Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
153 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Convolutional Spiking Neural Networks for Detecting Anticipatory Brain Potentials Using Electroencephalogram (2208.06900v3)

Published 14 Aug 2022 in cs.NE

Abstract: Spiking neural networks (SNNs) are receiving increased attention because they mimic synaptic connections in biological systems and produce spike trains, which can be approximated by binary values for computational efficiency. Recently, the addition of convolutional layers to combine the feature extraction power of convolutional networks with the computational efficiency of SNNs has been introduced. This paper studies the feasibility of using a convolutional spiking neural network (CSNN) to detect anticipatory slow cortical potentials (SCPs) related to braking intention in human participants using an electroencephalogram (EEG). Data was collected during an experiment wherein participants operated a remote-controlled vehicle on a testbed designed to simulate an urban environment. Participants were alerted to an incoming braking event via an audio countdown to elicit anticipatory potentials that were measured using an EEG. The CSNN's performance was compared to a standard CNN, EEGNet and three graph neural networks via 10-fold cross-validation. The CSNN outperformed all the other neural networks, and had a predictive accuracy of 99.06 percent with a true positive rate of 98.50 percent, a true negative rate of 99.20 percent and an F1-score of 0.98. Performance of the CSNN was comparable to the CNN in an ablation study using a subset of EEG channels that localized SCPs. Classification performance of the CSNN degraded only slightly when the floating-point EEG data were converted into spike trains via delta modulation to mimic synaptic connections.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (60)
  1. Christensen, D. V. et al. 2022 roadmap on neuromorphic computing and engineering. \JournalTitleNeuromorphic Computing and Engineering 2, 022501, DOI: 10.1088/2634-4386/ac4a83 (2022).
  2. Spiking neural networks: Background, recent development and the neucube architecture. \JournalTitleNeural Processing Letters 52, 1675–1701, DOI: 10.1007/s11063-020-10322-8 (2020).
  3. Spiking neural networks for computational intelligence: An overview. \JournalTitleBig Data and Cognitive Computing 5, 67, DOI: 10.3390/bdcc5040067 (2021).
  4. Deep learning in spiking neural networks. \JournalTitleNeural Networks 111, 47–63, DOI: 10.1016/j.neunet.2018.12.002 (2019).
  5. Davies, M. et al. Advancing neuromorphic computing with loihi: A survey of results and outlook. \JournalTitleProceedings of the IEEE 109, 911–934, DOI: 10.1109/jproc.2021.3067593 (2021).
  6. Deep spiking convolutional neural network trained with unsupervised spike-timing-dependent plasticity. \JournalTitleIEEE Transactions on Cognitive and Developmental Systems 11, 384–394, DOI: 10.1109/TCDS.2018.2833071 (2019).
  7. Feature extraction using spiking convolutional neural networks. \JournalTitleProceedings of the International Conference on Neuromorphic Systems DOI: 10.1145/3354265.3354279 (2019).
  8. Training deep spiking convolutional neural networks with stdp-based unsupervised pre-training followed by supervised fine-tuning. \JournalTitleFrontiers in Neuroscience 12, DOI: 10.3389/fnins.2018.00435 (2018).
  9. Deep spiking convolutional neural network for single object localization based on deep continuous local learning. \JournalTitle2021 International Conference on Content-Based Multimedia Indexing (CBMI) DOI: 10.1109/cbmi50038.2021.9461880 (2021).
  10. Convolutional spiking neural network model for robust face detection. In Proceedings of the 9th International Conference on Neural Information Processing, 2002. ICONIP ’02., vol. 2, 660–664 vol.2, DOI: 10.1109/ICONIP.2002.1198140 (2002).
  11. Unsupervised speech recognition through spike-timing-dependent plasticity in a convolutional spiking neural network. \JournalTitlePLOS ONE 13, DOI: 10.1371/journal.pone.0204596 (2018).
  12. A new spiking convolutional recurrent neural network (scrnn) with applications to event-based hand gesture recognition. \JournalTitleFrontiers in Neuroscience 14, DOI: 10.3389/fnins.2020.590164 (2020).
  13. Classification of alzheimer’s disease using deep convolutional spiking neural network. \JournalTitleNeural Processing Letters 53, 2649–2663, DOI: 10.1007/s11063-021-10514-w (2021).
  14. A survey of automotive driving assistance systems technologies. In 2018 International Conference on Artificial Intelligence and Data Processing (IDAP), 1–12, DOI: 10.1109/IDAP.2018.8620826 (2018).
