Intelligence based improvement of personalized treatment and clinical workflow support (IMPACT)

Info graphics multi-modal machine learning

In this VINNOVA / ITEA3 / AIDA project, the goal is to improve detection, segmentation and treatment of small brain tumours, such as metastases, by combining novel diffusion imaging with deep learning. The project involves collaboration with Elekta, Inovia and SyntheticMR.

Brain tumours severely affect the quality of life for a large number of people. Brain tumours can be treated in different ways; using radiation therapy, chemotherapy, and surgery. In this project, the focus is on radiation therapy, and the ultimate goal is to treat small brain tumours before they grow to large tumours.

Existing approaches for detecting and segmenting brain tumours use 2D or 3D deep learning (convolutional neural networks), using different types of structural magnetic resonance imaging (MRI), for example T1 weighted images, T2 weighted images and T1 weighted images with gadolinium contrast. In this project, we will also take advantage of advanced diffusion imaging techniques featuring general gradient waveforms to further increase our information about the underlying micro structures. Brain tumour patients will be scanned at CMIV (Center for Medical Image Science and Visualization), the brain tumours will be annotated by an expert, and different CNNs will then be trained for automatic detection and segmentation.

Project website on

Research groupShow/Hide content