Multi-modal mri image synthesis via gan with multi-scale gate mergence
Abstract
Multi-modal magnetic resonance imaging (MRI) plays a critical role in clinical diagnosis and treatment nowadays. Each modality of MRI presents its own specific anatomical features which serve as complementary information to other modalities and can provide rich diagnostic information. However, due to the limitations of time consuming and expensive cost, some image sequences of patients may be lost or corrupted, posing an obstacle for accurate diagnosis. Although current multi-modal image synthesis approaches are able to alleviate the issues to some extent, they are still far short of fusing modalities effectively. In light of this, we propose a multi-scale gate mergence based generative adversarial network model, namely MGM-GAN, to synthesize one modality of MRI from others. Notably, we have multiple down-sampling branches corresponding to input modalities to specifically extract their unique features. In contrast to the generic multi-modal fusion approach of averaging or maximizing operations, we introduce a gate mergence (GM) mechanism to automatically learn the weights of different modalities across locations, enhancing the task-related information while suppressing the irrelative information. As such, the feature maps of all the input modalities at each down-sampling level, i.e., multi-scale levels, are integrated via GM module. In addition, both the adversarial loss and the pixel-wise loss, as well as gradient difference loss (GDL) are applied to train the network to produce the desired modality accurately. Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art multi-modal image synthesis methods.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Zhan et al. proposed MGM-GAN, which leveraged a gate emergence mechanism to learn important weights across different modalities, enabling synthesis of incomplete modalities in MRI.
“Zhan et al. proposed MGM-GAN, which leveraged a gate emergence mechanism to learn important weights across different modalities, enabling the synthesis of incomplete modalities in MRI [61].”
From Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis (2025) · cites this paper as [61]
Cites
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404 · see citing works
Random forest regression for magnetic resonance image synthesis
Jog, A., Carass, A., Roy, S., Pham, D. L., & Prince, J. L. (2017) · Medical Image Analysis · doi:10.1016/j.media.2016.08.009 · Cited by 212 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Deep learning based synthesis of MRI, CT and PET: Review and analysis
Dayarathna, S., Islam, K. T., Uribe, S. A., Yang, G., Hayat, M., & Chen, Z. (2023) · Medical Image Analysis · doi:10.1016/j.media.2023.103046 · Cited by 285 · see citing works
GAN review: Models and medical image fusion applications
Zhou, T. Y., Li, Q., Lu, H., Cheng, Q., & Zhang, X. (2022) · Information Fusion · doi:10.1016/j.inffus.2022.10.017 · Cited by 269 · see citing works
CoLa-Diff: Conditional Latent Diffusion Model for Multi-Modal MRI Synthesis
Jiang, L., Mao, Y., Chen, X., Wang, X.-F., & Li, C. (2023) · International Conference on Medical Image Computing and Computer-Assisted Intervention · doi:10.48550/arxiv.2303.14081 · Cited by 133
“Recent generative adversarial networks (GANs) and variants, e.g., MM-GAN [22], DiamondGAN [12] and ProvoGAN [29], have been successful based on multi-modal MRI, further improved by introducing multi-modal coding [30], enhanced architecture [6], and novel learning strategies [28].”
BackgroundD2-Net: Dual Disentanglement Network for Brain Tumor Segmentation With Missing Modalities
Yang, Q., Guo, X., Chen, Z., Woo, P. Y. M., & Yuan, Y.-X. (2022) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2022.3175478 · Cited by 117 · see citing works
“A rational reason is that these commonly used reconstruction constraints may derive degenerate solutions and fail to decompose features well, while adversarial training based models [10]–[12], [20] lack explicit guidance of modality-specific information.”
