Is this a continuous marker?

AUC shows the predictive ability of the marker, the higher the AUC the more predictive the marker. In your case AUC~0.8 and this is not bad. To find the cutoff there are different methods. One of them is to use the point where the maximum of sensitivity plus specificity is obtained (Youden index), beyond other clinical considerations. To find this point out, you need to use the last table you have attached. Calculate sensitivity+specificity values for different cutoffs (as given by SPSS) and choose the one where this sum is maximized.

After finding that cut-off point, consider those with values bigger (or less) than this cut-off as diseased and others as healthy. Based on this classification, you calculate sensitivity, specificity (Sens & Spes are already given), PPV, NPV and accuracy for that cut-off point.

A possible criterion to obtain the optimal cut-off point is the Youden's J statistic: https://en.wikipedia.org/wiki/Youden's_J_statistic

Dear WA Gadalla,

Here is a summary acout ROC area.

ROC Area [1] - as described in Bradley (1997) and Fawcett (2006), ROC curve shows the tradeoff between sensitivity (true positive rate) and specificity (true negative rate), any increase in sensitivity will be accompanied by a decrease in specificity. The closer the curve follows the left-hand border and then the top border of the ROC space, the more accurate the test and the closer the curve comes to the 45-degree diagonal of the ROC space, the less accurate the test.

Accuracy is measured by the area under the ROC curve. An area of 1 represents a perfect test; an area of .5 represents a worthless test. A rough guide for classifying the accuracy of a diagnostic test is the traditional academic point system:

.90-1 = excellent (A) .80-.90 = good (B)  70-.80 = fair (C) 60-.70 = poor (D)                                       50-.60 = fail (F)

ROC Area is oftern used to evaluate the performance of ML methods.

Bradley, A. P. (1997). The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern recognition, 30(7), 1145-1159.‏

Fawcett, T. (2006). An introduction to ROC analysis. Pattern recognition letters,27(8), 861-874.‏‏

[1] All data shown was taken from http://gim.unmc.edu/dxtests website.

Best regards,

Yaakov

Classification Assessment Methods

This paper introduces a detailed explanation with numerical examples many classification assessment methods or classification measures such as:

Accuracy, sensitivity, specificity, ROC curve, Precision-Recall curve, AUC score and many other metrics. In this paper, many details about the ROC curve, PR curve, and Detection Error Trade-off (DET) curve. Moreover, many details about some measures which are suitable for imbalanced data are explained.

Your comments are highly appreciated.

https://ac.els-cdn.com/S2210832718301546/1-s2.0-S2210832718301546-main.pdf?_tid=20353999-8c84-422a-8191-fd1059469272&acdnat=1536185805_03b19e159f0675a01f35512961973eb5

Presentation of the paper is here

https://www.researchgate.net/publication/327403649_Classification_assessment_methods_a_detailed_tutorial