  15. Rashid, M. et al. Current status, challenges, and possible solutions of eeg-based brain-computer interface: A comprehensive review. \JournalTitleFrontiers in Neurorobotics 14, DOI: 10.3389/fnbot.2020.00025 (2020).
  16. EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces. \JournalTitleJournal of Neural Engineering 15, 056013, DOI: 10.1088/1741-2552/aace8c (2018).
  17. Driver’s cognitive state classification toward brain computer interface via using a generalized and supervised technology. \JournalTitleThe 2010 International Joint Conference on Neural Networks (IJCNN) DOI: 10.1109/ijcnn.2010.5596835 (2010).
  18. Eeg-based cognitive state monitoring and predition by using the self-constructing neural fuzzy system. \JournalTitleProceedings of 2010 IEEE International Symposium on Circuits and Systems DOI: 10.1109/iscas.2010.5536955 (2010).
  19. A multimodal approach to estimating vigilance using eeg and forehead eog. \JournalTitleJournal of Neural Engineering 14, 026017, DOI: 10.1088/1741-2552/aa5a98 (2017).
  20. Prediction of fatigue-related driver performance from eeg data by deep riemannian model. \JournalTitle2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) DOI: 10.1109/embc.2017.8037774 (2017).
  21. Detection of self-paced reaching movement intention from eeg signals. \JournalTitleFrontiers in Neuroengineering 5, DOI: 10.3389/fneng.2012.00013 (2012).
  22. Single trial analysis of slow cortical potentials: A study on anticipation related potentials. \JournalTitleJournal of Neural Engineering 10, 036014, DOI: 10.1088/1741-2560/10/3/036014 (2013).
  23. Single trial prediction of self-paced reaching directions from eeg signals. \JournalTitleFrontiers in Neuroscience 8, DOI: 10.3389/fnins.2014.00222 (2014).
  24. Haufe, S. et al. Eeg potentials predict upcoming emergency brakings during simulated driving. \JournalTitleJournal of Neural Engineering 8, 056001, DOI: 10.1088/1741-2560/8/5/056001 (2011).
  25. Action prediction based on anticipatory brain potentials during simulated driving. \JournalTitleJournal of Neural Engineering 12, 066006, DOI: 10.1088/1741-2560/12/6/066006 (2015).
  26. Spiking neural networks for classification of brain-computer interface and image data. In 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 3624–3629, DOI: 10.1109/BIBM52615.2021.9669864 (2021).
  27. Pals, M. et al. Demonstrating the viability of mapping deep learning based eeg decoders to spiking networks on low-powered neuromorphic chips. In 2021 43rd Annual International Conference of the IEEE Engineering in Medicine Biology Society (EMBC), 6102–6105, DOI: 10.1109/EMBC46164.2021.9629621 (2021).
  28. Luo, Y. et al. Eeg-based emotion classification using spiking neural networks. \JournalTitleIEEE Access 8, 46007–46016, DOI: 10.1109/ACCESS.2020.2978163 (2020).
  29. Spiking neural network for augmenting electroencephalographic data for brain computer interfaces. \JournalTitleFrontiers in Neuroscience 15, DOI: 10.3389/fnins.2021.651762 (2021).
  30. Brain-inspired spiking neural networks for decoding and understanding muscle activity and kinematics from electroencephalography signals during hand movements. \JournalTitleScientific Reports 11, DOI: 10.1038/s41598-021-81805-4 (2021).
  31. Eeg classification with spiking neural network: Smaller, better, more energy efficient. \JournalTitleSmart Health 24, 100261, DOI: 10.1016/j.smhl.2021.100261 (2022).
  32. Deep learning of eeg data in the neucube brain-inspired spiking neural network architecture for a better understanding of depression. \JournalTitleNeural Information Processing 195–206, DOI: 10.1007/978-3-030-36718-3_17 (2019).
  33. Chen, X. et al. Toward open-world electroencephalogram decoding via deep learning: A comprehensive survey. \JournalTitleIEEE Signal Processing Magazine 39, 117–134, DOI: 10.1109/MSP.2021.3134629 (2022).
  34. Merolla, P. A. et al. A million spiking-neuron integrated circuit with a scalable communication network and interface. \JournalTitleScience 345, 668–673, DOI: 10.1126/science.1254642 (2014).
