GAN-based anomaly detection in multi-modal MRI images
S. Benson, Regina G. H. Beets‐Tan
This paper uses GANs trained on healthy MRI images to detect tumours in patient scans without prior knowledge of tumour location, achieving AUC > 0.75 for anomalies > 4 cm².
bioRxiv · Jul 10, 2020 · PreprintBioinformatics