Interactive Manipulation and Visualization of 3D Brain MRI for Surgical Training

Zichen Gui, Siddharth Jha, Benjamin Delbos, Richard Moreau, Rémi Chalard, Arnaud Lelevé, Irene Cheng

Abstract: In modern medical diagnostics, magnetic resonance imaging (MRI) is an important technique that provides insights into anatomical structures. This paper presents a comprehensive methodology focusing on streamlining the segmentation, reconstruction, and visualization of MRI data for surgical training. Segmentation involves the extraction of anatomical regions with the help of state-of-the-art deep learning algorithms. Then, 3D reconstruction converts segmented data from the previous step into multiple 3D representations. Finally, the visualization stage provides efficient and interactive 2D and 3D MRI data presentations. After integrating these three steps, the proposed system is found to augment the interpretability of the anatomical information from MRI scans. Compared with other medical platforms such as 3D Slicer, our specialized system features less user effort and greater extensibility. It was originally designed and implemented for haptic feedback simulation in surgical training, but we believe it can also be an effective tool for surgical planning, joint learning, and other purposes.