J Korean Med Sci.  2020 Jun;35(24):e171. 10.3346/jkms.2020.35.e171.

Drawing Guidelines for Receiver Operating Characteristic Curve in Preparation of Manuscripts

  • 1Department of Orthopedic Surgery, Seoul Sacred Heart General Hospital, Seoul, Korea


The appropriate plot effectively conveys the author's conclusions to the readers. The Journal of Korean Medical Science provides a series of special articles to show you how to make consistent and excellent plots easier. In the second article, drawing receiver operating characteristic (ROC) curve is introduced. A ROC curve is a graphic plot that illustrates the diagnostic ability as its discrimination threshold is varied. It is widely used as logistic regression analysis as machine learning becomes widespread. It has great visual effect in comparing various diagnostic tools.


Receiver Operating Characteristic Curve


  • Fig. 1 First tool for ROC. (A) Initial screen and example data. (B) One ROC plot. (C) Three ROC curves. (D) One ROC curve with CIs.ROC = receiver operating characteristic, AUC = area under curve, CI = confidence interval.

  • Fig. 2 Second tool. (A) Initial screen of classifier plot. (B) Lists of available curves. (C) Density plot.AUC = area under curve, CI = confidence interval.

  • Fig. 3 Third tool. (A) Initial screen of calibration plot. (B) Available thresholds options. (C) Select multiple threshold methods. (D) Various optimal thresholds result. (E) Various optimal thresholds represented on plot.ROC = receiver operating characteristic, AUC = area under curve.

  • Fig. 4 Fourth tool. (A) ROC from table or model. (B) Calculated specificity and sensitivity using data. (C) Select color of line and grid. (D) Data for logistic regression. (E) Result of logistic regression. (F) Diagnostic table as a result of logistic regression. (G) ROC curve from logistic regression.ROC = receiver operating characteristic.

  • Fig. 5 How to upload data and download plots. (A) How to download the plot. (B) How to upload your own data.ROC = receiver operating characteristic.


1. Kim J. Drawing guideline for JKMS manuscript (01) Kaplan-Meier curve and survival analysis. J Korean Med Sci. 2019; 34(8):e35. PMID: 30833878.
2. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988; 44(3):837–845. PMID: 3203132.
Full Text Links
  • JKMS
export Copy
  • Twitter
  • Facebook
Similar articles
Copyright © 2024 by Korean Association of Medical Journal Editors. All rights reserved.     E-mail: koreamed@kamje.or.kr