Project Overview
“Unlocking the patterns hidden within neural data requires a seamless fusion of domain expertise and powerful computational tools.”
MULTISAB2 is a specialized desktop and web-based platform built to accelerate research in cognitive neuroscience. Neuroscientists often work with massive, noisy EEG datasets that are difficult to manage and analyze. This project was created to provide them with an integrated environment to filter, process, and extract meaningful insights from their data.
My role was to architect the core processing engine, developing the signal processing algorithms and implementing machine learning models to classify brain states and identify significant neural events automatically.
Key Capabilities
🧠 Advanced Signal Processing Pipeline
I built a multi-stage pipeline for EEG data pre-processing, including temporal/spatial filtering, artifact removal (like eye-blinks), and feature extraction (e.g., power spectral density, event-related potentials).
🤖 Machine Learning for Pattern Recognition
The platform integrates custom machine learning models to classify cognitive states, detect anomalies, and predict subject responses based on EEG patterns, achieving high accuracy on benchmark datasets.
📈 Interactive Data Visualization
Researchers can explore the processed data through interactive charts and topographic maps of brain activity, allowing for intuitive discovery of patterns and validation of results.
🔄 Batch Processing & Reproducibility
The system was designed to run analyses on large cohorts of data in batches and to save all processing parameters, ensuring that every step of the research is fully documented and reproducible.
Technologies Used
- Core Language: Python
- Signal Processing & ML: MNE-Python, Scikit-learn, PyTorch, Pandas, NumPy
- Data Visualization: Matplotlib, Seaborn, Plotly/Dash for interactive web dashboards
- Application Framework: Custom Python backend with a streamlined user interface
- Workflow: Jupyter Notebooks for research & prototyping, packaged Python scripts for the core application