IDOCS: intelligent distributed ontology consensus system--the use of machine learning in retinal drusen phenotyping

George Thomas; Michael A Grassi; John R Lee; Albert O Edwards; Michael B Gorin; Ronald Klein; Thomas L Casavant; Todd E Scheetz; Edwin M Stone; Andrew B Williams

doi:10.1167/iovs.06-1022

Back

IDOCS: intelligent distributed ontology consensus system--the use of machine learning in retinal drusen phenotyping

Journal article

Peer reviewed

IDOCS: intelligent distributed ontology consensus system--the use of machine learning in retinal drusen phenotyping

George Thomas, Michael A Grassi, John R Lee, Albert O Edwards, Michael B Gorin, Ronald Klein, Thomas L Casavant, Todd E Scheetz, Edwin M Stone and Andrew B Williams

Investigative ophthalmology & visual science, Vol.48(5), pp.2278-2284

05/2007

DOI: 10.1167/iovs.06-1022

PMID: 17460291

View Online

Abstract

To use the power of knowledge acquisition and machine learning in the development of a collaborative computer classification system based on the features of age-related macular degeneration (AMD). A vocabulary was acquired from four AMD experts who examined 100 ophthalmoscopic images. The vocabulary was analyzed, hierarchically structured, and incorporated into a collaborative computer classification system called IDOCS. Using this system, three of the experts examined images from a second set of digital images compiled from more than 1000 patients with AMD. Images were annotated, and features were identified and defined. Decision trees, a machine learning method, were trained on the data collected and used to extract patterns. Interrelationships between the data from the different clinicians were investigated. Six drusen classes in the structured vocabulary were largely sufficient to describe all the identified features. The decision trees classified the data with 76.86% to 88.5% accuracy and distilled patterns in the form of hierarchical trees composed of 5 to 15 nodes. Experts were largely consistent in their characterization of soft, and to a lesser extent, hard drusen, but diverge in definition of other drusen. Size and crystalline morphology were the main determinants of drusen type across all experts. Machine learning is a powerful tool for the characterization of disease phenotypes. The creation of a defined feature set for AMD will facilitate the development of an IDOCS-based classification system.

Algorithms

Computational Biology

Phenotype

User-Computer Interface

Macular Degeneration - classification

Vocabulary, Controlled

Artificial Intelligence

Humans

Middle Aged

Retinal Drusen - classification

Aged

Pattern Recognition, Automated - methods

Details

Title: Subtitle: IDOCS: intelligent distributed ontology consensus system--the use of machine learning in retinal drusen phenotyping
Creators: George Thomas
Michael A Grassi
John R Lee
Albert O Edwards
Michael B Gorin
Ronald Klein
Thomas L Casavant - University of Iowa, Center for Bioinformatics and Computational Biology
Todd E Scheetz - University of Iowa, Center for Bioinformatics and Computational Biology
Edwin M Stone - University of Iowa, Ophthalmology and Visual Sciences
Andrew B Williams - Spelman College
Resource Type: Journal article
Publication Details: Investigative ophthalmology & visual science, Vol.48(5), pp.2278-2284
DOI: 10.1167/iovs.06-1022
PMID: 17460291
NLM abbreviation: Invest Ophthalmol Vis Sci
ISSN: 0146-0404
eISSN: 1552-5783
Publisher: United States
Grant note: EY014467 / NEI NIH HHS 5 P20 MD000215 / NIMHD NIH HHS EY013688-03 / NEI NIH HHS
Language: English
Date published: 05/2007
Academic Unit: Roy J. Carver Department of Biomedical Engineering; Electrical and Computer Engineering; Iowa Neuroscience Institute; Ophthalmology and Visual Sciences
Record Identifier: 9983979986302771

Metrics

23 Record Views

4 Times Cited - Web of Science