
LVSegmenter: Left Ventricle Segmentation
2025Custom model for left ventricle segmentation in cardiac MRI. Based on a Kaggle-style challenge (IMA205: Apprentissage pour l'image et la reconnaissance d'objets).
Custom model for left ventricle segmentation in cardiac MRI. Based on a Kaggle-style challenge (IMA205: Apprentissage pour l'image et la reconnaissance d'objets).
Classifying paintings into figurative vs abstract using ResNet50 variants (single-branch, two-branch, dense fusion). Best model achieved ~90% macro F1-score.
Implementation of Gaussian-based ADSN models for inpainting microtextures. Evaluated parameters w and k_max across RGB, grayscale and complex textures; achieved realistic completions outperforming some existing methods.
Second prototype of a 'magic box' device for a card game aimed at cognitive development in children. Built with Arduino-based hardware (RFID, light, sound) and evaluated in classroom settings, with tools for data collection and analysis. Conducted as an academic extension project with Engineering and Psychology faculties.
Animatronic robotic face capable of imitating emotions and tracking humans with eyes and neck. Developed with Arduino Mega, ROS Noetic, PID control and face_recognition.