plotFROC: Draw FROC curves by two parameters a and b In BayesianFROC: FROC Analysis by Bayesian Approaches

Description

Plot FROC curves based on two parameters a and b.

Usage

 ```1 2 3 4 5 6 7 8``` ```plotFROC( a, b, mesh.for.drawing.curve = 10000, upper_x = 1, upper_y = 1, lower_y = 0 ) ```

Arguments

 `a` An arbitrary real number. It is no need to require any assumption, but I use such as `a`=μ/σ, where μ is a mean of signal distribution and σ is its standard deviation in the bi-normal assumption. `b` An arbitrary positive real number. I use such as `b`=1/σ, where σ is a standard deviation of signal distribution in the bi-noraml assumption. `mesh.for.drawing.curve` A positive large integer, indicating number of dots drawing the curves, Default =10000. `upper_x` A positive real number, indicating the frame size of drawing picture. `upper_y` A positive real number, indicating the frame size of drawing picture. `lower_y` A positive real number, indicating the frame size of drawing picture.

Details

FROC curve is the alternative notion of ROC curve in signal detection theory.

The definition of FROC curve is

(x(t),y(t) ) = (t, 1 - Φ( b* Φ^{-1}(exp(-t)) -a ) )

where, Φ() is the cumulative distribution function of the standard Gaussian distribution and Φ^{-1}() is its inverse mapping.

Revised 2019 NOv 27

Examples

 ```1 2 3``` ```dark_theme() plotFROC(0.1,0.2) ```

BayesianFROC documentation built on Jan. 13, 2021, 5:22 a.m.