Advancement and trends in medical image analysis using deep learning
IEEE EMBS is co-sponsoring a talk with UBC by Dr. Fausto Milletari, NVIDIA.
Abstract: In this talk, I will discuss advancements and trends in medical image analysis using deep learning. I will discuss V-Net and the impact of the fully convolutional neural network for 3D image segmentation and then highlight limitations and shortcoming of this class of algorithms. I will then present novel approaches and ongoing research aiming to integrate statistical shape prior into CNN-based predictions and our latest MICCAI contributions, termed Coarse to Fine Context Memory (CFCM) that aims to integrate features extracted at different depths through a memory mechanism based on RNNs. Finally, I will talk about TOMAAT, an open source package allowing easy deployment of research DL algorithms in the cloud.
Date and Time
Location
Hosts
Registration
- Date: 14 Jun 2018
- Time: 02:00 PM to 03:00 PM
- All times are (GMT-08:00) Canada/Pacific
-
Add Event to Calendar
- ICICIS
- 2366 Main Mall, University of British Columbia
- Vancouver, British Columbia
- Canada V6T 1Z4
- Room Number: Rm 288, ICICS Board Room, 2nd floor
- Contact Event Host
- Co-sponsored by University of British Columbia
Speakers
Dr. Milletari of NVIDIA
Advancement and trends in medical image analysis using deep learning
Abstract: In this talk, I will discuss advancements and trends in medical image analysis using deep learning. I will discuss V-Net and the impact of the fully convolutional neural network for 3D image segmentation and then highlight limitations and shortcoming of this class of algorithms. I will then present novel approaches and ongoing research aiming to integrate statistical shape prior into CNN-based predictions and our latest MICCAI contributions, termed Coarse to Fine Context Memory (CFCM) that aims to integrate features extracted at different depths through a memory mechanism based on RNNs. Finally, I will talk about TOMAAT, an open source package allowing easy deployment of research DL algorithms in the cloud.
Biography:
Biography: Fausto Milletari is a Senior Solutions Architect at NVIDIA and a Ph.D. graduate from the Technical University of Munich. Fausto's major research topic is the segmentation of medical images. He has pioneered the usage of 3D convolutional neural networks to segment MRI and Ultrasound volumetric data and other 3D deep-learning techniques. In addition, he works on a variety of other computer vision problems, such as object tracking and detection. His work focuses on pattern recognition and machine learning, and in particular on deep learning techniques. Several of his contributions have been presented in recent editions of MICCAI, IPCAI, BMVC, 3DV and in specialized journals. His current job at NVIDIA focuses on the application of artificial intelligence techniques in medicine at the boundary between research and commercial applications.
Address:NVIDIA,