Patch vs. Global Image-Based Unsupervised Anomaly Detection in MR Brain Scans of Early Parkinsonian Patients

TL;DR

This paper compares patch-based and global image-based unsupervised anomaly detection methods in MRI scans of early Parkinson's patients, highlighting the advantages of patch-based approaches for subtle brain lesion detection.

Contribution

It introduces a comparison between a spatial auto-encoder and a patch-fed siamese auto-encoder for Parkinson's anomaly detection, demonstrating the benefits of patch-based methods.

Findings

SAE outperforms the spatial auto-encoder on average

Patch-based models better capture local anomaly patterns

Global models may lose fine structural details

Abstract

Although neural networks have proven very successful in a number of medical image analysis applications, their use remains difficult when targeting subtle tasks such as the identification of barely visible brain lesions, especially given the lack of annotated datasets. Good candidate approaches are patch-based unsupervised pipelines which have both the advantage to increase the number of input data and to capture local and fine anomaly patterns distributed in the image, while potential inconveniences are the loss of global structural information. We illustrate this trade-off on Parkinson's disease (PD) anomaly detection comparing the performance of two anomaly detection models based on a spatial auto-encoder (AE) and an adaptation of a patch-fed siamese auto-encoder (SAE). On average, the SAE model performs better, showing that patches may indeed be advantageous.