Deep Learning-Based AI for Glioma and Cerebrovascular Segmentation in MRI

Precise segmentation of gliomas and cerebral blood vessels in MRI is critical for safe and effective neurosurgical planning, as well as for accurate diagnosis and monitoring in patients with brain tumors. This PhD project focuses on developing and refining deep learning models to accurately delineate both tumor tissue and vascular structures in multiparametric MRI scans.The research will involve the design and optimization of advanced neural network architectures using modern frameworks such as MONAI and PyTorch, with an emphasis on clinical applicability. Publicly available datasets, including BraTS and UPenn-GBM, will be used to train and validate the models to ensure their reliability and generalizability across different patient populations.In addition to algorithm development, the project will explore the spatial relationships between gliomas and surrounding vasculature to inform surgical strategies and reduce intraoperative risks. Advanced preprocessing techniques and data augmentation methods will be employed to address challenges such as data variability and limited sample sizes.This work aims to translate deep learning innovation into practical tools that support neurosurgeons in achieving greater precision during tumor resection, ultimately contributing to safer procedures and better outcomes for patients.

• We are looking for a highly motivated, self-driven team player with:

• A strong background in computer science, biomedical engineering, or a closely related discipline.

• Familiarity with a range of deep neural network architectures, as well as emerging models such as LLMs and VLMs

• Proficiency in Python and hands-on experience with deep learning frameworks such as PyTorch or TensorFlow.

• Familiarity with medical imaging modalities, particularly MRI.

• Strong analytical and problem-solving skills and a passion for interdisciplinary research at the intersection of AI and healthcare.

This project is eligible for funding but is subject to eligibility criteria & funding availability.

Page expires: 28th April 2026