Harmonizing Flows: Unsupervised MR Harmonization Based on Normalizing Flows

Beizaee, Farzad; Desrosiers, Christian; Lodygensky, Gregory A.; Dolz, José

doi:10.1007/978-3-031-34048-2_27

Cited by 3 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Distribution differences in covariates, such as demographics and clinical diagnoses, across datasets are inevitable. Most deep learning harmonization approaches directly align distributions of latent representations across datasets without explicitly modeling covariate differences (Dewey et al, 2019; Zuo et al, 2021; Beizaee et al, 2023; Liu et al, 2023).…”

Section: Methodsmentioning

confidence: 99%

“…Site information is then concatenated to the latent representations to reconstruct the MRI data via a decoder. Generative adversarial networks (Dewey et al, 2019; Zhao et al, 2019; Modanwal et al, 2020; Bashyam et al, 2021), normalizing flow (Wang et al, 2021; Beizaee et al, 2023) and federated learning (Dinsdale et al, 2022) have also been explored. However, existing DNN approaches typically overlook the inclusion of covariates, which are explicitly controlled in mixed effects harmonization models (Fortin et al, 2017, 2018; Chen et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DeepResBat: deep residual batch harmonization accounting for covariate distribution differences

An,

Zhang,

Wulan

et al. 2024

Preprint

View full text Add to dashboard Cite

Pooling MRI data from multiple datasets requires harmonization to reduce undesired inter- site variabilities, while preserving effects of biological variables (or covariates). The popular harmonization approach ComBat uses a mixed effect regression framework that explicitly accounts for covariate distribution differences across datasets. There is also significant interest in developing harmonization approaches based on deep neural networks (DNNs), such as conditional variational autoencoder (cVAE). However, current DNN approaches do not explicitly account for covariate distribution differences across datasets. Here, we provide mathematical results, suggesting that not accounting for covariates can lead to suboptimal harmonization outcomes. We propose two DNN-based harmonization approaches that explicitly account for covariate distribution differences across datasets: covariate VAE (coVAE) and DeepResBat. The coVAE approach is a natural extension of cVAE by concatenating covariates and site information with site- and covariate-invariant latent representations. DeepResBat adopts a residual framework inspired by ComBat. DeepResBat first removes the effects of covariates with nonlinear regression trees, followed by eliminating site differences with cVAE. Finally, covariate effects are added back to the harmonized residuals. Using three datasets from three different continents with a total of 2787 participants and 10085 anatomical T1 scans, we find that DeepResBat and coVAE outperformed ComBat, CovBat and cVAE in terms of removing dataset differences, while enhancing biological effects of interest. However, coVAE hallucinates spurious associations between anatomical MRI and covariates even when no association exists. Therefore, future studies proposing DNN-based harmonization approaches should be aware of this false positive pitfall. Overall, our results suggest that DeepResBat is an effective deep learning alternative to ComBat.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

DeepResBat: deep residual batch harmonization accounting for covariate distribution differences

An,

Zhang,

Wulan

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

DeepResBat: Deep residual batch harmonization accounting for covariate distribution differences

An,

Zhang,

Wulan

et al. 2025

Medical Image Analysis

View full text Add to dashboard Cite

Med-cDiff: Conditional Medical Image Generation with Diffusion Models

Hung,

Zhao,

Zheng

et al. 2023

Bioengineering

View full text Add to dashboard Cite

Conditional image generation plays a vital role in medical image analysis as it is effective in tasks such as super-resolution, denoising, and inpainting, among others. Diffusion models have been shown to perform at a state-of-the-art level in natural image generation, but they have not been thoroughly studied in medical image generation with specific conditions. Moreover, current medical image generation models have their own problems, limiting their usage in various medical image generation tasks. In this paper, we introduce the use of conditional Denoising Diffusion Probabilistic Models (cDDPMs) for medical image generation, which achieve state-of-the-art performance on several medical image generation tasks.

show abstract

Harmonizing Flows: Unsupervised MR Harmonization Based on Normalizing Flows

Cited by 3 publications

References 27 publications

DeepResBat: deep residual batch harmonization accounting for covariate distribution differences

DeepResBat: deep residual batch harmonization accounting for covariate distribution differences

DeepResBat: Deep residual batch harmonization accounting for covariate distribution differences

Med-cDiff: Conditional Medical Image Generation with Diffusion Models

Contact Info

Product

Resources

About