Anomaly Detection Using Generative Models and Sum-Product Networks in Mammography Scans

Funding Information: VRVis is funded by BMK, BMDW, Styria, SFG, Tyrol and Vienna Business Agency in the scope of COMET-Competence Centers for Excellent Technologies (879730) which is managed by FFG. Thanks go to AGFA HealthCare, project partner of VRVis, for providing valuable input. Martin Trapp acknowledges funding from the Academy of Finland (347279).Unsupervised anomaly detection models that are trained solely by healthy data, have gained importance in recent years, as the annotation of medical data is a tedious task. Autoencoders and generative adversarial networks are the standard anomaly detection methods that are utilized to learn the data distribution. However, they fall short when it comes to inference and evaluation of the likelihood of test samples. We propose a novel combination of generative models and a probabilistic graphical model. After encoding image samples by autoencoders, the distribution of data is modeled by Random and Tensorized Sum-Product Networks ensuring exact and efficient inference at test time. We evaluate different autoencoder architectures in combination with Random and Tensorized Sum-Product Networks on mammography images using patch-wise processing and observe superior performance over utilizing the models standalone and state-of-the-art in anomaly detection for medical data.Peer reviewe