Skip to content
View AnastasChoudra's full-sized avatar
1 follower · 7 following
  • Reading

Block or report AnastasChoudra

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
anastaschoudra/README.md

👋 About Me

🌐 Portfolio Website: anastaschoudra.com

🧠 Background

Psychology, Neuroscience, and a growing folder of side-projects that mix the two with Python. I like turning raw biological data into something a model can chew on—mostly for fun, always for learning.

Key Projects

  • Heart Failure Prediction: Built a neural network classifier using TensorFlow/Keras to predict patient survival based on clinical and demographic features.
  • Classifying X-Rays for COVID with Neural Networks – CNN ensemble over 2-D chest images.
  • Bone-Marrow Transplant (Children) – Survival Classification – logistic pipeline with tuned PCA, 76 % acc on 187 paediatric cases.
  • WHO Life Expectancy Analysis: Developed a scikit-learn pipeline to predict life expectancy using global health data.

    🎓 MSc Project

    My MSc thesis "Exploring Metacognitive Sensitivity through a 3-AFC Visual Motion Discrimination Task" : investigates how well individuals can evaluate their own cognitive processes. This research explores the relationship between confidence ratings and actual performance in visual tasks, providing insights into conscious awareness and decision-making.

Goals

I’m working toward solid hands-on experience with brain–computer interfaces. That means:

  • Deepening my grasp of real-time EEG/MEG preprocessing, artifact rejection and feature extraction
  • Experimenting with different neural architectures (CNN, RNN, transformer) to decode oscillatory or evoked signals into clean control commands
  • Investigating how neural dynamics relate to perception and decision-making by combining computational modelling with electrophysiological signal analysis

Contact

If you’re into neuro-data, ping me on LinkedIn.

Pinned Loading

  1. 3-AFC-Random-Dot-Kinematic-Task 3-AFC-Random-Dot-Kinematic-Task Public

    3-AFC random-dot kinematogram task designed to quantify human metacognitive sensitivity. The script supports 9 coherence levels (0–100 %), adaptive Gabor masking, automatic contingency-table output…

    MATLAB

  2. Bone-Marow-Transplants Bone-Marow-Transplants Public

    A scikit-learn pipeline that predicts survival after pediatric bone-marrow transplant. It auto-cleans 37 clinical variables, splits numeric vs low-cardinality categoricals, scales, one-hot encodes,…

    Python

  3. Classifying-X-Rays-For-COVID-with-Neural-Networks Classifying-X-Rays-For-COVID-with-Neural-Networks Public

    A lightweight, beginner-friendly CNN that classifies chest X-rays into COVID-19, Normal or Viral Pneumonia using only TensorFlow / Keras and 17 M parameters. Perfect as a baseline benchmark or teac…

    Python

  4. heart-failure-survival-prediction heart-failure-survival-prediction Public

    This repo hosts a compact, end-to-end example of predicting heart-failure mortality with a Keras neural network. A 12-feature clinical dataset is cleaned, scaled and one-hot encoded, then fed into …

    Python

  5. Vigenere-Cipher Vigenere-Cipher Public

    A small, well-tested Python implementation of the classic Vigenère cipher. Provides an importable module and a vigenere CLI, preserves letter case, leaves non-letter characters unchanged, and valid…

    Python

  6. Predicting-EV-Charging-Loads-with-Neural-Networks Predicting-EV-Charging-Loads-with-Neural-Networks Public

    I use PyTorch to train a neural network to predict residential electric vehicle charging loads using real-world data from apartment buildings in Norway.

    Python