BackgroundMethodInfluentialGenerative AI for brain image computing and brain network computing: a review
Gong, C., Jing, C., Chen, X., Pun, C. M., Huang, G., Saha, A., Nieuwoudt, M., Li, H.-X., Hu, Y., & Wang, S. (2023) · Frontiers in Neuroscience · doi:10.3389/fnins.2023.1203104 · Cited by 111 · see citing works
Diagnosis of brain diseases in fusion of neuroimaging modalities using deep learning: A review
Shoeibi, A., Khodatars, M., Jafari, M., Ghassemi, N., Moridian, P., Alizadehsani, R., Ling, S. H., Khosravi, A., Alinejad‐Rokny, H., Lam, H., et al. (2022) · Information Fusion · doi:10.1016/j.inffus.2022.12.010 · Cited by 106 · see citing works
Prezja, F., Paloneva, J., Pölönen, I., Niinimäki, E., & Äyrämö, S. (2022) · Scientific Reports · doi:10.1038/s41598-022-23081-4 · Cited by 76 · see citing works
Similarity and quality metrics for MR image-to-image translation
Dohmen, M., Klemens, M., Baltruschat, I. M., Truong, T., & Lenga, M. (2025) · Scientific Reports · doi:10.1038/s41598-025-87358-0 · Cited by 74 · see citing works
“Therefore, we first introduce normalization methods that are frequently used to bring MR images to a common intensity value scale or as prerequisite for certain metrics.”
Swin transformer-based GAN for multi-modal medical image translation
Yan, S., Wang, C.-Y., Chen, W., & Lyu, J. (2022) · Frontiers in Oncology · doi:10.3389/fonc.2022.942511 · Cited by 70 · see citing works
“multi-modal MR images of the same anatomical structure (1, 2).”
BackgroundChallenges in AI-driven Biomedical Multimodal Data Fusion and Analysis
Liu, J., Cen, X., Yi, C., Wang, F.-A., Ding, J., Cheng, J., Wu, Q., Gai, B., Zhou, Y., He, R., et al. (2025) · Genomics, Proteomics & Bioinformatics · doi:10.1093/gpbjnl/qzaf011 · Cited by 75 · see citing works
Assessing the efficacy of 3D Dual-CycleGAN model for multi-contrast MRI synthesis
Mahboubisarighieh, A., Shahverdi, H., Nesheli, S. J., Kermani, M. A., Niknam, M., Torkashvand, M., & Rezaeijo, S. M. (2024) · The Egyptian Journal of Radiology and Nuclear Medicine · doi:10.1186/s43055-024-01287-y · Cited by 53
3D multi-modality Transformer-GAN for high-quality PET reconstruction
Wang, Y., Luo, Y., Zu, C., Zhan, B., Jiao, Z., Wu, X., Zhou, J.-L., Shen, D.-G., & Zhou, L.-P. (2023) · Medical Image Anal. · doi:10.1016/j.media.2023.102983 · Cited by 50
Self-citationSun, K., Ding, J., Li, Q., Chen, W., Zhang, H., Sun, J., Jiao, Z.-Q., & Ni, X. (2024) · Quantitative Imaging in Medicine and Surgery · doi:10.21037/qims-24-9 · Cited by 42 · see citing works
“Another more typical solution is to synthesize the missing modalities by generative modeling and then using the complete modalities for segmentation (33,34).”
Yurt, M., Özbey, M., Dar, S., Tınaz, B., Oğuz, K., & Cukur, T. (2020) · Medical Image Anal. · doi:10.1016/j.media.2022.102429 · Cited by 40 · see citing works
Generalized Generative Deep Learning Models for Biosignal Synthesis and Modality Transfer
Dissanayake, T., Fernando, T., Denman, S., Sridharan, S., & Fookes, C. (2022) · IEEE journal of biomedical and health informatics · doi:10.1109/jbhi.2022.3223777 · Cited by 35 · see citing works
“Datagenerationandaugmentationaremajorapplicationareas ofgenerativemodeling[12],[13].”
