SpikeDeep-classifier: a deep-learning based fully automatic offline spike sorting algorithm

OBJECTIVE Advancements in electrode design have resulted in micro-electrode arrays with hundreds of channels for single cell recordings. In the resulting electrophysiological recordings, each implanted electrode can record spike activity (SA) of one or more neurons along with background activity (BA). The aim of this study is to isolate SA of each neural source. This process is called spike sorting or spike classification. Advanced spike sorting algorithms are time consuming because of the human intervention at various stages of the pipeline. Current approaches lack generalization because the values of hyperparameters are not fixed, even for multiple recording sessions of the same subject. In this study, a fully automatic spike sorting algorithm called "SpikeDeep-Classifier" is proposed. The values of hyperparameters remain fixed for all the evaluation data. APPROACH The proposed approach is based on our previous study (SpikeDeeptector) and a novel background activity rejector (BAR), which are both supervised learning algorithms and an unsupervised learning algorithm (K-means). SpikeDeeptector and BAR are used to extract meaningful channels and remove BA from the extracted meaningful channels, respectively. The process of clustering becomes straight-forward once the BA is completely removed from the data. Then, K-means with a predefined maximum number of clusters is applied on the remaining data originating from neural sources only. Lastly, a similarity-based criterion and a threshold are used to keep distinct clusters and merge similar looking clusters. The proposed approach is called cluster accept or merge (CAOM) and it has only two hyperparameters (maximum number of clusters and similarity threshold) which are kept fixed for all the evaluation data after tuning. MAIN RESULTS We compared the results of our algorithm with ground-truth labels. The algorithm is evaluated on data of human patients and publicly available labeled non-human primates (NHPs) datasets. The average accuracy of BAR on datasets of human patients is 92.3% which is further reduced to 88.03% after (K-means + CAOM). In addition, the average accuracy of BAR on a publicly available labeled dataset of NHPs is 95.40% which reduces to 86.95% after (K-mean + CAOM). Lastly, we compared the performance of the SpikeDeep-Classifier with two human experts, where SpikeDeep-Classifier has produced comparable results. SIGNIFICANCE The results demonstrate that "SpikeDeep-Classifier" possesses the ability to generalize well on a versatile dataset and henceforth provides a generalized well on a versatile dataset and henceforth provides a generalized and fully automated solution to offline spike sorting.

[1]  Kenneth D Harris,et al.  Towards reliable spike-train recordings from thousands of neurons with multielectrodes , 2012, Current Opinion in Neurobiology.

[2]  S. P. Lloyd,et al.  Least squares quantization in PCM , 1982, IEEE Trans. Inf. Theory.

[3]  Wen-Jyi Hwang,et al.  Spike Detection Based on Normalized Correlation with Automatic Template Generation , 2014, Sensors.

[4]  William S. Rhode,et al.  A neural network-based spike discriminator , 1994, Journal of Neuroscience Methods.

[5]  Yu Zeng,et al.  A Simple Deep Learning Method for Neuronal Spike Sorting , 2017 .

[6]  Sunghan Kim,et al.  Automatic spike detection based on adaptive template matching for extracellular neural recordings , 2007, Journal of Neuroscience Methods.

[7]  Jeremy F. Magland,et al.  A Fully Automated Approach to Spike Sorting , 2017, Neuron.

[8]  Vijay Viswam,et al.  A 1024-Channel CMOS Microelectrode Array With 26,400 Electrodes for Recording and Stimulation of Electrogenic Cells In Vitro , 2014, IEEE Journal of Solid-State Circuits.

[9]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[10]  H. Eichenbaum,et al.  Extracellular neural recording with multichannel microelectrodes , 1986 .

[11]  M. Abeles,et al.  Multispike train analysis , 1977, Proceedings of the IEEE.

[12]  Klaus Schilling,et al.  Automatic template generation for solving n/sup 2/-problems on parallel systems with arbitrary topology , 1997, Proceedings of PDSE '97: 2nd International Workshop on Software Engineering for Parallel and Distributed Systems.

[13]  Xingyu Wang,et al.  Automatic spike detection based on real-time multi-channel template , 2011, 2011 4th International Conference on Biomedical Engineering and Informatics (BMEI).

