Effectively Interpreting Electroencephalogram Classification Using the Shapley Sampling Value to Prune a Feature Tree

Identifying the features that contribute to classification using machine learning remains a challenging problem in terms of the interpretability and computational complexity of the endeavor. Especially in electroencephalogram (EEG) medical applications, it is important for medical doctors and patients to understand the reason for the classification. In this paper, we thus propose a method to quantify contributions of interpretable EEG features on classification using the Shapley sampling value (SSV). In addition, a pruning method is proposed to reduce the SSV computation cost. The pruning is conducted on an EEG feature tree, specifically at the sensor (electrode) level, frequency-band level, and amplitude-phase level. If the contribution of a feature at a high level (e.g., sensor level) is very small, the contributions of features at a lower level (e.g., frequency-band level) should also be small. The proposed method is verified using two EEG datasets: classification of sleep states, and screening of alcoholics. The results show that the method reduces the SSV computational complexity while maintaining high SSV accuracy. Our method will thus increase the importance of data-driven approaches in EEG analysis.

[1]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[2]  Yitong Li,et al.  Targeting EEG/LFP Synchrony with Neural Nets , 2017, NIPS.

[3]  Ankur Taly,et al.  Axiomatic Attribution for Deep Networks , 2017, ICML.

[4]  Scott Lundberg,et al.  A Unified Approach to Interpreting Model Predictions , 2017, NIPS.

[5]  L. S. Shapley,et al.  17. A Value for n-Person Games , 1953 .

[6]  Erik Strumbelj,et al.  Explaining prediction models and individual predictions with feature contributions , 2014, Knowledge and Information Systems.

[7]  Jeffrey M. Hausdorff,et al.  Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol , 2000 .

[8]  Cengiz Öztireli,et al.  Towards better understanding of gradient-based attribution methods for Deep Neural Networks , 2017, ICLR.

[9]  Avanti Shrikumar,et al.  Learning Important Features Through Propagating Activation Differences , 2017, ICML.

[10]  Aamir Saeed Malik,et al.  A review on EEG-based methods for screening and diagnosing alcohol use disorder , 2018, Cognitive Neurodynamics.

[11]  Klaus-Robert Müller,et al.  Interpretable deep neural networks for single-trial EEG classification , 2016, Journal of Neuroscience Methods.

[12]  Wolfram Burgard,et al.  Deep learning with convolutional neural networks for EEG decoding and visualization , 2017, Human brain mapping.

[13]  Tonio Ball,et al.  Hierarchical internal representation of spectral features in deep convolutional networks trained for EEG decoding , 2017, 2018 6th International Conference on Brain-Computer Interface (BCI).

[14]  Lars Kai Hansen,et al.  Deep convolutional neural networks for interpretable analysis of EEG sleep stage scoring , 2017, 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP).

[15]  Markus H. Gross,et al.  Gradient-Based Attribution Methods , 2019, Explainable AI.

[16]  Alexander Binder,et al.  On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation , 2015, PloS one.

[17]  Gleb V. Tcheslavski,et al.  Alcoholism-related alterations in spectrum, coherence, and phase synchrony of topical electroencephalogram , 2012, Comput. Biol. Medicine.

[18]  Brent Lance,et al.  EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces , 2016, Journal of neural engineering.