Recurrence quantification analysis of resting state EEG signals in autism spectrum disorder - a systematic methodological exploration of technical and demographic confounders in the search for biomarkers

Abstract

Background: Autism spectrum disorder (ASD) is a neurodevelopmental disorder with a worldwide prevalence of 1-2%. In low-resource environments, in particular, early identification and diagnosis is a significant challenge. Therefore, there is a great demand for 'language-free, culturally fair' low-cost screening tools for ASD that do not require highly trained professionals. Electroencephalography (EEG) has seen growing interest as an investigational tool for biomarker development in ASD and neurodevelopmental disorders. One of the key challenges is the identification of appropriate multivariate, next-generation analytical methodologies that can characterise the complex, nonlinear dynamics of neural networks in the brain, mindful of technical and demographic confounders that may influence biomarker findings. The aim of this study was to evaluate the robustness of recurrence quantification analysis (RQA) as a potential biomarker for ASD using a systematic methodological exploration of a range of potential technical and demographic confounders.

Methods: RQA feature extraction was performed on continuous 5-second segments of resting state EEG (rsEEG) data and linear and nonlinear classifiers were tested. Data analysis progressed from a full sample of 16 ASD and 46 typically developing (TD) individuals (age 0-18 years, 4802 EEG segments), to a subsample of 16 ASD and 19 TD children (age 0-6 years, 1874 segments), to an age-matched sample of 7 ASD and 7 TD children (age 2-6 years, 666 segments) to prevent sample bias and to avoid misinterpretation of the classification results attributable to technical and demographic confounders. A clinical scenario of diagnosing an unseen subject was simulated using a leave-one-subject-out classification approach.

Results: In the age-matched sample, leave-one-subject-out classification with a nonlinear support vector machine classifier showed 92.9% accuracy, 100% sensitivity and 85.7% specificity in differentiating ASD from TD. Age, sex, intellectual ability and the number of training and test segments per group were identified as possible demographic and technical confounders. Consistent repeatability, i.e. the correct identification of all segments per subject, was found to be a challenge.

Conclusions: RQA of rsEEG was an accurate classifier of ASD in an age-matched sample, suggesting the potential of this approach for global screening in ASD. However, this study also showed experimentally how a range of technical challenges and demographic confounders can skew results, and highlights the importance of probing for these in future studies. We recommend validation of this methodology in a large and well-matched sample of infants and children, preferably in a low- and middle-income setting.

Keywords: Autism spectrum disorder; RQA; Recurrence quantification analysis; Resting state electroencephalography.

Fig. 1

EEG signal processing methodology

Fig. 2

Progress loop of sample population analysis

Fig. 3

Classification accuracy of different feature sets for full sample cross-validation run 1

Fig. 4

Classification accuracy of different feature sets for subsample cross-validation run 1

Fig. 5

Classification accuracy of different feature sets for age-matched sample cross-validation run 1

Fig. 6

Generalisation performance for age-matched sample cross-validation run 1

Fig. 7

Cross-validation performance for the (a) linear discriminant analysis (LDA), (b) multilayer layer perceptron (MLP) and (c) support vector machine (SVM) classifiers

Fig. 8

Feature shuffle analysis of feature set 1 for age-matched sample cross-validation run 1

Fig. 9

Optimal feature subset identification for age-matched sample cross-validation run 1

Fig. 10

Visualisation of the feature space in a 2D principal component (PC) subspace, (a) training features and (b) test features for age-matched sample cross-validation run 1

Fig. 11

Visualisation of the feature space in a 3D principal component (PC) subspace, (a) training features and (b) test features for age-matched sample cross-validation run 1

Fig. 12

Repeatability analysis for the (a) linear discriminant analysis (LDA), (b) multilayer layer perceptron (MLP) and (c) support vector machine (SVM) classifiers for age-matched sample cross-validation run 1

Fig. 13

Generalisation performance for age-matched sample leave-one-subject-out analysis

Fig. 14

Repeatability analysis for the (a) linear discriminant analysis (LDA), (b) multilayer layer perceptron (MLP) and (c) support vector machine (SVM) classifiers for age-matched sample leave-one-subject-out analysis