Training Neural Networks for Computer-Aided Diagnosis: Experience in the Intelligence Community

Neural networks are often used in computeraided diagnosis systems for detecting clinically significant objects. They have also been applied in the Intelligence Community to cue Image Analysts (IAs) for assisted target recognition and wide-area search. Given the similarity between the applications in the two communities, there are a number of common issues that must be considered when training these neural networks. Two such issues are: 1) exploiting information at multiple scales (e.g. context and detail structure) and 2) dealing with uncertainty (e.g. errors in truth data). Working with The University of Chicago, we address these two issues, transferring architectures and training algorithms, originally developed for assisting IAs in search applications, to improve computer-aided diagnosis for mammography. These include hierarchical pyramid/neural-network (HPNN) architectures that automatically learn and integrate multi-resolution features for improving microcalcification and mass detection in computer-aided diagnosis (CAD) systems. These networks are trained using an uncertain object position (UOP) error function for the supervised learning of image search/detection tasks when the position of the objects to be found is uncertain or ill-defined. Results show that the HPNN architecture trained using the UOP error function reduces the false positive rate of a mammographic CAD system by 30%-50% without significant loss in sensitivity. We conclude that the transfer of assisted target recognition technology from the Intelligence Community to the Medical Community can significantly impact the clinically utility of CAD systems.