Vai ai contenuti. | Spostati sulla navigazione | Spostati sulla ricerca | Vai al menu | Contatti | Accessibilità

| Crea un account

Camarda, Martina (2009) Neural spikes classification in multichannel recordings. [Tesi di dottorato]

Full text disponibile come:

[img]
Anteprima
Documento PDF
5Mb

Abstract (inglese)

In order to study the neural behavior, scientists employ multichannel probes where each electrode records a mixture of spike trains from surrounding neurons. A first necessary step is the individuation and the separation of signal from different sources, associating each detected spike to the neuron of origin. Many spike sorting algorithms based on different principles have been developed for this purpose, but there is still no consensus on which is the best method.


This thesis addresses the issue of impulsive signal classification in the Neurophysiological framework presenting a novel spike sorting algorithm named Multi-Channel Inversion for Spike Classification (MCI4SC). The new method exploits multichannel information related to neuron positions, and makes a distinctive use of the mixing matrix associated to the measurement channel.
In particular, inverting many matrices derived by the mixing one, the method is able to handle the disadvantageous, but typical, situation where there are more recorded neurons than recording sensors, under the reasonable hypothesis that the number of simultaneously firing neurons is lower than or equal to the number of sensors.
Another distinguishing feature of MCI4SC algorithm and its implementation is the use of the Wavelet Packet Transform. This tool has been used to estimate the ratio between spike amplitudes in different channels, thus leading to a consistent estimation of the mixing matrix components even in case of low signal to noise ratio.



The MCI4SC algorithm has been applied on experimental data with human supervision on the threshold setting. Good spike sorting results have been obtained for bursting Purkinje Cells with varying waveform and amplitude spikes, as well as for noisy neurons in Locust antennal lobe.
Compared with an algorithm based on the Markov Chain Monte Carlo, the MCI4SC algorithm has at least comparable efficiency with a much lower computational time, in addition to the important capability of overlapping spike resolution.
This makes the new algorithm presented in this thesis a reliable and competitive tool in the spike sorting context.

Abstract (italiano)

Per studiare il comportamento dei neuroni vengono utilizzate sonde multicanale in cui ciascun elettrodo misura la sovrapposizione di treni di spike generati dai neuroni circostanti. Un primo passo necessario è quello di individuare e separare i segnali provenienti da diverse sorgenti associando ciascuno spike al neurone che lo ha generato. A questo scopo sono stati sviluppati molti algoritmi di spike sorting (classificazione di spike) che si basano su differenti principi, ma nessun metodo è, finora, stato riconosciuto come migliore degli altri.

Questa tesi affronta il problema della classificazione di segnali impulsivi nel contesto Neurofisiologico presentando un nuovo algoritmo di spike sorting denominato Multi-Channel Inversion for Spike Classification (MCI4SC). Il nuovo metodo sfrutta l'informazione proveniente da più canali (legata alla posizione dei neuroni) e fa un uso distintivo della matrice di mixing associata al canale di misura. In particolare, invertendo più matrici derivate da quella di mixing, il metodo è in grado di gestire la sfavorevole ma tipica situazione in cui sono presenti più neuroni registrati che sensori, sotto la ragionevole ipotesi che il numero di neuroni contemporaneamente attivi sia minore od uguale al numero di sensori.
Un'altra caratteristica distintiva dell'algoritmo MCI4SC e della sua implementazione è l'uso della Trasformata Wavelet Packet. Questo strumento è stato impiegato per stimare il rapporto tra le ampiezze degli spike nei differenti canali, dando così luogo a una stima delle componenti della matrice di mixing che risulta consistente anche nel caso di basso rapporto segnale rumore.

