ImUnity: A generalizable VAE-GAN solution for multicenter MR image harmonization

• Published in: Medical image analysis Vol. 88; p. 102799
• Main Authors: Cackowski, Stenzel, Barbier, Emmanuel L., Dojat, Michel, Christen, Thomas
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.08.2023; Elsevier
• Subjects: Artificial Intelligence; Bioengineering; Brain; Computer Science; Data harmonization; Deep Adversarial Network; Imaging; Life Sciences; Radiomic features; Self-supervised learning; Brain; Deep Adversarial Network; Data harmonization; Radiomic features; Self-supervised learning
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2023.102799

ImUnity is an original 2.5D deep-learning model designed for efficient and flexible MR image harmonization. A VAE-GAN network, coupled with a confusion module and an optional biological preservation module, uses multiple 2D slices taken from different anatomical locations in each subject of the training database, as well as image contrast transformations for its training. It eventually generates ‘corrected’ MR images that can be used for various multi-center population studies. Using 3 open source databases (ABIDE, OASIS and SRPBS), which contain MR images from multiple acquisition scanner types or vendors and a large range of subjects ages, we show that ImUnity: (1) outperforms state-of-the-art methods in terms of quality of images generated using traveling subjects; (2) removes sites or scanner biases while improving patients classification; (3) harmonizes data coming from new sites or scanners without the need for an additional fine-tuning and (4) allows the selection of multiple MR reconstructed images according to the desired applications. Tested here on T1-weighted images, ImUnity could be used to harmonize other types of medical images. [Display omitted] •We propose a modified VAE-GAN model for versatile T1-MRI domain adaptation.•ImUnity harmonizes multi-site data without the need for traveling subjects.•ImUnity removes site and scanner bias while preserving clinical information.•ImUnity reaches state of the art when harmonizing images from traveling subjects.•ImUnity improves clinical classification of Autism Spectrum Disorders.