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

TL;DR

This paper introduces a novel anomaly detection approach for mammography scans by combining autoencoders with sum-product networks, achieving superior performance over existing methods in medical anomaly detection.

Contribution

It proposes a new hybrid model that integrates autoencoders with sum-product networks for efficient and accurate anomaly detection in medical images.

Findings

Superior performance over standalone models

Effective patch-wise processing in mammography

State-of-the-art results in medical anomaly detection

Abstract

Unsupervised anomaly detection models which are trained solely by healthy data, have gained importance in the 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.