Researchers: Chao Li, Stephen Price, Pan Liu, Carola-Bibiane Schönlieb, Yuxin Li(Fudan University)

Glioblastoma is the most common primary malignant tumour in adults, characterised by poor outcomes. Evidence shows that a higher extent of resection is beneficial to patient survival. However, extended resection or radiation may subject patients to higher risks of neurologic deficits. Therefore, accurately targeting tumour invasion is crucial for treatment planning. 

Here we propose to develop a mathematical model to predict glioblastoma invasion using advanced multi-parametric MRI. We will improve the current mathematical modelling by integrating machine learning with image inpainting. The clinical relevance of this approach will be investigated in glioblastoma patients. The potential results of this study would improve the delineation of tumour invasion that are currently based on conventional imaging. This study could advance the scenario of mathematical image analysis and have significant clinical impact on personalized surgical treatment of patients.