Abstract:

Magnetic resonance imaging (MRI) and Computed Tomography (CT) are diagnostic imaging modalities used heavily to characterize brain disease. Artificial intelligence (AI), specifically deep learning, is a type of machine learning that involves training a computer to recognize patterns in data by processing large amounts of information through multiple layers of neural networks. Deep learning algorithms are well suited to image analysis tasks and have been applied to a wide range of medical imaging applications. To date, the Food and Drug Administration (FDA) has approved over 500 AI solutions as ‘medical devices’ that aid in the clinical decision making.

The presentation will focus on examples where AI deep learning models can be used to synthesize images that provide physiological information, for example cerebral blood flow maps from arterial spin labeling MRI, and cardiac-related brain pulsatility maps from blood oxygenation level dependent (BOLD) contrast images. We will also showcase the academic AI tools developed at the Computational Radiology and AI (CRAI.no) unit at the Oslo University Hospital, Norway. From among these tools, we will describe the development of a CT-based segmentation tool that contours all forms of intracranial hemorrhage from non-contrast CT images.

You can attend via Zoom: https://umich.zoom.us/j/93856074220
passcode: 847305