Wednesday, September 12, 2007

Tutorial document on image annotation

I just returned from the APIII2007 meeting in Pittsburgh, PA, where I gave a workshop on image annotation with RDF (concentrating on pathology images).

Bill Moore and I prepared a detailed white paper on the subject, which is available now at:

http://www.julesberman.info/spec2img.htm

The white paper is distibuted under the GNU Free Documentation License and can be copied, distributed, and cross-posted.

Table of Contents



Implementing an RDF Schema for Pathology Images

* Title
* How to read this document
* Conveying textual information with photomicrographs
* The Laboratory Digital Imaging Project (2004-2007)
* Increasing levels of complexity
o Level 1 - Free-text description inserted into image header
o Level 2 - Dublin Core descriptors inserted into image header
o Level 3 - Insert RDF document into image file
o Level 4 - Insert your image into RDF document
o Level 5 - Point to your image file from an RDF document.
o Level 6 - Combining of 1 through 5, with multiple images/annotations

How to do it
* How to acquire Perl and/or Ruby
* How to view a jpeg image
o How to view a jpeg image in Perl
o How to view a jpeg image in Ruby
+ Preliminaries
+ Displaying a JPEG image in Ruby
+ Manipulating an image in Ruby
* How to do Level 1 - Free-text description inserted into image header
o Perl
+ Preliminaries
+ Perl example
o Ruby
+ Ruby examples
# Inserting character strings into a JPEG header
# Inserting a text file into a JPEG header
* How to do Level 2 - Dublin Core descriptors inserted into image header
o Dublin core metadata is a special kind of text.
* How to do Level 3 - Insert RDF document into image file
o Introduction to RDF
+ Background
+ The definition of Meaning
+ Introduction to RDF syntax: RDF triples
* How to do Level 4 - Insert your image into an RDF document
o Example in Perl
o Base64 in Ruby
o Nota bene
* How to do Level 5 - Point to your image file from an RDF document.
* How to do Level 6 - Combining of 1 through 5, with multiple images/annotations
o Multiple RDF documents pointing to image file
o Multiple RDF files pointing to multiple image files

Dealing with DICOM

* What is DICOM?
* Standards are inter-convertible
* The DICOM header
* Manipulating DICOM with Ruby
o DICOM To JPEG Conversion - ezDicom And DCM2JPG And JPEG2DCM.
o Ruby scripts for converting DICOM to JPEG
o Extracting And Inserting Jpeg And Dicom Image Headers.
* RDF image specification contrasted with DICOM
* Porting between an RDF specification and a data standard

Advanced RDF tutorial (Can be skipped unless you have a special interest in subject)

* Common Data Elements (CDEs)
o ISO-11179 Specification for CDEs
* RDF Schemas
* A template for CDEs that can be transformed to RDF Schema elements
* Classes and Properties
* Creating Properties
* Specifying a datatype from within an RDF Schema Property Element
o Creating the external XSD document to datatype our property ranges
* The differences between Classes and Properties
* Creating instances of classes
* Preserving namespaces for Classes and Properties
* Unique subject identifiers
* LSID Examples:
* The RDF Schema for pathology images
* Data Specifications contrasted with Data Standards
* Creating your own RDF Schema
* Using an RDF Schema in your RDF documents
* A few publicly available RDF schemas
* Validating a specification
* Owl and DAML
* Problems with RDF
* Summary of advanced RDF section

-Jules Berman
My book, Principles of Big Data: Preparing, Sharing, and Analyzing Complex Information was published in 2013 by Morgan Kaufmann.



I urge you to explore my book. Google books has prepared a generous preview of the book contents. If you like the book, please request your librarian to purchase a copy of this book for your library or reading room.

tags: big data, metadata, data preparation, data analytics, data repurposing, datamining, data mining