Area under ROC curve

Figure 5 shows that during query phase the accuracy is 95 %. The area under ROC curve indicates the accuracy. If the area is 1, then it represents the perfect test and an area of. 5 or less is useless test. Accuracy of the test can also be considered as point system like a range of. 90-1 as excellent,. 08-.09 as good,. 70-.80 as fair, etc. The closer the curve is to the left and top borders, the more it is accurate. A 45-degree diagonal is a random classifier if the curve is near diagonal, then the system is less accurate. The blue circle in Fig. 5 is the cut-off point which indicates that the accuracy of this lecture video retrieval system is 95 %. 

Schirrmeister et al. prospectively evaluated the clinical value of planar bone scans, SPECT and [ F]-labeled sodium fluoride in 53 patients with newly diagnosed lung cancer [193], Twelve of the 53 patients turned out to have bone metas-tases. [ F]-fluoride-PET detected all patients with bone metastases, whereas bone scan and SPECT produced false-negative results (6 vs. 1). An area under the curve analysis (ROC) proved p F]-fluoride-PET to be the most accurate whole-body imaging modality for screening of bone metastases in this study. 

Cl, confidence interval TIA, transient ischemic attack CT, computed tomography ECG, electrocardiography AUC-ROC, area under the curve of the received operator characteristic... 

It has been advocated that the area under the ROC curve is a relative measure of a tesfs performance. A Wilcoxon statistic (or equivalently the Mann-Whitney U-Test) statists cally determines which ROC curve has more area under it. Less computationally intensive alternatives, which are no longer necessary, have been described. These methods are particularly helpful when the curves do not intersect. When the ROC curves of two laboratory tests for the same disease intersect, they may offer quite different performances even though the areas under their curves are identical. The performance depends on the region of the curve (i.e., high sensitivity versus high specificity) chosen. Details on how to compare statistically individual points on two curves have been developed elsewhere. ... 

The aim of BPS-MS approach presented here can be summarized by comparing the simulated receiver operator characteristic (ROC) curves shown in Figure 1. Each additional dimension of analysis (a distinct pulse shape, that yields distinct mass spectrum) enlarges the area under the curve and therefore the confidence of the measurement. 

Table 6.5 summarizes the results for each outcome prediction variable for the evaluation step. Accuracies obtained using the ANN model was inferior to those using SVM, RF, and LDA. Based on McNemar s test, fc-NN performed the worst for predicting source of bleeding. LDA appeared to demonstrate a good overall performance with regards to sensitivity, specificity, PPV, NPV, and accuracy (Table 6.5). ROC (receiver operating characteristic) curves are calculated and areas under the curve (AUC) are compared for each of the models. An ROC curve is a...

Fig. 6.6. The graph on the left shows two ROC curves that have equal area under the curve (AUC), but the curves cross. One system is superior if high sensitivity is desirable, while the other system would be preferred for high specificity. On the right is an illustration of partial area under the ROC (pAUC), where it is assumed that high sensitivity is desirable. The shaded area divided by the area above the 0.9 value line is djAUC, the area under the ROC curve above a sensitivtty of 90%. When high sensitivity is desired, ,AUC shows the advantage for the blue curve, when compared with using AUC for the whole curve...

Receiver operating characteristic (ROC) curves for the two modalities, based on a continuous likelihood of malignancy scale for each finding from 0 to 100%, were not significantly different (p = 0.18). The trend in area under the curve favored film (Fig. 9.1). 

In addition, the Area Under the Curves (AUC) of the Receiver Operating Characteristic (ROC) curves were calculated as a quantitative measure for sensitivity and specificity of BIS to estimate the BAC. In a ROC curve, the True Positive Rate (TPR, sensitivity) is plotted against the False Positive Rate (FPR, 1-specificity). The TPR defines how many correct positive results occur among all positive samples and the FPR defines how many incorrect positive results occur among all negative samples. The AUC is then calculated as follows ... 

Table 1. Area under the curve (measurement point 3) for the ROCs. 

Dependence of the FN, FP, TN, and TP on the threshold value is elearly reflected by a Receiver Operator Characteristic (ROC) curve [48] in the TPR-FPR coordinates (tme positive rate versus false positive rate), where TRP=TP/(TP+TN) and FPR=FP/(FP+TN). The laiger the area under the curve (AUC), the higher is the classifier efficiency. 

In any case, this approach uses several additional assumptions. For this reason in the last time in ML the recognition accuracy criterion of the Area Under the ROC Curve AUC), which is free of additional assumptions, becomes very popular. Mathematically, AUC equals the... 

---