Joint Mining of Biological Text and Images: Case Studies

Posted in Science on July 31, 2008

Joint Mining of Biological Text and Images: Case Studies

Lecture slides:

  • Systems Biology and Location Proteomics
  • Automated Interpretation
  • Supervised Learning of High-Resolution Subcellular Location Patterns
  • The Challenge
  • Feature-based, Supervised learning approach
  • Boland & Murphy 2001
  • Machine Learning Methods
  • Evaluating Classifiers
  • 2D Classification Results
  • Human Classification Results
  • Computer vs. Human
  • 3D HeLa cell images
  • Unsupervised Learning to Identify High-Resolution Protein Patterns
  • Location Proteomics
  • What Now?
  • Chen et al 2003; Chen and Murphy 2005
  • slide18
  • Nucleolar Proteins
  • Punctate Nuclear Proteins
  • Predominantly Nuclear Proteins with Some Punctate Cytoplasmic Staining
  • Nuclear and Cytoplasmic Proteins with Some Punctate Staining
  • Image Content-based Retrieval and Interpretation of Micrographs from On-line Journal Articles
  • Objectives of SLIF
  • Example page from biomedical literature
  • SLIF components
  • Overview: Image processing in SLIF
  • Overview: Image processing in SLIF01
  • Overview: Text Processing in SLIF Find entity names in text, and panel labels in text and the image. Match panels labels in text to panel labels on the image. Associate entity names to textual pane
  • Some major challenges
  • Graphical model: Panel typing
  • SLIF programmatic interface
  • Acknowledgments
  • Brian Athey (UMich), CMU: Bob Murphy

Author: Robert Murphy, Carnegie Mellon University

Watch Video

Tags: Science, Lectures, Computer Science, VideoLectures.Net, Bioinformatics, Text Mining, Image Analysis