Receiver operating characteristic curve applied in screening for transfusion-transmitted virus
Yin Yonghu
Abstract
Yin Yonghu
Abstract
Objective To promote ROC curve and to evaluate the decided thresholds and diagnostic performance in TTVs screening in blood station. Methods Enzyme-linked immunosorbent assay was used to detect antibody for hepatitis C virus( anti-HCV) in 56 human plasma( positive 34,negative 22). The cutoff value and accuracy between two diagnostic tests via the ROC curve were determined. Results The best cut point of manufacturer A was 2. 459,with values of sensitivity and specificity 0. 73 and 0. 97,respectively. The area under ROC curve was 0. 9082,which suggested a higher level of diagnostic performance. The test accuracy of manufacturer B was superior to A. Conclusion An optimal statistical index of sensitivity,specificity and Youden value could be calculated by ROC curve. ROC curve leads to more objective results and can evaluate the accuracy and compare between two diagnostic tests in a more reliable way than other indices.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Objective To promote ROC curve and to evaluate the decided thresholds and diagnostic performance in TTVs screening in blood station. Methods Enzyme-linked immunosorbent assay was used to detect antibody for hepatitis C virus( anti-HCV) in 56 human plasma( positive 34,negative 22). The cutoff value and accuracy between two diagnostic tests via the ROC curve were determined. Results The best cut point of manufacturer A was 2. 459,with values of sensitivity and specificity 0. 73 and 0. 97,respectively. The area under ROC curve was 0. 9082,which suggested a higher level of diagnostic performance. The test accuracy of manufacturer B was superior to A. Conclusion An optimal statistical index of sensitivity,specificity and Youden value could be calculated by ROC curve. ROC curve leads to more objective results and can evaluate the accuracy and compare between two diagnostic tests in a more reliable way than other indices.
Key concepts: Receiver operating characteristic, Youden's J statistic, Cutoff, Sensitivity (control systems), Area under the curve, Diagnostic accuracy, Area under curve, Medicine