GTC ON-DEMAND

 
SEARCH SESSIONS
SEARCH SESSIONS

Search All
 
Refine Results:
 
Year(s)

SOCIAL MEDIA

EMAIL SUBSCRIPTION

 
 

GTC ON-DEMAND

Presentation
Media
Abstract:
Well present a variety of different computational paradigms to fuse information from microscopic images of tissue biopsies, corresponding genomic data, and patient and familial histories. Subjective clinical diagnosis is often based on multimodal information from microscopic and molecular information as well as data from patient and familial histories. But most recent work in objective pathology image analysis doesnt take into account additional information that can influence diagnosis or prognosis. Well demonstrate that fusing multimodal information significantly improves survival prediction, characterization, and prognostication. This paradigm can also be used to identify new biomarkers and morphological features that can lead to the development of new grading schemes.
Well present a variety of different computational paradigms to fuse information from microscopic images of tissue biopsies, corresponding genomic data, and patient and familial histories. Subjective clinical diagnosis is often based on multimodal information from microscopic and molecular information as well as data from patient and familial histories. But most recent work in objective pathology image analysis doesnt take into account additional information that can influence diagnosis or prognosis. Well demonstrate that fusing multimodal information significantly improves survival prediction, characterization, and prognostication. This paradigm can also be used to identify new biomarkers and morphological features that can lead to the development of new grading schemes.  Back
 
Topics:
Computer Vision, Healthcare and Life Sciences
Type:
Talk
Event:
GTC Washington D.C.
Year:
2019
Session ID:
DC91206
Download:
Share:
 
Abstract:
We'll talk about overcoming the barrier to obtaining adequate data to generate highly accurate models in medical imaging. Deep learning has proven valuable in creating analytic tools for medical imaging that have been cleared in multiple countries and are used in daily clinical work. But one barrier to additional solutions has been a lack of available data. Data are often difficult to obtain, especially in the quantities needed to generate highly accurate models. We'll discuss data augmentation techniques and explain how three different institutions have overcome this barrier.
We'll talk about overcoming the barrier to obtaining adequate data to generate highly accurate models in medical imaging. Deep learning has proven valuable in creating analytic tools for medical imaging that have been cleared in multiple countries and are used in daily clinical work. But one barrier to additional solutions has been a lack of available data. Data are often difficult to obtain, especially in the quantities needed to generate highly accurate models. We'll discuss data augmentation techniques and explain how three different institutions have overcome this barrier.  Back
 
Topics:
AI in Healthcare, Advanced AI Learning Techniques, Medical Imaging & Radiology
Type:
Talk
Event:
GTC Silicon Valley
Year:
2019
Session ID:
S9995
Streaming:
Download:
Share:
 
Abstract:
We demonstrate that labeled synthetic medical images can be used to train deep networks for accurate cancer diagnostics, specifically in applications where annotated data is limited due to privacy concerns, underrepresentation of rare conditions, limited availability of experts, etc. Deep networks trained on limited data also suffer from the cross patient network adaptability problem where networks trained on one patient often fail to generalize to other patients. We show that by using synthetically generated medical images, we can train accurate deep networks for cancer diagnostics in two different medical imaging applications: a) polyp detection and classification in endoscopy for colorectal cancer detection b) breast cancer grading of histopathology images.
We demonstrate that labeled synthetic medical images can be used to train deep networks for accurate cancer diagnostics, specifically in applications where annotated data is limited due to privacy concerns, underrepresentation of rare conditions, limited availability of experts, etc. Deep networks trained on limited data also suffer from the cross patient network adaptability problem where networks trained on one patient often fail to generalize to other patients. We show that by using synthetically generated medical images, we can train accurate deep networks for cancer diagnostics in two different medical imaging applications: a) polyp detection and classification in endoscopy for colorectal cancer detection b) breast cancer grading of histopathology images.   Back
 
Topics:
AI in Healthcare, Artificial Intelligence and Deep Learning
Type:
Talk
Event:
GTC Washington D.C.
Year:
2018
Session ID:
DC8150
Streaming:
Share:
 
 
Previous
  • Amazon Web Services
  • IBM
  • Cisco
  • Dell EMC
  • Hewlett Packard Enterprise
  • Inspur
  • Lenovo
  • SenseTime
  • Supermicro Computers
  • Synnex
  • Autodesk
  • HP
  • Linear Technology
  • MSI Computer Corp.
  • OPTIS
  • PNY
  • SK Hynix
  • vmware
  • Abaco Systems
  • Acceleware Ltd.
  • ASUSTeK COMPUTER INC
  • Cray Inc.
  • Exxact Corporation
  • Flanders - Belgium
  • Google Cloud
  • HTC VIVE
  • Liqid
  • MapD
  • Penguin Computing
  • SAP
  • Sugon
  • Twitter
Next