Fig. 3: ExoVita pancreas performance in the independent validation set. | Communications Medicine

Fig. 3: ExoVita pancreas performance in the independent validation set.

From: Development of a blood-based extracellular vesicle classifier for detection of early-stage pancreatic ductal adenocarcinoma

Fig. 3: ExoVita pancreas performance in the independent validation set.

a ExoVita Score distribution across the case (n = 30) and control (n = 83) cohorts in the validation set. The box plots represent the median and interquartile range with each dot overlayed representing a patient in the validation set. The whiskers represent the min and max of the distribution. The dotted line represents the threshold selected to obtain 91% specificity from the training set. The receiver operating characteristic (ROC) curve analysis for the validation set is shown on the right. The 95% CI on the Area Under the Curve (AUC) is determined by bootstrapping in a total of 2000 repetitions. b Comparison of performance for the training and validation cohorts broken down by stage using the ExoVita Score threshold selected in the training set. The error bars represent the two-sided 95% Wilson confidence interval.

Back to article page