作者: Weijie Chen , Adam Wunderlich , Nicholas Petrick , Brandon D. Gallas
关键词: Sizing 、 Statistical hypothesis testing 、 Data modeling 、 Monte Carlo method 、 Binary number 、 Algorithm 、 Type I and type II errors 、 State (computer science) 、 Binary data 、 Medicine
摘要: We treat multireader multicase (MRMC) reader studies for which a reader’s diagnostic assessment is converted to binary agreement (1: agree with the truth state, 0: disagree state). present mathematical model simulating MRMC data desired correlation structure across readers, cases, and two modalities, assuming expected probability of equal modalities (P1=P2). This can be used validate coverage probabilities 95% confidence intervals (of P1, P2, or P1−P2 when P1−P2=0), type I error superiority hypothesis test, size noninferiority test (which assumes P1=P2). To illustrate utility our simulation model, we adapt Obuchowski–Rockette–Hillis (ORH) method analysis data. Moreover, use ORH sizing in setting. Our software package publicly available on Google code project hosting site simulation, analysis, validation,