Nancy A Obuchowski1, Stephen L Hillis. 1. Department of Quantitative Health Sciences, Cleveland Clinic Foundation, OH 44195, USA. obuchon@ccf.org
Abstract
OBJECTIVE: Calculating the sample size for a multireader, multicase study of readers' diagnostic accuracy is complicated. Studies in which patients can have multiple findings, as is common in many computer-aided detection (CAD) studies, are particularly challenging to design. MATERIALS AND METHODS: We modified existing methods for sample size estimation for multireader, multicase studies to accommodate multiple findings on the same case. We use data from two large multireader, multicase CAD studies as ballpark estimates of parameter values. RESULTS: Sample size tables are presented to provide an estimate of the number of patients and readers required for a multireader, multicase study with multiple findings per case; these estimates may be conservative for many CAD studies. Two figures can be used to adjust the number of readers when there is some data on the between-reader variability. CONCLUSION: The sample size tables are useful in determining whether a proposed study is feasible with the available resources; however, it is important that investigators compute sample size for their particular study using any available pilot data.
OBJECTIVE: Calculating the sample size for a multireader, multicase study of readers' diagnostic accuracy is complicated. Studies in which patients can have multiple findings, as is common in many computer-aided detection (CAD) studies, are particularly challenging to design. MATERIALS AND METHODS: We modified existing methods for sample size estimation for multireader, multicase studies to accommodate multiple findings on the same case. We use data from two large multireader, multicase CAD studies as ballpark estimates of parameter values. RESULTS: Sample size tables are presented to provide an estimate of the number of patients and readers required for a multireader, multicase study with multiple findings per case; these estimates may be conservative for many CAD studies. Two figures can be used to adjust the number of readers when there is some data on the between-reader variability. CONCLUSION: The sample size tables are useful in determining whether a proposed study is feasible with the available resources; however, it is important that investigators compute sample size for their particular study using any available pilot data.
Authors: Abraham H Dachman; Nancy A Obuchowski; Jeffrey W Hoffmeister; J Louis Hinshaw; Michael I Frew; Thomas C Winter; Robert L Van Uitert; Senthil Periaswamy; Ronald M Summers; Bruce J Hillman Journal: Radiology Date: 2010-07-27 Impact factor: 11.105
Authors: Lori E Dodd; Robert F Wagner; Samuel G Armato; Michael F McNitt-Gray; Sergey Beiden; Heang-Ping Chan; David Gur; Geoffrey McLennan; Charles E Metz; Nicholas Petrick; Berkman Sahiner; Jim Sayre Journal: Acad Radiol Date: 2004-04 Impact factor: 3.173
Authors: Nicholas Petrick; Berkman Sahiner; Samuel G Armato; Alberto Bert; Loredana Correale; Silvia Delsanto; Matthew T Freedman; David Fryd; David Gur; Lubomir Hadjiiski; Zhimin Huo; Yulei Jiang; Lia Morra; Sophie Paquerault; Vikas Raykar; Frank Samuelson; Ronald M Summers; Georgia Tourassi; Hiroyuki Yoshida; Bin Zheng; Chuan Zhou; Heang-Ping Chan Journal: Med Phys Date: 2013-08 Impact factor: 4.071
Authors: Steven J Skates; Michael A Gillette; Joshua LaBaer; Steven A Carr; Leigh Anderson; Daniel C Liebler; David Ransohoff; Nader Rifai; Marina Kondratovich; Živana Težak; Elizabeth Mansfield; Ann L Oberg; Ian Wright; Grady Barnes; Mitchell Gail; Mehdi Mesri; Christopher R Kinsinger; Henry Rodriguez; Emily S Boja Journal: J Proteome Res Date: 2013-10-28 Impact factor: 4.466
Authors: Darren Boone; Susan Mallett; Justine McQuillan; Stuart A Taylor; Douglas G Altman; Steve Halligan Journal: PLoS One Date: 2015-09-10 Impact factor: 3.240