Computer-aided Glioblastoma and Intracranial Aneurysms Treatment Response Quantification in Neuroradiology

Purpose and goal:

Improved diagnosis and treatment can reduce mortality in glioblastoma and intracranial aneurysms. Manual quantification of treatment follow-up of glioblastoma and intracranial aneurysms is very time consuming. This project aims to develop methods based on artificial intelligence in neuroradiology, with a focus on methods for creating reference segmentations as well as allowing and improving quantification in treatment follow-up of glioblastoma and intracranial aneurysms.

Expected results and effects: A better change detection and quantification tool in neuroimaging will * lead to more precise and accurate diagnosis of brain cancers * permit diagnosis at an earlier stage, which makes it possible to start the treatment earlier

* lead to improved quantification of treatment response, which enables earlier adjustment of therapy when needed, in turn improving outcome. Approach and implementation:

The project is based on the following work packages, which will be carried out in collaboration between the PIs in Sweden and India. * Work Package (WP) 1: Project Management * WP 2: Ground Truth Creation Tool * WP 3: High Variability Image Data Processing * WP 4: Implementation and User Interface (UI) Development

* Clinical Adaptation Project 1: Deep learning based interactive segmentation and volumetric quantification of brain tumor * Clinical Adaptation Project 2: Improve and automate quantification of intracranial aneurysm remnants