A deep learning-based multisite neuroimage harmonization framework established with a traveling-subject dataset

Tian, Dezheng; Zeng, Zilong; Sun, Xian; Tong, Qiqi; Li, Huanjie; He, Hongjian; Gao, Jia‐Hong; He, Yong; Xia, Mingrui

doi:10.1016/j.neuroimage.2022.119297

Cited by 28 publications

(19 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Papers were also categorised according to the harmonisation features: raw signals ( n = 29) and image-derived features ( n = 12). Of the 28 studies on sMRI, 24 harmonised raw intensity signals, while four studies focused on derived measures, including brain volumes [ 25 , 46 ], cortical thickness [ 36 ] and mixed image-based features [ 66 ]. Four studies harmonised the raw diffusion signals of dMRI through dictionary representation [ 31 ] and spherical harmonics representation [ 29 , 39 , 48 ], while three studies focused on derived features [ 28 , 41 , 65 ].…”

Section: Resultsmentioning

confidence: 99%

“…Studies that used explicit methods commonly employed statistical tests to compute the probabilities of differences. The most commonly used metrics were Cohen’s d value (e.g., [ 36 , 41 ]), Pearson’s r value (e.g., [ 41 ]), and Kullback–Leibler divergence (e.g., [ 31 , 46 ]). Implicit methods, on the other hand, used the t-SNE algorithm to directly visualise the data distributions based on the features extracted from the trained network.…”

Section: Resultsmentioning

confidence: 99%

“…ML methods have provided efficient solutions for explicitly generating harmonised MRI data across different domains. The most widely used models are GANs and autoecoders, which have shown promising results in reducing multi-site variation through image-to-image synthesis [ 40 , 46 , 49 ]. GANs perform domain translation by learning domain-invariant features [ 34 , 80 ].…”

Section: Discussionmentioning

confidence: 99%

“…Autoencoder-based methods aim to harmonise data in terms of disentangled representations [ 83 ]. All related work in this review attempted to extract scanner-related features for harmonisation [ 43 , 44 , 45 , 46 , 47 , 48 , 49 ]. Similar to GANs, data from multiple sources and target domains are required for training.…”

Section: Discussionmentioning

confidence: 99%

See 3 more Smart Citations

Machine Learning for Brain MRI Data Harmonisation: A Systematic Review

et al. 2023

View full text Add to dashboard Cite

Background: Magnetic Resonance Imaging (MRI) data collected from multiple centres can be heterogeneous due to factors such as the scanner used and the site location. To reduce this heterogeneity, the data needs to be harmonised. In recent years, machine learning (ML) has been used to solve different types of problems related to MRI data, showing great promise. Objective: This study explores how well various ML algorithms perform in harmonising MRI data, both implicitly and explicitly, by summarising the findings in relevant peer-reviewed articles. Furthermore, it provides guidelines for the use of current methods and identifies potential future research directions. Method: This review covers articles published through PubMed, Web of Science, and IEEE databases through June 2022. Data from studies were analysed based on the criteria of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Quality assessment questions were derived to assess the quality of the included publications. Results: a total of 41 articles published between 2015 and 2022 were identified and analysed. In the review, MRI data has been found to be harmonised either in an implicit (n = 21) or an explicit (n = 20) way. Three MRI modalities were identified: structural MRI (n = 28), diffusion MRI (n = 7) and functional MRI (n = 6). Conclusion: Various ML techniques have been employed to harmonise different types of MRI data. There is currently a lack of consistent evaluation methods and metrics used across studies, and it is recommended that the issue be addressed in future studies. Harmonisation of MRI data using ML shows promises in improving performance for ML downstream tasks, while caution should be exercised when using ML-harmonised data for direct interpretation.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Machine Learning for Brain MRI Data Harmonisation: A Systematic Review

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Deep learning frameworks have been employed in a broad spectrum of medical image (post-) reconstruction applications ranging from improving image quality in magnetic resonance imaging (MRI), CT, and [ 18 F]FDG PET/CT [316-319], quality control and the identification of EARL-compliance [320], reduction of scan time [318; 319; 321] or reconstruction time [322], and the transformation from one modality to another, generating synthetic low-dose CTs based on non-attenuation-corrected PET images [323], zero-echo-time MRI [324], or synthetic CT based on MRI for radiotherapy planning [325]. For PET harmonisation, we could learn from (predominantly T1-weighted) MRI, where deep learning is used for the harmonisation in multicentre effects [326][327][328]. In the years to come, these algorithms will be developed even further, which requires large amounts of data.…”

Section: Acquisition and Reconstructionmentioning

confidence: 99%

Radiomics in [18F]FDG PET/CT: A leap in the dark?

Noortman¹

View full text Add to dashboard Cite

be reproduced, stored in a retrieval system or transmitted in any form or by any means without permission of the author. Alle rechten voorbehouden. Niets uit deze uitgave mag worden vermenigvuldigd, in enige vorm of op enige wijze, zonder voorafgaande schriftelijke toestemming van de auteur. RADIOMICS IN [18 F]FDG PET/CT: A LEAP IN THE DARK? PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit Twente, op gezag van de rector magnificus, prof. dr. ir. A. Veldkamp, volgens besluit van het College voor Promoties in het openbaar te verdedigen op woensdag 10 mei 2023 om 12.45 uur door

show abstract

Addressing multi‐site functional MRI heterogeneity through dual‐expert collaborative learning for brain disease identification

Fang

Potter

et al. 2023

Human Brain Mapping

View full text Add to dashboard Cite

Several studies employ multi‐site rs‐fMRI data for major depressive disorder (MDD) identification, with a specific site as the to‐be‐analyzed target domain and other site(s) as the source domain. But they usually suffer from significant inter‐site heterogeneity caused by the use of different scanners and/or scanning protocols and fail to build generalizable models that can well adapt to multiple target domains. In this article, we propose a dual‐expert fMRI harmonization (DFH) framework for automated MDD diagnosis. Our DFH is designed to simultaneously exploit data from a single labeled source domain/site and two unlabeled target domains for mitigating data distribution differences across domains. Specifically, the DFH consists of a domain‐generic student model and two domain‐specific teacher/expert models that are jointly trained to perform knowledge distillation through a deep collaborative learning module. A student model with strong generalizability is finally derived, which can be well adapted to unseen target domains and analysis of other brain diseases. To the best of our knowledge, this is among the first attempts to investigate multi‐target fMRI harmonization for MDD diagnosis. Comprehensive experiments on 836 subjects with rs‐fMRI data from 3 different sites show the superiority of our method. The discriminative brain functional connectivities identified by our method could be regarded as potential biomarkers for fMRI‐related MDD diagnosis.

show abstract

A deep learning-based multisite neuroimage harmonization framework established with a traveling-subject dataset

Cited by 28 publications

References 49 publications

Machine Learning for Brain MRI Data Harmonisation: A Systematic Review

Machine Learning for Brain MRI Data Harmonisation: A Systematic Review

Radiomics in [18F]FDG PET/CT: A leap in the dark?

Addressing multi‐site functional MRI heterogeneity through dual‐expert collaborative learning for brain disease identification

Contact Info

Product

Resources

About