[Eeglablist] Microstate Analysis

[Bathelt, Johannes]

Dear Pierre,
Thank you for these hints. However, the k-means clustering and the hierarchical clustering algorithms do not quite cut it. The atomize and agglomerate hierarchical clustering algorithm is a modified version of hierarchical clustering that takes into account that the EEG samples are not independent, i.e. maps that are closer together in time are also more similar. I tried to reproduce the clustering from Cartool with k-means and hierarchical clustering in Matlab and did not obtain similar results. So, I would need an implementation of the AAHC algorithm specifically. 

Yours

Joe

____________________________________
Joe Bathelt, MSc
PhD candidate
Developmental Cognitive Neuroscience Unit
UCL Institute of Child Health
+44 020 7905 42749
johannes.bathelt.10 at ucl.ac.uk

On 24 Jul 2013, at 20:07, "Megevand, Pierre" <PMegevand at NSHS.edu> wrote:

> Hi Joe,
> 
> The Statistics toolbox of MATLAB comes with functions for cluster analysis, including both hierarchical clustering and also k-means clustering (I gather you're referring to the Cartool program from Geneva University, which includes these 2 types of clustering). You can have a look at this introduction on the MATLAB website: http://www.mathworks.com/help/stats/introduction-to-cluster-analysis.html, or look for "cluster analysis" in the MATLAB documentation.
> 
> Relevant functions for hierarchical clustering include: 
> pdist
> linkage
> cluster
> clusterdata (combines the 3 previous ones)
> 
> I don't know that some has come up with wrappers for these functions that cater specifically to EEG data.
> 
> I hope that this helps.
> 
> Pierre
> --
> Pierre Mégevand, MD, PhD
> Post-doctoral research fellow
> Laboratory for multimodal human brain mapping
> Hofstra North Shore-LIJ School of Medicine
> Manhasset, NY, USA
> ________________________________________
> From: Johannes Bathelt [johannes.bathelt.10 at ucl.ac.uk]
> Sent: Wednesday, July 24, 2013 6:08 AM
> To: eeglablist at sccn.ucsd.edu
> Subject: [Eeglablist] Microstate Analysis
> 
> Dear List,
> I would like to compare EEG microstate analysis with some other methods. Does anyone know if there is an implementation in EEGLAB/Matlab based on 'atomize and agglomerate hierarchical clustering'? There is a windows programme by the authors of the main paper about it, but it makes it rather difficult to transfer the data back and forth between Matlab and this. Thank you for your help!
> 
> Yours
> 
> Joe
> 
> 
> 
> __________________________________________________________
> Joe Bathelt, MSc
> Ph.D. candidate
> 
> Contact Information
> T:  +44 (0) 20 7905 2749
> E: johannes.bathelt.10 at ucl.ac.uk<mailto:johannes.bathelt.10 at ucl.ac.uk>
> A:    Developmental Cognitive Neuroscience Unit
> UCL Institute of Child Health
>  30 Guilford Street
> London WC1N 1EH
> 
> 
> Great Ormond Street Hospital - Developmental Vision Service<http://www.gosh.nhs.uk/medical-conditions/clinical-specialties/neurodisability-information-for-parents-and-visitors/clinics-and-services/developmental-vision-clinic/>
> Centre for Developmental Cognitive Neuroscience <https://www.ucl.ac.uk/cdcn/ourresearch/projects/socio-cognitiveskillsrecruitment>
> 
> 
> The information contained in this electronic e-mail transmission and any attachments are intended only for the use of the individual or entity to whom or to which it is addressed, and may contain information that is privileged, confidential and exempt from disclosure under applicable law. If the reader of this communication is not the intended recipient, or the employee or agent responsible for delivering this communication to the intended recipient, you are hereby notified that any dissemination, distribution, copying or disclosure of this communication and any attachment is strictly prohibited. If you have received this transmission in error, please notify the sender immediately by telephone and electronic mail, and delete the original communication and any attachment from any computer, server or other electronic recording or storage device or medium. Receipt by anyone other than the intended recipient is not a waiver of any attorney-client, physician-patient or other privilege.
>