[14]  Konrad P. Körding,et al.  Linear-nonlinear-time-warp-poisson models of neural activity , 2018, Journal of Computational Neuroscience.

[15]  Sotiris B. Kotsiantis,et al.  Supervised Machine Learning: A Review of Classification Techniques , 2007, Informatica.

[16]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[17]  David B. Dunson,et al.  Multichannel Electrophysiological Spike Sorting via Joint Dictionary Learning and Mixture Modeling , 2013, IEEE Transactions on Biomedical Engineering.

[18]  R. Andersen,et al.  Decoding motor imagery from the posterior parietal cortex of a tetraplegic human , 2015, Science.

[19]  Martin Vinck,et al.  Unsupervised clustering of temporal patterns in high-dimensional neuronal ensembles using a novel dissimilarity measure , 2018, bioRxiv.

[20]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[21]  Gerhard Gritsch,et al.  F08. Automatic spike detection in intracerebral depth electrode recordings , 2018, Clinical Neurophysiology.

[22]  M. Spira,et al.  Multi-electrode array technologies for neuroscience and cardiology. , 2013, Nature nanotechnology.

[23]  Pierre Yger,et al.  A spike sorting toolbox for up to thousands of electrodes validated with ground truth recordings in vitro and in vivo , 2018, eLife.

[24]  Timothy J. Blanche,et al.  Python for Large-Scale Electrophysiology , 2009, Front. Neuroinform..

[25]  Gregory Apker,et al.  Multimodal representation of limb endpoint position in the posterior parietal cortex. , 2013, Journal of neurophysiology.

[26]  Eran Stark,et al.  Large-scale, high-density (up to 512 channels) recording of local circuits in behaving animals. , 2014, Journal of neurophysiology.

[27]  Tobias Glasmachers,et al.  Limits of End-to-End Learning , 2017, ACML.

[28]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[29]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[30]  M S Lewicki,et al.  A review of methods for spike sorting: the detection and classification of neural action potentials. , 1998, Network.

[31]  Daniel N Hill,et al.  Quality Metrics to Accompany Spike Sorting of Extracellular Signals , 2011, The Journal of Neuroscience.

[32]  Jiwen Dong,et al.  Simple convolutional neural network on image classification , 2017, 2017 IEEE 2nd International Conference on Big Data Analysis (ICBDA)(.

[33]  Valérie Ventura,et al.  To sort or not to sort: the impact of spike-sorting on neural decoding performance , 2014, Journal of neural engineering.

[34]  Vijay Viswam,et al.  In Vitro Multi-Functional Microelectrode Array Featuring 59 760 Electrodes, 2048 Electrophysiology Channels, Stimulation, Impedance Measurement, and Neurotransmitter Detection Channels , 2017, IEEE Journal of Solid-State Circuits.

[35]  R. Wightman,et al.  Microelectrodes for studying neurobiology. , 2008, Current opinion in chemical biology.

[36]  Tony A. Fields,et al.  Cerebral microdialysis combined with single-neuron and electroencephalographic recording in neurosurgical patients. Technical note. , 1999, Journal of neurosurgery.

[37]  Kenneth L. Shepard,et al.  A very large-scale microelectrode array for cellular-resolution electrophysiology , 2017, Nature Communications.

[38]  Andrew B Schwartz,et al.  Cortical neural prosthetics. , 2004, Annual review of neuroscience.

[39]  U. Frey,et al.  Microelectronic system for high-resolution mapping of extracellular electric fields applied to brain slices. , 2009, Biosensors & bioelectronics.

[40]  Dejan Markovic,et al.  Spike Sorting: The First Step in Decoding the Brain: The first step in decoding the brain , 2012, IEEE Signal Processing Magazine.

[41]  Ioannis Iossifidis,et al.  SpikeDeeptector: a deep-learning based method for detection of neural spiking activity , 2019, Journal of neural engineering.

[42]  A. Hodgkin,et al.  A quantitative description of membrane current and its application to conduction and excitation in nerve , 1952, The Journal of physiology.

[43]  Steven M Chase,et al.  Control of a brain–computer interface without spike sorting , 2009, Journal of neural engineering.

[44]  Wei Chu,et al.  Multi-category Classification by Soft-Max Combination of Binary Classifiers , 2003, Multiple Classifier Systems.

