Use of Sub-Ensembles and Multi-Template Observers to Evaluate Detection Task Performance for Data That are Not Multivariate Normal

Keyphrases

• Task Performance 100%
• Detector Response 100%
• Hotelling Observer 100%
• Multivariate Normal 100%
• Multi-template 100%
• Test Statistic 60%
• Reconstruction Parameters 60%
• Linear Observer 40%
• Linear Discriminant 40%
• Data Ensemble 40%
• Template Strategy 40%
• Observer Model 40%
• Image Quality 20%
• Medical Imaging 20%
• Optimal Reconstruction 20%
• Reconstructed Image 20%
• Area under the Receiver Operating Characteristic (AUROC) 20%
• SPECT Myocardial Perfusion Imaging 20%
• Pooled Testing 20%
• Single Template 20%
• Simulated Population 20%
• Homoscedasticity 20%

Computer Science

• Task Performance 100%
• Linear Discriminant 100%
• Image Quality 50%
• Reconstructed Image 50%
• Medical Imaging 50%
• Characteristic Curve 50%
• Single Template 50%