WIP: Steady State Visually Evoked Potential Signal Analyzer (SSVEP)

This repository is dedicated to the development of an advanced signal processing tool for analyzing Steady State Visually Evoked Potentials (SSVEP). SSVEP are brain responses induced by visual stimuli at specific flickering frequencies, widely used in BCI systems. Our tool aims to provide robust analysis capabilities to enhance the interpretation and application of SSVEP in neuroscientific research and practical applications. http://biosigpro.xyz/overview.html#dataset

Project Status

This project is currently in a work-in-progress stage. We are continuously refining our algorithms and improving the user interface.

Features

• Signal Processing: Implementation of various signal processing techniques to clean and extract meaningful data from raw SSVEP signals.
• CCA-Based Analysis: Utilizes Canonical Correlation Analysis to match SSVEP signals against known templates, enhancing the accuracy of frequency recognition.
• Filter Bank Approach: Incorporates a filter bank to analyze multiple frequency bands, enabling more precise frequency detection and better handling of signal artifacts.
• Cross-Platform Scripts: Includes both MATLAB and Python scripts to facilitate analysis, accommodating users with different programming preferences.

Dataset

The project utilizes a comprehensive SSVEP dataset, which has been converted into the .h5 format to improve compatibility and performance in Python environments:

• Description: EEG recordings from subjects exposed to LED stimuli at varied frequencies. The data is now available in .h5 format, allowing for efficient handling and processing in Python, alongside traditional MATLAB scripts.
• Access: The package provides the dataset.
• Download: You can download data from http://biosigpro.xyz/overview.html#dataset

Pipeline and Its Stages Overview

• Data Ingestion: Collecting and loading data from sources (modifiable) into the system.

• Data Validation: Ensuring the quality and consistency of the ingested data by applying validation rules.

• Data Transformation: Preprocessing the data, including feature engineering and scaling, to make it suitable for model training.

• Model Training: Training machine learning models on the transformed data to create a predictive model.

• Model Evaluation: Assessing the performance of the trained model using different evaluation metrics.

• Model Deployment: Deploying the trained model on a cloud platform, such as AWS (Amazon Web Services), to be accessed and used by the public or end-users via APIs or web services.

Getting Started

To begin using this project, clone the repository and follow the installation instructions below. The list of packages will become available.

git clone [email protected]:Biomeical-Signal-Processing/WIP-SSVEPA.git
cd WIP-SSVEPA
pip install -r requirements.txt

Demonstration of EEG signal analysis using Canonical Correlation Analysis

Canonical Correlation Analysis Matrix

Canonical Correlation Sample A1

Canonical Correlation Sample A1

Workflows that is followed for each integration

1. Update config.yaml
2. Update schema.yaml
3. Update params.yaml
4. Update the entity
5. Update the configuration manager in src config
6. Update the components
7. Update the pipeline
8. Update the main.py
9. Update the app.py