From Nemo

Contents 1 Introduction 2 Tools 2.1 NemoErpPatternExtraction 2.1.1 NemoErpPatternDecomposition 2.1.2 NemoErpPatternSegmentation 2.1.3 NemoErpMetricExtraction 2.2 NemoErpClustering 2.3 NemoErpMetaAnalysis 3 Data Sets 3.1 Sim ERP 3.2 GAF-LP1 4 References

Introduction

The NEMO ERP Analysis Toolkit includes tools for EEG/ERP and MEG data decomposition, and ontology-based mark-up, annotation, and labeling of patterns in EEG and MEG data. These tools have been implemented in MATLAB by Robert Frank, a mathematician and data analyst for NEMO.

The NEMO analysis pipeline has been designed to support robust analysis and interpretation of EEG/MEG data within a lab, as well as cross-lab, cross-experiment meta-analysis. The current pipeline comprises four steps:

• Step 1: Decomposing ERP data (continuous spatiotemporal patterns are transformed into discrete patterns for analysis)
  • Data decomposition methods (PCA/ICA and related factor analytic methods)
  • Data windowing/segmentation methods (Microstate Analysis, a family of techniques for automated windowing)

• Step 2: Marking up the analysis results
  • Each pattern that is output from Step 1 is annotated with labels that relate the functional, spatial and temporal characteristics of the pattern to concepts in NEMO ontologies.
  • Ontologically structured data are then written to RDF and stored within a relational database, which we refer to as the NEMO ontology database.
  • RDF data can be queried in Virtuoso using a SPAQRL query engine.

• Step 3: Clustering the observed patterns within and across experiment datasets

• Step 4: Labeling the the new cross-experiment clusters using NEMO ontologies.

Each item in the above list is a step/phase in the processing pipeline and is associated with a set of MATLAB scripts. These script collections (aka tools) will work together to perform their assigned role and can be considered a single tool in the toolkit. The inputs and outputs of each tool will be compatible with those tools in the NEMO ERP Analysis Toolkit.

Tools

The Wiki download links have been deprecated as of 2010-11-02. Beginning 2010-11-03, all tools can be downloaded from the sourceforge repository. Bug and feature requests may be submitted to the sourceforge tracker.

The NEMO ERP Analysis Toolkit will be compatible with other MATLAB toolkits that are widely used in the EEG/MEG communities, including Makeig & Delorme's EEGLAB, Dien's ERP PCA Toolkit, Oostenveld's Fieldtrip, and (in future work) SPM8.

NemoErpPatternExtraction

NemoErpPatternDecomposition

NemoErpPatternDecomposition is a MATLAB class-based toolkit for decomposing Event Related Potential (ERP) data into ICA / PCA components for subsequent analysis.

Notes:

• The NEMO_ERP_Pattern_Decomposition class-based toolkit may be downloaded from the sourceforge repository.
• Latest release version: 2010-11-11
  • Includes EEGLAB version 9023b, EP Toolkit version 2.20, and FieldTrip version 20101014
  • Please see demonstration scripts in NEMO_ERP_Pattern_Decomposition/_Tutorial/DemoScript/EGIreader or EPreader for information on running the toolkit. The documentation is currently being revised.

NemoErpPatternSegmentation

NemoErpPatternSegmentation is a MATLAB class-based toolkit for segmenting Event Related Potential (ERP) data into microstates for subsequent analysis.

Notes:

• The NEMO_ERP_Pattern_Segmentation class-based toolkit may be downloaded from the sourceforge repository.
• Latest release version: 2010-11-13
  • Please see demonstration scripts in NEMO_ERP_Pattern_Segmentation/_Tutorial/DemoScript for information on running the toolkit. The documentation is currently being revised.

NemoErpMetricExtraction

NemoErpMetricExtraction is a MATLAB class-based toolkit for extracting a core suite of spatiotemporal metrics from the PCA / ICA / Microstate (under development) components of Event Related Potential data.

Notes:

• The NEMO_ERP_Metric_Extraction class-based toolkit may be downloaded from the sourceforge repository.
• Latest release version: 2010-11-11
  • Please see demonstration scripts in NEMO_ERP_Metric_Extraction/_Tutorial/DemoScript for information on running the toolkit. The documentation is currently being revised.

NemoErpClustering

COMING SOON!

NemoErpMetaAnalysis

COMING SOON!

Data Sets

Sim ERP

Simulated ERP data set, described in detail in internal report.

Download: Sim_ERP_Data.zip

GAF-LP1

Temporal PCA of 42 subject GAF-LP1 data set.

• Note: Please see ReadMe for PCA parameter settings.

Download: ReadMe

Download: GAF_LP1_tPCA_42subj_GrandAvg

Download: GAF_LP1_tPCA_Subj001_Subj010

Download: GAF_LP1_tPCA_Subj011_Subj020

Download: GAF_LP1_tPCA_Subj021_Subj030

Download: GAF_LP1_tPCA_Subj031_Subj040

Download: GAF_LP1_tPCA_Subj041_Subj042

References

Documentation on the NEMO ERP Analysis Toolkit is packaged with each [download] and may be found in the _Documentation subfolder for each NEMO ERP Analysis toolbox.

Literature references for each of the analysis techniques may be found on our References page.