JSM 2005 Online Program

Abstract #304778

This is the preliminary program for the 2005 Joint Statistical Meetings in Minneapolis, Minnesota. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2005); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

The Program has labeled the meeting rooms with "letters" preceding the name of the room, designating in which facility the room is located:

Minneapolis Convention Center = “MCC” Hilton Minneapolis Hotel = “H” Hyatt Regency Minneapolis = “HY”

Back to main JSM 2005 Program page

Legend:

= Applied Session,

= Theme Session,

= Presenter

Activity Number:	147
Type:	Contributed
Date/Time:	Monday, August 8, 2005 : 10:30 AM to 12:20 PM
Sponsor:	Section on Health Policy Statistics
Abstract - #304778
Title:	Integration of Computed Patient Anatomic and Case Data with Clinical Insight: An Ontology for "Case Type," Combining Computed Statistics with Clinician-provided Text and Diagnoses
Author(s):	Christopher Overton and Michael Zakharevich*+
Companies:	Align Technology and Align Technology
Address:	881 Martin Ave, Santa Clara, CA, 95050,
Keywords:	ontology ; clustering ; clinical outcomes ; text mining ; hierarchical modeling ; diagnosis
Abstract:	Recently popular clinical ontology (classifications) efforts seem not to integrate smoothly with modern statistical clustering. We discuss efforts at Align Technology to superimpose conceptual ontology structures from dental clinicians with computed, nearly hierarchical Bayesian networks. Our clusters (i.e., patients, doctors) should have low variance and thus good predictive behavior and intrinsic conceptual meaning, facilitating cluster assignment early in treatment and providing clinical explanatory power for doctors and internal technologists. In addition, we require flexibility to rebalance the hierarchy or "grow it downward" into common areas of uncertainty. Underlying data sources include features computed from 3D patient scans, doctor diagnoses and instructions, internal technician communications, and derived predictors. Our classification system combines a frontend using the Protégé ontology editor with a custom database backend linking data sources. We discuss early missteps and later successes in building the ontology in accordance with statistical classification and the value of coarser vs. finer classifications to probabilistic modeling.

The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2005 program