Evaluation of Interpretability for Deep Learning algorithms in EEG Emotion Recognition: A case study in Autism

Mayor-Torres, Juan Manuel; Medina-DeVilliers, Sara; Clarkson, Tessa; Lerner, Matthew D.; Riccardi, Giuseppe

doi:10.48550/arxiv.2111.13208

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These input nodes correspond to the cortical regions, in which the neural activity is substantially different between normal controls and RLS patients. Recent advances in machine learning, especially deep neural networks, have enabled improved analysis of multidimensional data such as high-density EEG, and are extensively adopted for pattern recognition and the estimation of neural information [ 15 , 16 , 17 , 18 ].…”

Section: Introductionmentioning

confidence: 99%

Explainable Machine-Learning-Based Characterization of Abnormal Cortical Activities for Working Memory of Restless Legs Syndrome Patients

Kim

Seo

et al. 2022

Sensors

View full text Add to dashboard Cite

Restless legs syndrome (RLS) is a sensorimotor disorder accompanied by a strong urge to move the legs and an unpleasant sensation in the legs, and is known to accompany prefrontal dysfunction. Here, we aimed to clarify the neural mechanism of working memory deficits associated with RLS using machine-learning-based analysis of single-trial neural activities. A convolutional neural network classifier was developed to discriminate the cortical activities between RLS patients and normal controls. A layer-wise relevance propagation was applied to the trained classifier in order to determine the critical nodes in the input layer for the output decision, i.e., the time/location of cortical activities discriminating RLS patients and normal controls during working memory tasks. Our method provided high classification accuracy (~94%) from single-trial event-related potentials, which are known to suffer from high inter-trial/inter-subject variation and low signal-to-noise ratio, after strict separation of training/test/validation data according to leave-one-subject-out cross-validation. The determined critical areas overlapped with the cortical substrates of working memory, and the neural activities in these areas were correlated with some significant clinical scores of RLS.

show abstract