Construire une IUG universelle pour la gestion de l’hétérogénéité des formats d’imagerie cérébrale via conversion automatique et visualisation 3D intégrées

Details

Supervisors

Macq, Benoît ; Dricot, Laurence

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] en sciences informatiques, à finalité spécialisée

Abstract

enThe analysis of brain imaging data today relies on a multitude of specialized software tools. This diversity is accompanied by significant heterogeneity in the file formats used, which complicates the visualization, conversion, or exploitation of these data within standardized workflows. This thesis offers a technical solution to part of this problem by developing an interactive software tool that combines two essential features: three-dimensional visualization of brain imaging files (anatomical, functional, and diffusion magnetic resonance imaging) and automatic conversion between several common formats (TRK, TCK, FBR, VOI, VMR, NIfTI). Designed in Python with a modular and extensible approach, this tool is based on an intuitive graphical interface and leverages robust open-source libraries such as PyVista, DIPY, NumPy, PyQt6, and Nibabel. Particular attention was paid to the management of coordinate systems and transformation matrices to ensure spatial consistency of the data after conversion. The entire project is distributed under an open-source license and documented in a public repository, with the aim of serving as a bridge between heterogeneous tools without disrupting existing practices. This work thus aligns with an open science approach and aims to improve the accessibility, interoperability, and reusability of neuroimaging data.