L'algoritmo MCI4SC è stato applicato a dati sperimentali impostando manualmente le soglie. Buoni risultati di classificazione sono stati ottenuti sia nel caso di Purkinje Cells con spikes di differenti ampiezze e forme d'onda, sia nel caso di neuroni nel lobo antennale della Locusta con segnale molto rumoroso.
Nel confronto con un algoritmo basato sul metodo Markov Chain Monte Carlo, l'algoritmo MCI4SC presenta una efficienza almeno comparabile con un più basso tempo computazionale, oltre alla importante capacità di risolvere spike sovrapposti.
Ciò rende il nuovo algoritmo, presentato in questa tesi, uno strumento affidabile e competitivo nel contesto dello spike sorting.

Statistiche Download - Aggiungi a RefWorks
Tipo di EPrint:Tesi di dottorato
Relatore:Beghi, Alessandro
Dottorato (corsi e scuole):Ciclo 21 > Scuole per il 21simo ciclo > INGEGNERIA DELL'INFORMAZIONE > AUTOMATICA E RICERCA OPERATIVA
Data di deposito della tesi:NON SPECIFICATO
Anno di Pubblicazione:2009
Parole chiave (italiano / inglese):spike sorting, multichannel recording, amplitude ratio, wavelet.
Settori scientifico-disciplinari MIUR:Area 09 - Ingegneria industriale e dell'informazione > ING-INF/05 Sistemi di elaborazione delle informazioni
Struttura di riferimento:Dipartimenti > Dipartimento di Ingegneria dell'Informazione
Codice ID:2093
Depositato il:10 Mar 2010 10:23
Simple Metadata
Full Metadata
EndNote Format

Bibliografia

I riferimenti della bibliografia possono essere cercati con Cerca la citazione di AIRE, copiando il titolo dell'articolo (o del libro) e la rivista (se presente) nei campi appositi di "Cerca la Citazione di AIRE".
Le url contenute in alcuni riferimenti sono raggiungibili cliccando sul link alla fine della citazione (Vai!) e tramite Google (Ricerca con Google). Il risultato dipende dalla formattazione della citazione.

[1] M. A. Abdul-Ghani, T. A. Valiante, and P. S. Pennefather. Sr2+ and Cerca con Google

quantal events at excitatory synapses between mouse hippocampal neurons Cerca con Google

in culture. Journal of Physiology, 495(1):113–125, August 1996. Cerca con Google

[2] M. Abeles and M. H. Goldstein. Multispike train analysis. Proceedings of Cerca con Google

the IEEE, 65(5):762–773, May 1977. Cerca con Google

[3] A. F. Atiya. Recognition of multi-unit neural signals. IEEE Transactions Cerca con Google

on Biomedical Engineering, 39(7):723–729, July 1992. Cerca con Google

[4] I. N. Bankman, K. O. Johnson, and W. Schneider. Optimal detection, Cerca con Google

classification, and superposition resolution in neural waveform recordings. Cerca con Google

IEEE Transaction on Biomedical Engineering, 40(8):836–841, August 1993. Cerca con Google

[5] A. Ben-Israel and T. N. E. Greville. Generalized inverses: theory and Cerca con Google

applications. John Wiley and Sons, New York, 1974. Cerca con Google

[6] H. Bergman and M. R. DeLong. A personal computer-based spike detector Cerca con Google

and sorter: implementation and evaluation. Journal of Neuroscience Cerca con Google

Methods, 41(3):187–97, March 1992. Cerca con Google

[7] D. S. Bernstein. Matrix mathematics: theory, facts, and formulas with Cerca con Google

application to linear systems theory. Princeton University press, Princeton, Cerca con Google

2005. Cerca con Google

[8] M. Blatt, S. Wiseman, and E. Domany. Superparamagnetic clustering of Cerca con Google

data. Physical Review Letters, 76(18):3251–3254, April 1996. Cerca con Google

[9] J. M. Bower, Y.-F. Wong, and J. Banik. Neural networks for template Cerca con Google

matching: Application to real-time classification of the action potentials of Cerca con Google

real neurons. In Neural Information Processing Systems, pages 103–113. Cerca con Google