[45]  R. Quiroga.  Concept cells: the building blocks of declarative memory functions , 2012, Nature Reviews Neuroscience.

[46]  José Carlos Príncipe,et al.  Ascertaining the importance of neurons to develop better brain-machine interfaces , 2004, IEEE Transactions on Biomedical Engineering.

[47]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[48]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[49]  P. M. Horton,et al.  Spike sorting based upon machine learning algorithms (SOMA) , 2007, Journal of Neuroscience Methods.

[50]  Mark A. Clements,et al.  Spike Sorting by Joint Probabilistic Modeling of Neural Spike Trains and Waveforms , 2014, Comput. Intell. Neurosci..

[51]  Matthew Fellows,et al.  On the variability of manual spike sorting , 2004, IEEE Transactions on Biomedical Engineering.

[52]  Andrew Jackson,et al.  Minimum requirements for accurate and efficient real-time on-chip spike sorting , 2014, Journal of Neuroscience Methods.

[53]  Tomoki Fukai,et al.  Spike sorting of heterogeneous neuron types by multimodality-weighted PCA and explicit robust variational Bayes , 2012, Front. Neuroinform..

[54]  Cyrille Rossant,et al.  Spike sorting for large, dense electrode arrays , 2015 .

[55]  E. Fetz,et al.  Volitional control of single cortical neurons in a brain–machine interface , 2011, Journal of neural engineering.

[56]  Michael J. Berry,et al.  Mapping a Complete Neural Population in the Retina , 2012, The Journal of Neuroscience.

[57]  Kenneth D. Harris,et al.  High-Dimensional Cluster Analysis with the Masked EM Algorithm , 2013, Neural Computation.

[58]  Daniel Micol,et al.  Nev2lkit: a New Open Source Tool for Handling neuronal Event Files from Multi-electrode recordings , 2014, Int. J. Neural Syst..

[59]  Linh Hoang,et al.  1/f Neural Noise Reduction and Spike Feature Extraction Using a Subset of Informative Samples , 2010, Annals of Biomedical Engineering.

[60]  Kenneth D. Harris,et al.  Fast and accurate spike sorting of high-channel count probes with KiloSort , 2016, NIPS.

[61]  Fei-Fei Li,et al.  ImageNet: A large-scale hierarchical image database , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[62]  Robert E. Kass,et al.  Comparison of brain–computer interface decoding algorithms in open-loop and closed-loop control , 2010, Journal of Computational Neuroscience.

[63]  Spencer Kellis,et al.  Hand Shape Representations in the Human Posterior Parietal Cortex , 2015, The Journal of Neuroscience.

[64]  Johannes Stallkamp,et al.  The German Traffic Sign Recognition Benchmark: A multi-class classification competition , 2011, The 2011 International Joint Conference on Neural Networks.

[65]  Anders Krogh,et al.  A Simple Weight Decay Can Improve Generalization , 1991, NIPS.

[66]  Shie Mannor,et al.  The cross entropy method for classification , 2005, ICML.

[67]  Mikhail A. Lebedev,et al.  How to read neuron-dropping curves? , 2014, Front. Syst. Neurosci..

[68]  Ning Qian,et al.  On the momentum term in gradient descent learning algorithms , 1999, Neural Networks.

[69]  A. Lambacher,et al.  Identifying firing mammalian neurons in networks with high-resolution multi-transistor array (MTA) , 2011 .

[70]  Rodrigo Quian Quiroga,et al.  Past, present and future of spike sorting techniques , 2015, Brain Research Bulletin.

[71]  E. Wanke,et al.  Exact distinction of excitatory and inhibitory neurons in neural networks: a study with GFP-GAD67 neurons optically and electrophysiologically recognized on multielectrode arrays , 2012, Front. Neural Circuits.

[72]  Jorge Cadima,et al.  Principal component analysis: a review and recent developments , 2016, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[73]  Kenneth D Harris,et al.  Improving data quality in neuronal population recordings , 2016, Nature Neuroscience.

[74]  S. Herculano‐Houzel.  The Human Brain in Numbers: A Linearly Scaled-up Primate Brain , 2009, Front. Hum. Neurosci..