Statistical Inference on the Accuracy of Computer-Aided Diagnosis (CAD) Using Artificial Intelligence-Based Systems

• Activity Number: 384 - Recent Statistical Advances in Diagnostic Medicine
• Type: Invited
• Date/Time: Wednesday, August 5, 2020 : 1:00 PM to 2:50 PM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Xiao-Hua Zhou*
• Companies: Peking University
• Keywords: Artificial neural networks; Classification; Deep learning; Machine learning; Sensitivity; Specificity

Abstract

In the digital age of medicine computer-aided diagnosis (CAD) is commonly used in the routine clinical work for the detection and diagnosis of various diseases in different imaging modalities, such as the differential diagnosis of lung nodules and interstitial lung diseases in chest radiography. Methods based on Artificial neural networks (ANN), such as deep learning methods, are also commonly used to develop the computer algorithm for the CAD. In this talk,I will outline assessment methodologies and statistical issues for the performance of AI-based CAD systems. I will also discuss some new methodological development in assessing the performance of AI-based CAD systems.