FEIT Research Project Database

Deep learning algorithms for stroke image analysis


Project Leader: Marimuthu Palaniswami
Staff: Aravinda Rao, Nandakishor Desai
Collaborators: Bernard Yan (Medicine, Dentistry and Health Sciences)
Primary Contact: Marimuthu Palaniswami (palani@unimelb.edu.au)
Keywords: computer vision; deep learning; magnetic resonance imaging MRI; neuroimaging
Disciplines: Electrical & Electronic Engineering
Domains:
Research Centre: ARC Research Network on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP)

One in six people worldwide experience a stroke in their lifetime. 15 million people suffer from stroke and 5.8 million die each year. An estimated 6.7 million deaths were due to stroke in 2015. Computed Tomography (CT) scans are used by clinicians to assess the extent of stroke and provide right treatment to save brain tissue. However, current neuroimaging techniques have variable diagnostics accuracies.

The project aims to develop new Deep Learning-based Artificial Intelligence (AI) tools to provide better precision for diagnosing stroke patients. The project would also utilise Magnetic Resonance Imaging (MRI) scans (as a gold standard) to validate the new models.