Mirai/Sybil Research FAQ

Here are some clarifying responses to questions that we get often about terminology that is often used in research on Mirai and Sybil.

What is an AUC?

AUC stands for ‘Area Under the Curve’ and typically refers to the Area Under the Receiver Operating Characteristic Curve (AUROC). The ROC curve plots the true positive rate against the false positive rate at various classification thresholds, providing a visual representation of a model’s ability to distinguish between classes. The AUC summarizes this curve into a single number ranging from 0 to 1. An AUC of 1.0 indicates perfect classification, while an AUC of 0.5 suggests no discriminative ability, which is equivalent to random guessing. Although AUC is a measure of a model’s discriminative performance, it should not be interpreted as percentage accuracy, e.g., an AUC of 0.5 does not mean the model is 50% accurate.

What is a retrospective study?

A retrospective study is a study conducted on past data. In the case of Mirai and Sybil, this means the study was conducted on past patients whose future diagnoses were already known so researchers were able to validate the accuracy of the tool.

What is a prospective study?

A prospective study is conducted on patients whose future diagnoses are not yet known. In the case of Mirai and Sybil, this would be the equivalent of conducting a clinical trial.

Can I enroll in a prospective study?

Unfortunately, we are not able to enroll patients in a prospective study as this is entirely up to the hospital/institution and depends on what types of patients they want to enroll. If you are interested in having Mirai or Sybil run on your mammogram or CT scan, we recommend asking your local hospital to assist!