ECG Recordings as Predictors of very early Autism Likelihood: A Machine Learning Approach

Deepa Tilwani, Jessica Bradshaw, Amit Sheth and Christian O’Reilly

This repository contains the code used for the paper entitled "ECG Recordings as Predictors of very early Autism Likelihood: A Machine Learning Approach" and in revision for Bioengineering. To cite :

@article{tilwani2023ecg,
  title={ECG Recordings as Predictors of Very Early Autism Likelihood: A Machine Learning Approach},
  author={Tilwani, D and Bradshaw, J and Sheth, AP and O'Reilly, C},
  publisher={Preprints},
  year={2023}
  doi={https://doi.org/10.20944/preprints202305.0713.v1},
}

Abstract

In recent years, there has been a rise in the prevalence of autism spectrum disorder (ASD). 1 The diagnosis of ASD requires behavioral observation and standardized testing completed by highly 2 trained experts. Early intervention for ASD can begin as early as 1-2 years of age, but ASD diagnoses 3 are not typically made until ages 2-5 years, thus delaying the start of intervention. There is an 4 urgent need for non-invasive biomarkers to detect ASD in infancy. While previous research using 5 physiological recordings has focused on brain-based biomarkers of ASD, this study investigated the 6 potential of electrocardiogram (ECG) recordings as an ASD biomarker in 3-6-month-old infants. We 7 recorded heart activity of infants at typical and elevated familial likelihood for ASD during naturalistic 8 interactions with objects and caregivers. After obtaining the ECG signals, features such as heart rate 9 variability (HRV) and sympathetic and parasympathetic activities were extracted from them. Then 10 we evaluated the effectiveness of multiple machine learning classifiers for the classification of ASD 11 likelihood. Our findings support our hypothesis that infant ECG signals contain a significant amount 12 of information about ASD familial likelihood. Among the various machine learning algorithms tested, KNN performed best according to sensitivity (0.70 ± 0.117), f1-score (0.689 ± 0.124), precision (0.717 ± 0.128 14 ), accuracy (0.70 ± 0.117, p-value = 0.02), and ROC (0.686 ± 0.122, p-value = 0.06). These results suggest that ECG signals contains relevant information about the likelihood of an infant to 16 develop ASD. Future studies should consider the potential of information contained in ECG, and 17 other indices of autonomic control, for the development of biomarkers of ASD in infancy.