American Institue of Physics, Denver, Colorado, USA, 1987. Cerca con Google

[10] E. N. Brown, R. E. Kass, and P. P. Mitra. Multiple neural spike train Cerca con Google

data analysis: state-of-the-art and future challenges. Nature Neuroscience,7(5):456–461, May 2004. Cerca con Google

[11] G. D. Brown, S. Yamadab, and Sejnowskic T. J. Independent componentanalysis at the neural cocktail party. Trends in Neurosciences, 24(1):54–63,January 2001. Cerca con Google

[12] C. S. Burrus, R. A. Gopinath, and H. Guo. Introduction to Wavelets andWavelet Transforms: A Primer. Prentice Hall PTR, Upper Saddle River(NJ), 1998. Cerca con Google

[13] G. Buzs´aki. Large-scale recording of neuronal ensembles. Nature Neuro-science, 7(5):446–451, May 2004. Cerca con Google

[14] W. H. Calvin. Some simple spike separation techniques for simultaneouslyrecorded neurons. Electroencephalography and Clinical Neurophysiology,34(1):94–96, January 1973. Cerca con Google

[15] J.-F. Cardoso. Blind signal separation: statistical principles. Proceedingsof the IEEE, 86(10):2009–2025, October 1998. Cerca con Google

[16] R. Chandra and L. M. Optican. Detection, classification, and superpositionresolution of action potentials in multiunit single-channel recordings byan on-line real-time neural network. IEEE Transactions on BiomedicalEngineering, 44(5):403–412, May 1997. Cerca con Google

[17] M. I. Chelaru and M. S. Jog. Spike source localization with tetrodes. Jour-nal of Neuroscience Methods, 142(2):305–315, March 2005. Cerca con Google

[18] H. Cho, D. Corina, J. F. Brinkley, G. A. Ojemann, and L. G. Shapiro. Anew template matching method using variance estimation for spike sorting. Cerca con Google

Neural Engineering, 2005. Conference Proceedings. 2nd International IEEEEMBS Conference, pages 225–228, March 2005. Cerca con Google

[19] J. H. Choi, D. Kim, and T. Kim. A new overlapping resolution methodfor multi-channel spike sorting. In Conference Proceedings. 2nd Interna-tional IEEE EMBS Conference on Neural Engineering, 2005, pages 683–686, March 2005. Cerca con Google

[20] S. Choi, A. Cichocki, H.M. Park, and S. Y. Lee. Blind source separation andindependent component analysis: A review. Neural Information Processing- Letters and Reviews, 6(1):1–57, January 2005. Cerca con Google

[21] J. D. Clements and J. M. Bekkers. Detection of spontaneous synaptic eventswith an optimally scaled template. Biophysical Journal, 73(1):220–229, July1997. Cerca con Google

[22] R. R. Coifman, Y. Meyer, and M. V. Wickerhauser. Adapted waveformanalysis, wavelet-packets and applications. In Robert E. O’Malley Jr., editor,ICIAM 91: Proceedings of the 2nd International Conference on Indus-trial and Applied Mathematics, 8–12 July, 1991, pages 41–50, Philadelphia,1992. SIAM, SIAM Press. Cerca con Google

[23] R. R. Coifman and M.V. Wickerhauser. Entropy based algorithms for bestbasis selection. IEEE Transactions on Information Theory, 32:712–718,March 1992. Cerca con Google

[24] I. Daubechies. Ten Lectures on Wavelets (C B M S - N S F Regional Con-ference Series in Applied Mathematics). Society for Industrial and AppliedMathematics, Philadephia, December 1992. Cerca con Google

[25] M. Delescluse and C. Pouzat. SpikeOMatic Tutorial 2 — SpikeOMaticgoes MCMC. http://www.biomedicale.univ-paris5.fr/physcerv/C Pouzat/Code folder/SpikeOMaticTutorial2.pdf, July 2005. Vai! Cerca con Google

[26] M. Delescluse and C. Pouzat. Efficient spike-sorting of multi-state neuronsusing inter-spike intervals information. Journal of Neuroscience Methods,150(1):16–29, January 2006. Cerca con Google

