Deep Learning for Medical Imaging

Description

Deep learning is the field of machine learning that studies and develops artificial neural networks capable of learning several layers of representation (features) from raw data. These methods have delivered new levels of performance in the field of computer vision. More recently, they have become popular in medical imaging systems, such as for the segmentation of various types of tissues in medical imagery.

Scope

In this tutorial, we will provide an introduction to deep learning, covering both theory and practice. On the theory side, we will describe the most common concepts found in today’s deep learning research, with a focus on convolutional neural networks. On the practical side, we will describe how the TensorFlow library can be used to apply deep learning in medical imaging, covering both TensorFlow basics and specific use cases in medical imaging applications.

Schedule - September 10th

Morning (9:00 - 12:00) [Slides: part1, part2, part3]

• 9:00 - 10:00
  • Motivation - Why deep learning for medical imaging [15 mins]
  • What is a neural network [45 mins]
• 10:00 - 10:30
  • Conference coffee break
• 10:30 - 12:30
  • Training deep neural networks [50 mins]
  • Break [10 mins]
  • Convolutional neural networks [50 mins]
  • Future Challenges [10 mins]

Lunch break (12:30 - 14:30)

Afternoon (14:30 - 17:30) [Slides]

• 14:30 - 15:30
  • Introduction to TensorFlow and TensorBoard [60 mins]
• 15:30 - 16:00
  • Conference coffee break
• 16:00 - 17:30
  • Case study 1: Diabetic Retinopathy [40 mins]
  • Case study 2: Breast Cancer Metastasis [40 mins]
  • Wrap up [10 mins]

Organizers

• Hugo Larochelle, Google Brain
• Yun Liu, Google Brain
• Derek Wu, Google Brain

For more information on this tutorial, contact us at tutorial-deep@miccai2017.org .