Welcome to my GitHub! I'm a Neural Data Scientist at Alljoined Inc. I am an expert in EEG preprocessing, neural decoding and identifying neural correlates and neural biomarkers of cognitive, emotional, and clinical processes.
I was until recently a PostDoctoral Fellow at Chapman University, specializing in data science with focus on computational neuroscience research. Originally from Brazil, I got my PhD with focus on Psychology with focus on Cognitive Neuroscience, but fell in love with Data Science and Computational Neuroscience. My expertise is in bringing actionable insights from complex multimodal data, and accelerating data processing, following rigorous scientific procedures and standards.
I am passionate about developing scripts to process and analyze EEG/EMG data using Matlab, R, and Python. My research focuses on:
- 🔭 Current Projects:
- Image Reconstruction from ERP
- Emotional Decoding from Images and Videos
- Triangulating the Neural Correlates of Consciousness
- Identifying Novel Neural Correlates of Consciousness in Inattentional Blindness
- Programming Languages: Matlab, R, Python, JavaScript, SQL
- Techniques: Machine Learning (SciKit-Learning), Deep Neural Networks (Keras-Tensorflow, Pytorch), Time-series analysis (Numpy/Pandas/Torch/MNE), Data plotting (Seaborn/Matplotlib)
- Project 1: Decoding EEG/ERP Across Tasks and Time - A moving-window cross-validation implementation for ML decoding and testing over time - pipeline to and compare multiple models' AUC, accuracy, f1-score (Jupyter Notebook - Python [Keras, TensorFlow, Scikit-Learn, XGboost, Numpy, Scipy, Pandas, Pickle, Matplotlib, Seaborn, MNE])
- Project 2: Using CNN to decode EEG - Created a 3D Convolutional Neural Network to decode brain signals over time (Jupyter Notebook - Python [Keras, TensorFlow, Numpy, Pandas, Pickle, Matplotlib, Seaborn, MNE])
- Project 3: QuickLab - A comprehensive toolkit for EEGLAB for processing and analyzing EEG data (MATLAB, [EEGLAB, Fieldtrip, Signal Processing Toolbox, Parallel Computing, Cuda-ICA])
- Project 4: XTTS read-outloud - A Chrome extension to train voice models using XTTS2 and read outloud any text on chrome. This is a fork, but I redesigned the whole interface (HTML, JavaScript, CSS, JSON)
- Project 5: Mental Imagery and Tracking Experiments - Created an 5 tasks for my Master's Thesis experiments, including multi-object tracking, mental rotation, and mouse-object traking (PHP, HTML, JavaScript, CSS, SQL)
- 😄 Pronouns: He/Him/His
- ⚡ Fun fact: I am from Brazil, but I love hockey!
Feel free to explore my repositories and reach out if you have any questions or collaboration ideas!