[27] E. H. D’Hollander and G. A. Orban. Spike recognition and on-line classificationby unsupervised learning system. IEEE Transaction on BiomedicalEngineering, 26(5):279–284, May 1979. Cerca con Google

[28] J. C. Dill, P. C. Lockemann, and K. Nab. An attempt to analyze multi-unitrecordings. Electroencephalography and Clinical Neurophysiology, 28(1):79–82, January 1970. Cerca con Google

[29] G. J. Dinning and A. C. Sanderson. Real-time classification of multiunitneural signals using reduced feature sets. IEEE Transaction on BiomedicalEngineering, 28(12):804–812, December 1981. Cerca con Google

[30] K. L. Drake, K. D.Wise, J. Farraye, D. J. Anderson, and S. L. Bement. Performanceof planar multisite microprobes in recording extracellular singleunitintracortical activity. IEEE Transactions on Biomedical Engineering,35(9):719–732, September 1988. Cerca con Google

[31] R. O. Duda and P. E. Hart. Pattern Classification and Scene Analysis.,chapter 6, pages 217–219. John Wiley and Sons, New York, 1973. Cerca con Google

[32] M. S. Fee, P. P. Mitra, and D. Kleinfeld. Automatic sorting of multiple unitneuronal signals in the presence of anisotropic and non-gaussian variability.Journal of Neuroscience Methods, 69(2):175–188, November 1996. Cerca con Google

[33] M. S. Fee, P. P. Mitra, and D. Kleinfeld. Variability of extracellular spikewaveforms of cortical neurons. Journal of Neurophysiology, 76(6):3823–3833, December 1996. Cerca con Google

[34] G. L. Gerstein, M. J. Bloom, I. E. Espinosa, S. Evanczuk, and M. R. Turner.Design of a laboratory for multineuron studies. IEEE Transactions onSystems, Man, and Cybernetics, 13(5):668–677, 1983.1 Cerca con Google

[35] G. L. Gerstein and W. A. Clark. Simultaneous studies of firing patterns inseveral neurons. Science, 143(3612):1325–1327, March 1964. Cerca con Google

[36] E. M. Glaser. Separation of neuronal activity by waveform analysis. Ad-vances in Biomedical Engineering, 1:77–136, 1971. Cerca con Google

[37] E. M. Glaser and W. B. Marks. On-line separation of interleaved pulsesequences. Data Acquisition and Processing in Biology and Medicine, 5:137–156, 1968. Cerca con Google

[38] C. Gold, D. A. Henze, C. Koch, and G. Buzs´aki. On the origin of theextracellular action potential waveform: A modeling study. Journal ofNeurophysiology, 95(5):3113–3128, February 2006. Cerca con Google

[39] C. M. Gray, P. E. Maldonado, M. Wilson, and B. McNaughton. Tetrodesmarkedly improve the reliability and yield of multiple single-unit isolationfrom multi-unit recordings in cat striate cortex. Journal of NeuroscienceMethods, 63(1):43–54, December 1995. Cerca con Google

[40] K. D. Harris, D. A. Henze, J. Csicsvari, H. Hirase, and G. Buzs´aki. Accuracyof tetrode spike separation as determined by simultaneous intracellular andextracellular measurements. Journal of Neurophysiology, 84(1):401–414,July 2000. Cerca con Google

[41] E. Hulata, R. Segev, and E. Ben-Jacob. A method for spike sorting and detectionbased on wavelet packets and shannon’s mutual information. Jour-nal of Neuroscience Methods, 117(1):1–12, May 2002. Cerca con Google

[42] M. S. Jog, C. I. Connolly, Y. Kubota, D. R. Iyengar, L. Garrido, R. Harlan,and Graybiel A. M. Tetrode technology: advances in implantable hardware,neuroimaging, and data analysis techniques. Journal of Neuroscience Meth-ods, 117(2):141–152, June 2002. Cerca con Google

