bcROCsurface

bcROCsurface implements bias-corrected estimation of ROC surfaces and their volumes to adjust for verification bias when evaluating diagnostic tests with three ordered disease statuses.

Key Features:

• Bias correction: Corrects for verification bias in the estimation of ROC surfaces and their volumes.
• ROC surface and volume estimation: Provides estimation of the ROC surface and the volume under the surface (VUS) for three-class diagnostic tests.
• Three-class ordered disease support: Explicitly supports analyses for ordered disease categories (e.g., non-diseased, intermediate, diseased).
• Missing data handling: Adjusts for missingness in true disease status arising from verification processes.
• Utility functions: Offers functions to perform verification-bias-corrected ROC surface analyses on complex datasets.

Scientific Applications:

• Medical diagnostics evaluation: Assess diagnostic test accuracy across three ordered disease states while accounting for verification bias.
• Marker development and validation: Support development, validation, and improvement of diagnostic markers evaluated in three-class settings.

Methodology:

Applies statistical techniques to adjust for missing true disease status data and produce verification-bias-corrected ROC surface and volume estimates.