Trevor,
This may help at least indirectly. Alternatively, if regions in which
abnormalites
are found are discrete and well defined you might consider
a two-sample proportional odds model with clustering within subject
[working independence model with cluster sandwich or cluster
bootstrap intracluster correlation correction.]
That way the lowest value of the ordinal scale would be
especially meaningful.
-Frank Harrell
One possi
@ARTICLE{alb97gen,
author = {Albert, Paul S. and Follmann, Dean A. and Barnhart, Huiman
X.},
year = 1997,
title = {A generalized estimating equation approach for modeling
random
length binary vector data},
journal = Biometrics,
volume = 53,
pages = {1116-1124},
annote = {correlated binary data;GEE;shared parameter models;joint
modeling
of severity and frequency;compound endpoints}
}
> Trevor Hastie wrote:
>
> I am looking for some Splus or R code to do a FROC analysis, in
> particular
> a comparison of the diagnostic ability of two groups. From what I have
> been able to glean from the couple
> of papers that my radioligist consultee has given me to read, FROC
> stands for "Free Response ROC",
> and it arises in radiology where the screener may find none, one, or
> more than one
> abnormality, and rates each of them. This distinguishes it from the
> case where ROC
> analysis is popular, where the unit of measurement is the X-ray or
> image.
>
> I am open to other approaches as well.
> Thanks
>
> --------------------------------------------------------------------
> Trevor Hastie hastie@stat.stanford.edu
> Professor, Department of Statistics, Stanford University
> Phone: (650) 725-2231 (Statistics) Fax: (650) 725-8977
> (650) 498-5233 (Biostatistics) Fax: (650) 725-6951
> URL: http://www-stat.stanford.edu/~hastie
> address: room 104, Department of Statistics, Sequoia Hall
> 390 Serra Mall, Stanford University, CA 94305-4065
> --------------------------------------------------------------------
--
Frank E Harrell Jr Prof. of Biostatistics & Statistics
Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences
U. Virginia School of Medicine http://hesweb1.med.virginia.edu/biostat
|