Machine Learning

Developed deep learning solutions using AFT-Net to correct motion artifacts in MRI scans. Created simulated motion-corrupted brain MRI data and trained models to reconstruct high quality, motion-free images directly from raw k-space data. Promising results improving structural similarity over 0.9 against ground truth. Deep learning on raw k-space shows potential for efficient, accurate MRI motion correction.

Developed and optimized machine learning models to predict RNA expression from chromatin accessibility data and protein levels from RNA expression in individual cells. Preprocessed large single-cell genomics datasets, implemented regression algorithms, and tuned hyperparameters. Achieved high accuracy on test data, demonstrating the feasibility of predicting across modalities. Overall, this project showcases expertise in genomics, predictive modeling, and working with big data.

With a group of 3 other students, we competed on Kaggle during 2 weeks with other teams from our school at EURECOM to analyse and classify tweets' sentiments into positive, neutral or negative labels.

With 3 other classmates we worked on the prediction of worldwide temperature starting with data from dry and mild temperate climate locations.

Following a Machine Learning course, I conducted with two classmates a project to predict ratings given by users to French red wines on the platform Vivino.com