In the article titled “Automatic Characterization of the Physiological Condition of the Carotid Artery in 2D Ultrasound Image Sequences Using Spatiotemporal and Spatiospectral 2D Maps” [1], there were errors in the sensitivity and specificity values and missing information on how they were derived from Figures , , and .
Information on how these values were obtained is explained as follows: True positive (TP) is the number of sick individuals who are correctly identified as sick. False positive (FP) is the number of healthy individuals incorrectly identified as sick. True negative (TN) is the number of healthy individuals correctly identified as healthy. False negative (FN) is the number of sick individuals incorrectly identified as healthy. Sensitivity = TP/(TP + FN) = TP/(total number of sick individuals). Specificity = TN/(TN + FP) = TN/(total number of healthy individuals).
The thresholds (the separation lines) in Figures and are calculated as the mean value of all sick and healthy individuals in the study. From Figure , the following values are obtained: FN = 15 TP = 69 − 15 = 54 FP = 12 TN = 47 − 12 = 35Then Sensitivity = TP/69 = 0.7826 Specificity = TN/47 = 0.7447
From Figure , the following values are obtained: FN = 5 TP = 69 − 5 = 64 FP = 15 TN = 47 − 15 = 32Then Sensitivity = TP/69 = 0.9275 Specificity = TN/47 = 0.6809
And when combining Figures and , the following values are obtained: FN = 5 TP = 69 − 5 = 64 FP = 12 TN = 47 − 12 = 35Then Sensitivity = TP/69 = 0.9275 Specificity = TN/47 = 0.7447
This combining rule is equivalent to a logical “OR.” In Figures and , there are 5 out of 69 pathological elderly cases that are misclassified simultaneously in both these figures whereas the remaining 64 cases appear correctly classified in at least one of the two figures. This implies that FN = 5 and TP = 69 − 5 = 64, so the combined sensitivity is equal to . On the other hand, there are 12 out of 47 healthy elderly patients who happen to be misclassified simultaneously in both these figures, which implies that and , so the combined specificity is equal to .
From the green curve in Figure , the following values are obtained: FN = 15 TP = 69 − 15 = 54 FP = 12 TN = 47 − 12 = 35Then Sensitivity = TP/69 = 0.783 Specificity = TN/47 = 0.745
From the blue curve in Figure , the following (better) values are obtained: FN = 10 TP = 69 − 10 = 59 FP = 10 TN = 47 − 10 = 37Then Sensitivity = TP/69 = 0.855 Specificity = TN/47 = 0.787
Therefore, the sensitivity and specificity values should be corrected in the following sections:(i)In Abstract, the sensitivity value in the sentence “Automatic differentiation, between cases of these three categories, was achieved with a sensitivity of 97.1% and a specificity of 74.5%.” should be corrected to “Automatic differentiation, between cases of these three categories, was achieved with a sensitivity of 92.8% and a specificity of 74.5%.”(ii)In the fifth paragraph of Results, the specificity value in the sentence “In other words, by considering the resulting feature values and the chosen classification thresholds presented in these figures, a sensitivity of 92.8% and a specificity of 76.6% can be obtained.” should be corrected to “In other words, by considering the resulting feature values and the chosen classification thresholds presented in these figures, a sensitivity of 92.8% and a specificity of 74.5% can be obtained.”(iii)In the last paragraph of Results, the sensitivity value in the sentence “Therefore, the results obtained in that study were optimal achieving 100% accuracy, compared to 82.8% when applying the approach considering the normalized frequencies in the current work.” should be corrected to “Therefore, the results obtained in that study were optimal achieving 100% accuracy, compared to 85.5% when applying the approach considering the normalized frequencies in the current work.”(iv)In the fifth paragraph of Conclusions, the sensitivity value in the sentence “The proposed method for automated evaluation of the homogeneity and the variation in the patterns of the spatiospectral maps resulted in a sensitivity of 97.1% and a specificity of 74.5%, compared to 85.5% and 78.7%, respectively, when employing the method proposed in [.” should be corrected to “The proposed method for automated evaluation of the homogeneity and the variation in the patterns of the spatiospectral maps resulted in a sensitivity of 92.8% and a specificity of 74.5%, compared to 85.5% and 78.7%, respectively, when employing the method proposed in [].”