Data Scientist

4/2020 – 1/2024

PhD candidate specializing in developing and evaluating individualized treatment assignment strategies using deep learning algorithms in TensorFlow and Keras, including MLPs, GANs, Variational Autoencoders, and Graph Neural Networks (GNNs). Worked on a project assigning pain patients to three treatment groups using K-NN clustering and XGBoost, validated via an RCT. Collected data with REDCap and visualized patient characteristics in an R Shiny web app via the REDCap API. Post-RCT, proposed an enhanced treatment assignment algorithm incorporating model uncertainty using a novel GNN approach. Developed a project applying prototype learning and decision trees for explainable classification of pathology patches in PyTorch, enhancing model interpretability. Applied digital twins model using stable diffusion to real patient data for privacy protection in Pytorch. Trained a model to predict dropout of cancer patients using longitudinal data reaching 80% in precision score in Tensorflow.

Machine Learning, Computer Vision, Tensorflow, Keras, Pytorch, Data Science

10/2018 – 4/2019

Development and evaluation of a computer vision algorithm for long-term video prediction using neural networks in PyTorch.

Computer Vision, Pytorch

10/2014 – 7/2015

Created a computer program that was able, with the help of several web cameras, to scan information from a helicopter panel and then insert it into a table.

Computerprogrammierung