ROC Analysis        

John Eng, M.D.
The Russell H. Morgan Department of Radiology and Radiological Science
Johns Hopkins University, Baltimore, Maryland, USA

Web browser requirements for this page:  For Windows, Netscape Navigator 4.0 or Microsoft Internet Explorer 4.0 (later versions are preferred). For Macintosh, Netscape Navigator 4.0.x to 4.7.x. This page may not work with Microsoft Internet Explorer for the Macintosh or later versions of Netscape Navigator for the Macintosh, due to incomplete browser support for Internet standards. Please send any bugs, questions, comments, or suggestions to electronic mail will be answered.

Instructions:  To analyze your data, use the following steps to fill out the form below. This Web page contains JROCFIT, a Java program for calculating receiver operating characteristic (ROC) curves. JROCFIT is a direct translation of the ROCFIT program, the latter developed by Charles Metz and colleagues at the University of Chicago. JROCFIT duplicates all output of the ROCFIT program. See below for program translation details.

  1. Select the data format. (See explanation of data formats.)
  2. Paste or enter your data into the "Input Data" field or click the "Paste Example Data" button. Data may be pasted from programs such as Microsoft Excel or Word.
  3. Enter the number of rating categories. (See explanation of data formats.) This is not needed for the example data.
  4. Click the "Run Program" button.
Data Format:  Format 1   Format 2   Format 3

Number of Rating Categories:     (Not needed for example data.)

      

 Input Data:  (paste or enter)  Program Output:  (may be copied and pasted into other programs)

ROC Curve      Summary Statistics:      Points for Plotting:  (copy & paste into Excel)
 

ROC Curve Type:   Fitted   Empirical


Key for the ROC Plot
RED symbols and BLUE line:  The fitted ROC curve.
GRAY lines:  The 95% confidence interval of the fitted ROC curve.
BLACK symbols ± GREEN line:  The points making up the empirical ROC curve.
 

Details of the Translation:  The lavender box at the top of this page contains JROCFIT, a program for fitting receiver operating characteristic (ROC) curves using the maximum likelihood fit of a binormal model. JROCFIT is a literal Java translation of the Fortran source code for ROCFIT, a program developed and maintained by Dr. Charles Metz and colleagues at the University of Chicago. During translation, data input routines were added to JROCFIT in order to provide a more convenient user interface. The translation was performed with kind permission from Dr. Metz, but neither he nor his colleagues were otherwise involved in the translation, nor do they share any responsibility for the translation's accuracy. While the translation was performed as carefully as possible, the author assumes no responsibility for the accuracy of the program's operation, regardless of cause.

(Page last updated: 7/4/2002)