I'm Ricky, a Data Scientist from Ireland 
I'm currently a PhD student in the Empenn team in Inria, Rennes. I am working on techniques to create robust methods for the automatic segmentation of multiple sclerosis of lesions in spinal cord MRI. I am particularly interested in cases of "domain shift", i.e., where the distributions of data differ between training and testing sets, e.g., if we use different MRI scanners or different MRI sequences.
The repositories below contain the code for several projects during my masters (2020-2022).
shortcuts-skin-cancer is a project in Explainable AI, exploring how bias in data affects the predictions of an AI model trained to detect skin cancer from images of skin lesions. The corresponding paper has been published in MDPI Diagnostics.
detect-af concerns the automatic detection of Atrial Fibrillation episodes using Electrocardiogram (ECG) data. Relevant statistical features from the literature were identified & applied, and were used as inputs to decision tree and XGBoost models.
IdMind delves into EEG biometrics. Using the brainwaves recorded by an electro-encephalogram (EEG), a Convolutional Neural Network is trained to distinguish between the brain patterns of 20 individuals. The idea is to enable authentication using the signals from our brain, which are more difficult to spoof (vs. fingerprint or iris scans).
Track-Facial-Features-3D is a small project as part of an initial image processing course. Given videos of participants carrying out a set of tongue exercises, the objective is to track the movements of their tongue, nose and eyes from three different angles, and pinpoint the locations in 3D space.