File(s) under permanent embargo
Spike sorting using hidden markov models
conference contributionposted on 2013-01-01, 00:00 authored by Hailing Zhou, Shady MohamedShady Mohamed, Asim BhattiAsim Bhatti, Chee Peng LimChee Peng Lim, Nong Gu, Sherif Haggag, Saeid NahavandiSaeid Nahavandi
In this paper, hidden Markov models (HMM) is studied for spike sorting. We notice that HMM state sequences have capability to represent spikes precisely and concisely. We build a HMM for spikes, where HMM states respect spike significant shape variations. Four shape variations are introduced: silence, going up, going down and peak. They constitute every spike with an underlying probabilistic dependence that is modelled by HMM. Based on this representation, spikes sorting becomes a classification problem of compact HMM state sequences. In addition, we enhance the method by defining HMM on extracted Cepstrum features, which improves the accuracy of spike sorting. Simulation results demonstrate the effectiveness of the proposed method as well as the efficiency.