Impact Factor 3.648 | CiteScore 3.99
More on impact ›

Original Research ARTICLE Provisionally accepted The full-text will be published soon. Notify me

Front. Neurosci. | doi: 10.3389/fnins.2019.01222

High Accuracy Classification of School-aged Barbadian Children with Histories of Early Protein Malnutrition (PEM) using an Source Imaging based Age-adjusted

  • 1Clinical Hospital of Chengdu Brain Science Institute, China
  • 2The Clinical Hospital of Chengdu Brain Sciences Institute, University of Electronic Science and Technology of China, China
  • 3Cuban Neuroscience Center, Cuba
  • 4Neurology & Neurosurgery, McGill University, Canada
  • 5Chester M. Pierce MD Division of Global Psychiatry, Massachusetts General Hospital, United States
  • 6Judge Baker Children's Center, Harvard Medical School, United States
  • 7McGill Centre for Integrative Neuroscience, Montreal Neurological Institute, Canada

We provide an electroencephalographic (EEG) based statistical classifier that correctly identifies children with histories of Protein Energy Malnutrition (PEM) in the first year of life and distinguished them from controls with 0.82% accuracy (area under the ROC curve). Our previous result in the same participants achieved equivalent accuracy but was based on scalp quantitative EEG features which precluded anatomical interpretation. In contrast, we now employ BC-VARETA, a novel high-resolution EEG source imaging method with minimal leakage and maximal sparseness. This allowed us to identify, to our knowledge for the first time, a classifier in the source space. The EEGs was recorded in 1978 in 108 children who were 5 to 11 years and were participants in the 45+ year longitudinal Barbados Nutrition Study. The PEM cohort experienced moderate-severe protein-energy malnutrition limited to the first year of life; healthy controls were classmates who were matched by age, gender and handedness. In the current study, we employed a machine learning approach based on the elastic net to create a stable sparse classifier. Interestingly, the classifier was driven predominantly by nutrition group differences in alpha activity in the lingual gyrus. This structure is part of the pathway associated with generating alpha rhythm generation which increases with normal maturation. The PEM group showed a significant decrease in alpha activity, suggesting a developmental delay in brain maturation. Childhood malnutrition is still a serious worldwide public health problem. Its consequences are particularly severe when present during early development- deficits during this critical period are permanent and predict impaired cognitive and behavioral functioning over the lifespan. Our EEG source classifier may provide a functionally interpretable diagnostic technology to study the effects on the brain of malnutrition, with wide applicability in low resources settings.

Keywords: Children, EEG, Protein-energy malnutrition (PEM), Source analysis, Classification algorithm

Received: 05 Jun 2019; Accepted: 29 Oct 2019.

Copyright: © 2019 Bringas, Guo, Tang, Razzaq, Calzada-Reyes, Paz-Linares, Galan Garcia, Rabinowitz, Galler, Bosch-Bayard and Valdes-Sosa. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

* Correspondence: Prof. Maria L. Bringas, Clinical Hospital of Chengdu Brain Science Institute, Chengdu, Sichuan Province, China,