[43] K. H. Kim and S. J. Kim. A wavelet-based method for action potential detectionfrom extracellular neural signal recording with low signal-to-noiseratio. IEEE Transactions on Biomedical Engineering, 50(8):999–1011, August2003. Cerca con Google

[44] T. Kohonen. The self-organizing map. Proceedings of the IEEE, 78(9):1464–1480, 1990. Cerca con Google

[45] M. Laubach. Wavelet-based processing of neuronal spike trains prior todiscriminant analysis. Journal of Neuroscience Methods, 134(2):159–168,April 204. Cerca con Google

[46] J. C. Letelier and P. P. Weber. Spike sorting based on discrete wavelettransform coefficients. Journal of Neuroscience Methods, 101(2):93–106,September 2000. Cerca con Google

[47] M. S. Lewicki. Bayesian modeling and classification of neural signals. NeuralComputation, 6(5):1005–1030, September 1994. Cerca con Google

[48] M. S. Lewicki. A review of methods for spike sorting: the detection andclassification of neural action potentials. Computation in Neural Systems,9(4):R53–R78, November 1998. Cerca con Google

[49] S. G. Mallat. Multiresolution approximations and wavelet orthonormalbases of l2(r). Transactions of the American Mathematical Society,315(1):69–87, September 1989. Cerca con Google

[50] S. G. Mallat. A theory for multiresolution signal decomposition: the waveletrepresentation. IEEE Transactions on Pattern Analysis and MachineIntelligence, 11(7):674–693, July 1989. Cerca con Google

[51] S. G. Mallat. A Wavelet Tour of Signal Processing. Academic Press, SanDiego, 1999. Cerca con Google

[52] D.A. McCormick, Connors B.W., Lighthall J.W., and D.A. Prince. Comparativeelectrophysiology of pyramidal and sparsely spiny stellate neuronsof the neocortex. Journal of Neurophysiology, 54(4):782–806, 1985. Cerca con Google

[53] B. L. McNaughton, J. O’Keefe, and C. A. Barnes. The stereotrode: anew technique for simultaneous isolation of several single units in the centralnervous system from multiple unit records. Journal of NeuroscienceMethods, 8(4):391–397, August 1983. Cerca con Google

[54] Y. Meyer. Ondelettes et Op´erateurs. Hermann, Paris France, 1990. Cerca con Google

[55] D. J. Mishelevich. On-line real-time digital computer separation of extracellularneuroelectric signals. IEEE Transactions on Biomedical Engineering,17(2):147–150, April 1970. Cerca con Google

[56] Z. Nenadic and J. W. Burdick. Spike detection using the continuous wavelettransform. IEEE Transaction on Biomedical Engineering, 52(1):74–87,January 2005. Cerca con Google

[57] K. G. Oweiss and D. J. Anderson. Spike sorting: A novel shift and amplitudeinvariant technique. Neurocomputing, 44–46:1133–1139, July 2002. Cerca con Google

[58] K. G. Oweiss and D. J. Anderson. Spike superposition resolution in multichannelextracellular neural recordings: a novel approach. In ConferenceProceedings. First International IEEE EMBS Conference on Neural Engi-neering, 2003, pages 130–133, March 2003. Cerca con Google

[59] K. G. Oweiss and D. J. Anderson. Tracking signal subspace invariance forblind separation and classification of nonorthogonal sources in correlatednoise. EURASIP Journal on Advances in Signal Processing, 2007. Cerca con Google

[60] A. Pavlov, V. A. Makarov, I. Makarova, and F. Panetsos. Sorting of neuralspikes: When wavelet based methods outperform principal componentanalysis. Natural Computing, 6(3):269–281, September 2007. Cerca con Google

[61] C. Pouzat. SpikeOMatic Tutorial 1. http://www.biomedicale.univparis5.fr/ physcerv/ C Pouzat/ Code folder/ SpikeOMaticTutorial1.pdf,July 2005. Vai! Cerca con Google

[62] C. Pouzat, O. Mazor, and G. Laurent. Using noise signature to optimizespike-sorting and to assess neuronal classification quality. Journal of Neu-roscience Methods, 122(1):43–57, December 2002. Cerca con Google

