Harmonization Across Imaging Locations(HAIL): One-Shot Learning for Brain MRI

TL;DR

This paper introduces a one-shot learning approach using neural style transfer to harmonize brain MRI images across different clinical sites, effectively reducing site-specific variations while preserving anatomical details.

Contribution

It presents a novel one-shot harmonization method that prevents hallucinations in MRI images by combining neural style transfer with adaptive normalization, applicable to unseen sites.

Findings

Effective preservation of anatomical structures in MRI after harmonization

Successful adjustment of image intensities to match different clinical sites

Model generalizes well to unseen data from new sites

Abstract

For machine learning-based prognosis and diagnosis of rare diseases, such as pediatric brain tumors, it is necessary to gather medical imaging data from multiple clinical sites that may use different devices and protocols. Deep learning-driven harmonization of radiologic images relies on generative adversarial networks (GANs). However, GANs notoriously generate pseudo structures that do not exist in the original training data, a phenomenon known as "hallucination". To prevent hallucination in medical imaging, such as magnetic resonance images (MRI) of the brain, we propose a one-shot learning method where we utilize neural style transfer for harmonization. At test time, the method uses one image from a clinical site to generate an image that matches the intensity scale of the collaborating sites. Our approach combines learning a feature extractor, neural style transfer, and adaptive instance normalization. We further propose a novel strategy to evaluate the effectiveness of image harmonization approaches with evaluation metrics that both measure image style harmonization and assess the preservation of anatomical structures. Experimental results demonstrate the effectiveness of our method in preserving patient anatomy while adjusting the image intensities to a new clinical site. Our general harmonization model can be used on unseen data from new sites, making it a valuable tool for real-world medical applications and clinical trials.