N A Obuchowski1. 1. Cleveland Clinic Foundation, Department of Biostatistics, OH 44195-5196, USA.
Abstract
RATIONALE AND OBJECTIVES: Traditionally, multireader receiver operating characteristic (ROC) studies have used a "paired-case, paired-reader" design. The statistical power of such a design for inferences about the relative accuracies of the tests was assessed and compared with alternative designs. METHODS: The noncentrality parameter of an F statistic was used to compute power as a function of the reader and patient sample sizes and the variability and correlation between readings. RESULTS: For a fixed-power and Type I error rate, the traditional design reduces the number of verified cases required. A hybrid design, in which each reader interprets a different sample of patients, reduces the number of readers, total readings, and reading required per reader. The drawback is a substantial increase in the number of verified cases. CONCLUSION: The ultimate choice of study design depends on the nature of the tests being compared, limiting resources, a priori knowledge of the magnitude of the correlations and variability and logistic complexity.
RATIONALE AND OBJECTIVES: Traditionally, multireader receiver operating characteristic (ROC) studies have used a "paired-case, paired-reader" design. The statistical power of such a design for inferences about the relative accuracies of the tests was assessed and compared with alternative designs. METHODS: The noncentrality parameter of an F statistic was used to compute power as a function of the reader and patient sample sizes and the variability and correlation between readings. RESULTS: For a fixed-power and Type I error rate, the traditional design reduces the number of verified cases required. A hybrid design, in which each reader interprets a different sample of patients, reduces the number of readers, total readings, and reading required per reader. The drawback is a substantial increase in the number of verified cases. CONCLUSION: The ultimate choice of study design depends on the nature of the tests being compared, limiting resources, a priori knowledge of the magnitude of the correlations and variability and logistic complexity.
Authors: Brandon D Gallas; Heang-Ping Chan; Carl J D'Orsi; Lori E Dodd; Maryellen L Giger; David Gur; Elizabeth A Krupinski; Charles E Metz; Kyle J Myers; Nancy A Obuchowski; Berkman Sahiner; Alicia Y Toledano; Margarita L Zuley Journal: Acad Radiol Date: 2012-02-03 Impact factor: 3.173
Authors: Laurence Gaalaas; Donald Tyndall; André Mol; Eric T Everett; Ananta Bangdiwala Journal: Dentomaxillofac Radiol Date: 2015-12-16 Impact factor: 2.419
Authors: Robert M Nishikawa; Suddhasatta Acharyya; Constantine Gatsonis; Etta D Pisano; Elodia B Cole; Helga S Marques; Carl J D'Orsi; Dione M Farria; Kalpana M Kanal; Mary C Mahoney; Murray Rebner; Melinda J Staiger Journal: Radiology Date: 2009-04 Impact factor: 11.105
Authors: R Edward Hendrick; Elodia B Cole; Etta D Pisano; Suddhasatta Acharyya; Helga Marques; Michael A Cohen; Roberta A Jong; Gordon E Mawdsley; Kalpana M Kanal; Carl J D'Orsi; Murray Rebner; Constantine Gatsonis Journal: Radiology Date: 2008-04 Impact factor: 11.105