[63] V. J. Prochazka and H. H. Kornhuber. On-line multi-unit sorting withresolution of superposition potentials. Electroencephalography and clinicalneurophysiology, 34(1):91–93, January 1973. Cerca con Google

[64] R. Quian Quiroga, Z. Nadasdy, and Y. Ben-Shaul. Unsupervised spike detectionand sorting with wavelets and superparamagnetic clustering. NeuralComputation, 16(8):1661–1687, August 2004. Cerca con Google

[65] W. Rall. Electrophysiology of a dendritic neuron model. Biophysical Jour-nal, 2:145–167, March 1962. Cerca con Google

[66] S. Rebrik, S. Tzonev, and K. D. Miller. Analysis of tetrode recordings incat visual system. In CNS ’97: Proceedings of the sixth annual conferenceon Computational Neuroscience: trends in research, 1998, pages 491–496,New York, NY, USA, 1998. Plenum Press. Cerca con Google

[67] D. Rinberg, W. Bialek, H. Davidowitz, and N. Tishby. Spike sorting inthe frequency domain with overlap detection. arXiv:physics/0306056, June2003. Cerca con Google

[68] D. Rinberg, H. Davidowitz, and N. Tishby. Multi-electrode spike sortingby clustering transfer functions. In Proceedings of the 1998 conferenceon Advances in Neural Information Processing Systems II, pages 146–152,Cambridge, MA, USA, 1999. MIT Press. Cerca con Google

[69] M. Sahani. Latent variable models for neural data analysis. PhD thesis,Pasadena, CA, USA, 1999. Cerca con Google

[70] T. Sato, T. Suzuki, and K. Mabuchi. Fast automatic template matchingfor spike sorting based on davies-bouldin validation indices. Engineering inMedicine and Biology Society, 2007. 29th Annual International Conferenceof the IEEE, pages 3200–3203, August 2007. Cerca con Google

[71] E. M. Schmidt. An instrument for separation of multiple unit neuro-electricsignals. IEEE Transactions on Biomedical Engineering, 18(2):155–157,March 1971. Cerca con Google

[72] E. M. Schmidt. Computer separation of multi-unit neuroelectric data: areview. Journal of Neuroscience Methods, 12(2):95–111, December 1984. Cerca con Google

[73] W. Simon. The real time sorting of neuro-electric action potentials in multipleunit studies. Electroencephalography and Clinical Neurophysiology,18:192–195, February 1965. Cerca con Google

[74] S. Takahashi, Y. Anzai, and Y. Sakurai. Automatic sorting for multineuronalactivity recorded with tetrodes in the presence of overlappingspikes. Journal of Neurophysiology, 89(4):2245–2258, April 2003. Cerca con Google

[75] M. Vetterli and J. Kovacevic. Wavelets and Subband Coding. Prentice-Hall,Englewood Cliffs (NJ), 1995. Cerca con Google

[76] B. C. Wheeler and W. J. Heetderks. A comparison of techniques for classificationof multiple neural signals. IEEE Transactions on Biomedical En-gineering, 29(12):752–759, December 1982. Cerca con Google

[77] X. Yang and S. A. Shamma. A totally automated system for the detectionand classification of neural spikes. IEEE Transaction on BiomedicalEngineering, 35(10):806–816, October 1988. Cerca con Google

[78] P.-M. Zhang, J.-Y. Wu, Y. Zhou, P.-J. Liang, and J.-Q. Yuan. Spike sortingbased on automatic template reconstruction with a partial solution to theoverlapping problem. Journal of Neuroscience Methods, 135(1–2):55–65,May 2004. Cerca con Google

[79] G. Zouridakis and D. C. Tam. Multi-unit spike discrimination using wavelettransforms. Computers in Biology and Medicine, 27(1):9–18, January 1997. Cerca con Google

Download statistics

Solo per lo Staff dell Archivio: Modifica questo record