Difference between revisions of "2017 Winter Project Week/Deep Learning for Medical Image Computation"

From NAMIC Wiki
Jump to: navigation, search
 
(10 intermediate revisions by 3 users not shown)
Line 7: Line 7:
 
==Key Investigators==
 
==Key Investigators==
 
*Tina Kapur, BWH/HMS
 
*Tina Kapur, BWH/HMS
 +
*Steve Pieper, Isomics
 
*Abdul Al Halabi, nVIDIA
 
*Abdul Al Halabi, nVIDIA
 
*Abel Brown, nVIDIA
 
*Abel Brown, nVIDIA
Line 26: Line 27:
 
<!-- Approach and Plan bullet points -->
 
<!-- Approach and Plan bullet points -->
 
* 11 Projects that plan to use/investigate Deep Learning on Medical Images + clinical data
 
* 11 Projects that plan to use/investigate Deep Learning on Medical Images + clinical data
* Hands-on Tutorial on using GPUs for Deep Learning (Abel Brown, Abdul Alhalabi) (Monday 3-5pm, 5:30-7pm)
+
* Deep Learning hands-on tutorial: Image segmentation lab with TensorFlow (Abel Brown, Abdul Alhalabi) (Monday 3-5pm)
 +
* Deep Learning hands-on tutorial: Electronic health records lab with Keras (Abel Brown, Abdul Alhalabi) (Monday 5:30-7pm)
 
* Newly Launched MGH Clinical Data Science Center with focus on GPU and Machine Learning (Joshua Moore) (Monday 5-5:30pm)
 
* Newly Launched MGH Clinical Data Science Center with focus on GPU and Machine Learning (Joshua Moore) (Monday 5-5:30pm)
 
*Basic Concepts in Neural Network Methodology (Mohsen Ghafoorian) (Tuesday 1-3pm)
 
*Basic Concepts in Neural Network Methodology (Mohsen Ghafoorian) (Tuesday 1-3pm)
 
*Tensorflow Workshop (Luis Ibanez) (Wednesday 10am-12pm)
 
*Tensorflow Workshop (Luis Ibanez) (Wednesday 10am-12pm)
*Methodology Continued: Unet, Segmentation (Mohsen Ghafoorian) (Wednesday 3-3:45pm)
 
*DeepInfer Extension Manager for  an Open Source Repository for Deep Learning networks (Alireza Mehrtash) (Wednesday 3:45-4pm)
 
 
|
 
|
 
<!-- Progress and Next steps bullet points (fill out at the end of project week) -->
 
<!-- Progress and Next steps bullet points (fill out at the end of project week) -->
*
+
* All sessions were well attended (30-100 attendees). Specific feedback: google tensorflow tutorial should have been the first one in the series
 +
* MGH Center for Clinical Data Science presented and loaned high performance machines for the week
 +
* Successful use of prostate segmentation code across multiple institutions
 
|}
 
|}
  
 
==Background and References==
 
==Background and References==
 
<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->
 
<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->
 +
*[[File:Basic_concepts.pptx]] Presentation on basic concepts in neural networks (By Mohsen Ghafoorian)
 +
*[[File:State_of_the_art_CNNs.pptx]] Presentation on State-of-the-art CNN architectures (By Mohsen Ghafoorian)

Latest revision as of 13:05, 13 January 2017

Home < 2017 Winter Project Week < Deep Learning for Medical Image Computation

Key Investigators

  • Tina Kapur, BWH/HMS
  • Steve Pieper, Isomics
  • Abdul Al Halabi, nVIDIA
  • Abel Brown, nVIDIA
  • Joshua Moore, MGH Clinical Data Science Center
  • Mohsen Ghafoorian, Radboud University
  • Luis Ibanez, Google
  • Alireza Mehrtash, UBC/BWH

Project Description

Objective Approach and Plan Progress and Next Steps
  • To provide an introduction to proven concepts from Deep Learning that are making high impact in medical image computation.
  • 11 Projects that plan to use/investigate Deep Learning on Medical Images + clinical data
  • Deep Learning hands-on tutorial: Image segmentation lab with TensorFlow (Abel Brown, Abdul Alhalabi) (Monday 3-5pm)
  • Deep Learning hands-on tutorial: Electronic health records lab with Keras (Abel Brown, Abdul Alhalabi) (Monday 5:30-7pm)
  • Newly Launched MGH Clinical Data Science Center with focus on GPU and Machine Learning (Joshua Moore) (Monday 5-5:30pm)
  • Basic Concepts in Neural Network Methodology (Mohsen Ghafoorian) (Tuesday 1-3pm)
  • Tensorflow Workshop (Luis Ibanez) (Wednesday 10am-12pm)
  • All sessions were well attended (30-100 attendees). Specific feedback: google tensorflow tutorial should have been the first one in the series
  • MGH Center for Clinical Data Science presented and loaned high performance machines for the week
  • Successful use of prostate segmentation code across multiple institutions

Background and References