MADGAN: unsupervised medical anomaly detection GAN using multiple adjacent brain MRI slice reconstruction

Shin'Ichi Satoh,Kohei Murao,Yuki Shimahara,Changhee Han,Tomoyuki Noguchi,Leonardo Rundo,Evis Sala,Hideki Nakayama,Saori Koshino,Zoltán Ádám Milacski

doi:10.1186/s12859-020-03936-1

Abstract

BackgroundUnsupervised learning can discover various unseen abnormalities, relying on large-scale unannotated medical images of healthy subjects. Towards this, unsupervised methods reconstruct a 2D/3D single medical image to detect outliers either in the learned feature space or from high reconstruction loss. However, without considering continuity between multiple adjacent slices, they cannot directly discriminate diseases composed of the accumulation of subtle anatomical anomalies, such as Alzheimer’s disease (AD). Moreover, no study has shown how unsupervised anomaly detection is associated with either disease stages, various (i.e., more than two types of) diseases, or multi-sequence magnetic resonance imaging (MRI) scans.ResultsWe propose unsupervised medical anomaly detection generative adversarial network (MADGAN), a novel two-step method using GAN-based multiple adjacent brain MRI slice reconstruction to detect brain anomalies at different stages on multi-sequence structural MRI: (Reconstruction) Wasserstein loss with Gradient Penalty + 100 ell _1 loss—trained on 3 healthy brain axial MRI slices to reconstruct the next 3 ones—reconstructs unseen healthy/abnormal scans; (Diagnosis) Average ell _2 loss per scan discriminates them, comparing the ground truth/reconstructed slices. For training, we use two different datasets composed of 1133 healthy T1-weighted (T1) and 135 healthy contrast-enhanced T1 (T1c) brain MRI scans for detecting AD and brain metastases/various diseases, respectively. Our self-attention MADGAN can detect AD on T1 scans at a very early stage, mild cognitive impairment (MCI), with area under the curve (AUC) 0.727, and AD at a late stage with AUC 0.894, while detecting brain metastases on T1c scans with AUC 0.921.ConclusionsSimilar to physicians’ way of performing a diagnosis, using massive healthy training data, our first multiple MRI slice reconstruction approach, MADGAN, can reliably predict the next 3 slices from the previous 3 ones only for unseen healthy images. As the first unsupervised various disease diagnosis, MADGAN can reliably detect the accumulation of subtle anatomical anomalies and hyper-intense enhancing lesions, such as (especially late-stage) AD and brain metastases on multi-sequence MRI scans.

Highlights

Unsupervised learning can discover various unseen abnormalities, relying on large-scale unannotated medical images of healthy subjects
Han et al BMC Bioinformatics 2021, 22(Suppl 2):31 detect the accumulation of subtle anatomical anomalies and hyper-intense enhancing lesions, such as Alzheimer’s disease (AD) and brain metastases on multi-sequence magnetic resonance imaging (MRI) scans
Researchers reconstructed a single medical image via generative adversarial network (GAN) [14], autoencoders (AEs) [15], or combining them, since GANs can generate realistic images and AEs, especially variational AEs (VAEs), can directly map data onto its latent representation [16]; unseen images were scored by comparing them with reconstructed ones to discriminate a pathological image distribution

Summary

Introduction

Unsupervised learning can discover various unseen abnormalities, relying on large-scale unannotated medical images of healthy subjects. In medical imaging preparing such massive annotated datasets is often unfeasible [6, 7]; to tackle this pervasive problem, researchers have proposed various data augmentation techniques, including generative adversarial network (GAN)-based ones [8,9,10,11,12,13] ; alternatively, Rauschecker et al combined convolutional neural networks (CNNs), feature engineering, and expertknowledge Bayesian network to derive brain magnetic resonance imaging (MRI) differential diagnoses that approach neuroradiologists’ accuracy for 19 diseases Even exploiting these techniques, supervised learning still requires many images with pathological features, even for rare diseases, to make a reliable diagnosis; it can only detect already-learned specific pathologies. No study has shown so far how unsupervised anomaly detection is associated with either disease stages, various (i.e., more than 2 types of ) diseases, or multi-sequence MRI scans

Methods

Results

Discussion

Conclusion