding suggested that retinal ganglion cell function was not affected in this young ASD population.
Light-Adapted Electroretinogram Differences in Autism Spectrum Disorder
The first multi centre study with Ed Ritvo that identified a reduced b-wave amplitude in ASD children compared to age matched controls. The study suggested that the main difference was in the ON-retinal pathway with glutamate regulation a possible explanation for these findings.
Full-field electroretinogram in autism spectrum disorder
The first study since Ed Ritvo's pioneering 1988 study, that identified in adults with ASD a reduced b-wave amplitude. The study also confirmed reduced b-wave amplitudes under dark adapted conditions and identified a difference in the shape of the high frequency oscillatory potential components of the ERG under light adapted conditions.