FACTS (Find the Appropriate Clinical Trials) for You: A Computer-Based Decision Support System for Breast Cancer Patients

Abstract

The Find Appropriate Clinical Trials (FACTS) system was redesigned to make it more accurate and compliant with existing standards. An explicit data model of patient eligibility for breast cancer clinical trials was developed, and served as the basis for encoding eligibility criteria. Standard vocabularies were utilized to represent concepts used in the system, and to retrieve their hierarchical relationships. The system now uses Bayesian networks to handle missing patient information. Protocols are presented to the user ranked by the likelihood that the patient is eligible for each one of them. As a result of a detailed data model, most of the eligibility criteria in 10 clinical trial protocols taken from the National Cancer Institute database were encoded and the performance of the system was compared to that of two oncologists. In a preliminary evaluation, there was a good agreement between the system's selection of clinical trials and those of two oncologists (kappa 0.86, 0.76). All cases in which the system's selection of a protocol did not agree with any of the physicians were analyzed, and the system's limitations were identified. The disagreement on ranking the protocols (kappa 0.24, 0.14) is discussed.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 2001
Accession Number
ADA392468

Entities

People

  • Lucila Ohno-machado

Organizations

  • Brigham and Women's Hospital

Tags

DTIC Thesaurus Topics

  • Bayesian Networks
  • Breast Cancer
  • Clinical Trials
  • Computers
  • Databases
  • Decision Support Systems
  • Diseases And Disorders
  • Graphical User Interface
  • Health Services
  • Heart Failure
  • Language
  • Medical Personnel
  • Models
  • Neoplasms
  • Physicians
  • Standards
  • User Interface

Readers

  • Regression Analysis.
  • Systems Analysis and Design
  • Women's Health and Cancer Risk Research: African American Women and Pregnancy Outcomes.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Neural Networks