@inproceedings{ keshtkaran_unsupervised_2014,
  title = {Unsupervised Spike Sorting Based on Discriminative Subspace Learning},
  abstract = {Spike sorting is a fundamental preprocessing step
for many neuroscience studies which rely on the analysis of
spike trains. In this paper, we present two unsupervised spike
sorting algorithms based on discriminative subspace learning.
The first algorithm simultaneously learns the discriminative
feature subspace and performs clustering. It uses histogram
of features in the most discriminative projection to detect the
number of neurons. The second algorithm performs hierarchical
divisive clustering that learns a discriminative 1-dimensional
subspace for clustering in each level of the hierarchy until
achieving almost unimodal distribution in the subspace. The
algorithms are tested on synthetic and in-vivo data, and are
compared against two widely used spike sorting methods. The
comparative results demonstrate that our spike sorting methods
can achieve substantially higher accuracy in lower dimensional
feature space, and they are highly robust to noise. Moreover,
they provide significantly better cluster separability in the
learned subspace than in the subspace obtained by principal
component analysis or wavelet transform.},
  booktitle = {2014 Annual International Conference of the {IEEE} Engineering in Medicine and Biology Society ({EMBC})},
  author = {Keshtkaran, Mohammad Reza and Yang, Zhi},
  month = {August},
  year = {2014},
  file = {files/EMBC14_2224_FI (3).pdf:application/pdf},
  url_paper = {files/EMBC14_2224_FI (3).pdf}
}

{"_id":{"_str":"5405e1f168c1c0b80700049d"},"__v":0,"authorIDs":[],"author_short":["Keshtkaran","Reza, M.","Yang, Z."],"bibbaseid":"keshtkaran-reza-yang-unsupervisedspikesortingbasedondiscriminativesubspacelearning-2014","bibdata":{"abstract":"Spike sorting is a fundamental preprocessing step for many neuroscience studies which rely on the analysis of spike trains. In this paper, we present two unsupervised spike sorting algorithms based on discriminative subspace learning. The first algorithm simultaneously learns the discriminative feature subspace and performs clustering. It uses histogram of features in the most discriminative projection to detect the number of neurons. The second algorithm performs hierarchical divisive clustering that learns a discriminative 1-dimensional subspace for clustering in each level of the hierarchy until achieving almost unimodal distribution in the subspace. The algorithms are tested on synthetic and in-vivo data, and are compared against two widely used spike sorting methods. The comparative results demonstrate that our spike sorting methods can achieve substantially higher accuracy in lower dimensional feature space, and they are highly robust to noise. Moreover, they provide significantly better cluster separability in the learned subspace than in the subspace obtained by principal component analysis or wavelet transform.","author":["Keshtkaran","Reza, Mohammad","Yang, Zhi"],"author_short":["Keshtkaran","Reza, M.","Yang, Z."],"bibtex":"@inproceedings{ keshtkaran_unsupervised_2014,\n title = {Unsupervised Spike Sorting Based on Discriminative Subspace Learning},\n abstract = {Spike sorting is a fundamental preprocessing step\nfor many neuroscience studies which rely on the analysis of\nspike trains. In this paper, we present two unsupervised spike\nsorting algorithms based on discriminative subspace learning.\nThe first algorithm simultaneously learns the discriminative\nfeature subspace and performs clustering. It uses histogram\nof features in the most discriminative projection to detect the\nnumber of neurons. The second algorithm performs hierarchical\ndivisive clustering that learns a discriminative 1-dimensional\nsubspace for clustering in each level of the hierarchy until\nachieving almost unimodal distribution in the subspace. The\nalgorithms are tested on synthetic and in-vivo data, and are\ncompared against two widely used spike sorting methods. The\ncomparative results demonstrate that our spike sorting methods\ncan achieve substantially higher accuracy in lower dimensional\nfeature space, and they are highly robust to noise. Moreover,\nthey provide significantly better cluster separability in the\nlearned subspace than in the subspace obtained by principal\ncomponent analysis or wavelet transform.},\n booktitle = {2014 Annual International Conference of the {IEEE} Engineering in Medicine and Biology Society ({EMBC})},\n author = {Keshtkaran, Mohammad Reza and Yang, Zhi},\n month = {August},\n year = {2014},\n file = {files/EMBC14_2224_FI (3).pdf:application/pdf},\n url_paper = {files/EMBC14_2224_FI (3).pdf}\n}","bibtype":"inproceedings","booktitle":"2014 Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)","file":"files/EMBC14_2224_FI (3).pdf:application/pdf","id":"keshtkaran_unsupervised_2014","key":"keshtkaran_unsupervised_2014","month":"August","title":"Unsupervised Spike Sorting Based on Discriminative Subspace Learning","type":"inproceedings","url_paper":"files/EMBC14_2224_FI (3).pdf","year":"2014","bibbaseid":"keshtkaran-reza-yang-unsupervisedspikesortingbasedondiscriminativesubspacelearning-2014","role":"author","urls":{" paper":"http://keshtkaran.com/publication/files/EMBC14_2224_FI%20(3).pdf"},"downloads":3,"html":""},"bibtype":"inproceedings","biburl":"http://keshtkaran.com/publication/mypubs.bib","creationDate":"2014-09-02T15:27:45.885Z","downloads":3,"keywords":[],"search_terms":["unsupervised","spike","sorting","based","discriminative","subspace","learning","keshtkaran","reza","yang"],"title":"Unsupervised Spike Sorting Based on Discriminative Subspace Learning","year":2014,"dataSources":["wgKX5HhuqkGtWWkhk"]}