Share Email Print
cover

Proceedings Paper

The problem of ROC analysis without truth: the EM algorithm and the information matrix
Author(s): Sergey V. Beiden; Gregory Campbell; Kristen L. Meier; Robert F. Wagner
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Henkelman, Kay, and Bronskill (HKB) showed that although the problem of ROC analysis without truth is underconstrained and thus not uniquely solvable in one dimension (one diagnostic test), it is in principle solvable in two or more dimensions. However, they gave no analysis of the resulting uncertainties. The present work provides a maximum-likelihood solution using the EM (expectation-maximization) algorithm for the two- dimensional case. We also provide an analysis of uncertainties in terms of Monte Carlo simulations as well as estimates based on Fisher Information Matrices for the complete- and the missing-data problem. We find that the number of patients required for a given precision of estimate for the truth- unknown problem is a very large multiple of that required for the corresponding truth-known case.

Paper Details

Date Published: 14 April 2000
PDF: 9 pages
Proc. SPIE 3981, Medical Imaging 2000: Image Perception and Performance, (14 April 2000); doi: 10.1117/12.383099
Show Author Affiliations
Sergey V. Beiden, FDA Ctr. for Devices and Radiological Health (United States)
Gregory Campbell, FDA Ctr. for Devices and Radiological Health (United States)
Kristen L. Meier, FDA Ctr. for Devices and Radiological Health (United States)
Robert F. Wagner, FDA Ctr. for Devices and Radiological Health (United States)


Published in SPIE Proceedings Vol. 3981:
Medical Imaging 2000: Image Perception and Performance
Elizabeth A. Krupinski, Editor(s)

© SPIE. Terms of Use
Back to Top