Overview / Usage

Models are typically based on large amounts of data with annotated examples of
known markers aiming at automating detection.Performing unsupervised learning to
identify anomalies in imaging data as candidates for markers. By using deep convolutional generative adversarial network to learn a manifold of normal anatomical variability, we can achieve high accuracy in anomaly detection. Medical imaging enables the observation of markers correlating with disease status, and treatment response. Generative model will generate anomalies . The training procedure for Generative model is to maximize the probability of discriminative model making a mistake.