  35. Jo, S. H. et al. Nanoscale memristor device as synapse in neuromorphic systems. \JournalTitleNano Letters 10, 1297–1301, DOI: 10.1021/nl904092h (2010).
  36. Hardware-friendly neural network architecture for neuromorphic computing. \JournalTitleCoRR abs/1906.08853 (2019). 1906.08853.
  37. Neuromorphic artificial intelligence systems. \JournalTitleFrontiers in Neuroscience 16, DOI: 10.3389/fnins.2022.959626 (2022).
  38. One-shot learning with spiking neural networks. \JournalTitlebioRxiv. DOI: 10.1101/2020.06.17.156513 (2020).
  39. Eeg-gnn: Graph neural networks for classification of electroencephalogram (eeg) signals. \JournalTitle2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) DOI: 10.1109/embc46164.2021.9630194 (2021).
  40. Attention-based graph resnet for motor intent detection from raw eeg signals, DOI: 10.48550/ARXIV.2007.13484 (2020).
  41. Hou, Y. et al. Gcns-net: A graph convolutional neural network approach for decoding time-resolved eeg motor imagery signals, DOI: 10.48550/ARXIV.2006.08924 (2020).
  42. Eeg-based emotion recognition using regularized graph neural networks, DOI: 10.48550/ARXIV.1907.07835 (2019).
  43. Eeg-based detection of driver emergency braking intention for brain-controlled vehicles. \JournalTitleIEEE Transactions on Intelligent Transportation Systems 19, 1766–1773, DOI: 10.1109/tits.2017.2740427 (2018).
  44. Detection of braking intention in diverse situations during simulated driving based on eeg feature combination. \JournalTitleJournal of Neural Engineering 12, 016001, DOI: 10.1088/1741-2560/12/1/016001 (2014).
  45. Khaliliardali, Z. et al. Real-time detection of driver’s movement intention in response to traffic lights, DOI: 10.1101/443390 (2019).
  46. Eeg-based detection of braking intention under different car driving conditions. \JournalTitleFrontiers in Neuroinformatics 12, DOI: 10.3389/fninf.2018.00029 (2018).
  47. Detection of driver braking intention using eeg signals during simulated driving. \JournalTitleSensors 19, 2863, DOI: 10.3390/s19132863 (2019).
  48. Detecting driver’s braking intention using recurrent convolutional neural networks based eeg analysis. \JournalTitle2017 4th IAPR Asian Conference on Pattern Recognition (ACPR) DOI: 10.1109/acpr.2017.86 (2017).
  49. Graph neural networks in tensorflow and keras with spektral [application notes]. \JournalTitleIEEE Computational Intelligence Magazine 16, 99–106, DOI: 10.1109/mci.2020.3039072 (2021).
  50. Paszke, A. et al. Pytorch: An imperative style, high-performance deep learning library. In Wallach, H. et al. (eds.) Advances in Neural Information Processing Systems 32, 8024–8035 (Curran Associates, Inc., 2019).
  51. Abadi, M. et al. TensorFlow: Large-scale machine learning on heterogeneous systems. https://www.tensorflow.org/. (2015).
  52. Eshraghian, J. K. et al. Training spiking neural networks using lessons from deep learning. \JournalTitlearXiv preprint arXiv:2109.12894 (2021).
  53. Wilson, R. J. Introduction to graph theory (Prentice Hall, 2015).
  54. Wu, Z. et al. A comprehensive survey on graph neural networks. \JournalTitleIEEE Transactions on Neural Networks and Learning Systems 32, 4–24, DOI: 10.1109/tnnls.2020.2978386 (2021).
  55. Semi-supervised classification with graph convolutional networks, DOI: 10.48550/ARXIV.1609.02907 (2016).
  56. How powerful are graph neural networks?, DOI: 10.48550/ARXIV.1810.00826 (2018).
  57. Neuroelectrics. https://www.neuroelectrics.com. Accessed: 18-May-2022.
  58. Eeglab: An open source toolbox for analysis of single-trial eeg dynamics including independent component analysis. \JournalTitleJournal of Neuroscience Methods 134, 9–21, DOI: 10.1016/j.jneumeth.2003.10.009 (2004).
  59. Ben-Shachar, M. S. mattansb/tbt: Channel you inner error, DOI: 10.5281/zenodo.5948294 (2022).
  60. Bagdasarov, A. et al. Spatiotemporal dynamics of eeg microstates in four- to eight-year-old children: Age- and sex-related effects, DOI: 10.31234/osf.io/x35uf (2022).
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now