Ahmad, S., Ullah, Z., & Gwak, J. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111854 · Cited by 33 · see citing works
Zhu, L., He, Q., Huang, Y., Zhang, Z., Zeng, J.-M., Lu, L., Kong, W.-M., & Zhou, F.-Q. (2022) · Comput. Biol. Medicine · doi:10.1016/j.compbiomed.2022.105387 · Cited by 33 · see citing works
Qiu, C.-J., Song, Y.-Q., Liu, Y., Zhu, Y., Han, K., Sheng, V. S., & Liu, Z. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2023.105827 · Cited by 31 · see citing works
Farhan, A. S., Khalid, M., & Manzoor, U. (2025) · Frontiers Artif. Intell. · doi:10.3389/frai.2025.1525240 · Cited by 28 · see citing works
“MRI generates images of different contrasts because protons in the tissues vary in their relaxation rates (Zhan et al., 2022; Zhou et al., 2019).”
GAN Inversion for Data Augmentation to Improve Colonoscopy Lesion Classification
Golhar, M. V., Bobrow, T. L., Ngamruengphong, S., & Durr, N. J. (2022) · IEEE journal of biomedical and health informatics · doi:10.1109/jbhi.2024.3397611 · Cited by 26 · see citing works
“Furthermore, GANs have demonstrated significant potential in multi-modal image translation, facilitating the conversion be-tween CT and MRI [16], different MRI modalities [17], and low-dose CT denoising [18], among others.”
Diamantis, D. E., Gatoula, P., Koulaouzidis, A., & Iakovidis, D. K. (2023) · IEEE Access · doi:10.1109/access.2024.3366801 · Cited by 23 · see citing works
“Also, the qualitative results validate that multiscale information can enhance image quality, which has also been observed in the context of multiscale GANs [31].”
Mamo, A. A., Gebresilassie, B. G., Mukherjee, A., Hassija, V., & Chamola, V. (2024) · Cognitive Computation · doi:10.1007/s12559-024-10291-3 · Cited by 21 · see citing works
“Owing to the scarcity of large, labeled medical datasets due to privacy concerns and the complexities of data acquisition, GANs can be trained on existing medical images to synthesize new images [47, 48], effectively expanding the dataset and enhancing model generalization.”
Multimodal Medical Image Fusion Network Based on Target Information Enhancement
Zhou, Y.-T., Yang, X., Liu, S.-Q., & Yin, J.-P. (2024) · IEEE Access · doi:10.1109/access.2024.3402965 · Cited by 21 · see citing works
“On the contrary, deep learning methods CNN [16] and GAN [17] have the advantages of detailed edge texture information, reduced computational cost, and elimination of explicit fusion rule design.”
Mosqueira-Rey, E., Hernández-Pereira, E., Bobes-Bascarán, J., Alonso-Ríos, D., Pérez-Sánchez, A., Fernández-Leal, Á., Moret‐Bonillo, V., Vidal-Ínsua, Y., & Vázquez-Rivera, F. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-09197-2 · Cited by 23 · see citing works
“Given their excellent performance, they have also attracted the attention of researchers in the medical image fusion field, as exemplified by Fu et al. [19], Zhan et al. [91], Jiang et al. [38], and Guo et al. [23].”
Diao, Y.-Q., Li, F., & Li, Z. (2023) · Comput. Biol. Medicine · doi:10.1016/j.compbiomed.2023.107234 · Cited by 17 · see citing works
Bi, Y.-D., Abrol, A., Jia, S., Sui, J., & Calhoun, V. (2024) · NeuroImage · doi:10.1016/j.neuroimage.2024.120674 · Cited by 15 · see citing works
“They excel at simulating neuroimaging data, detecting disease-specific patterns, and enabling modality translation, such as converting T1-weighted to T2-weighted MRI scans (Dar et al., 2019; Zhan et al., 2021).”
Chen, L., Hu, R., Wu, M., & Zhou, X. (2023) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3610909 · Cited by 15 · see citing works
Liu, X.-F., Prince, J. L., Xing, F.-X., Zhuo, J., Reese, T., Stone, M., Fakhri, G. E., & Woo, J. (2023) · Medical Image Anal. · doi:10.1016/j.media.2023.102851 · Cited by 15 · see citing works
“Multi-modal neuroimaging is another area, where image modality translation is widely used to overcome the missing modality issue caused by motion during the acquisition process (Chartsias et al., 2017; Gu et al., 2019; Armanious et al., 2020; Zhou et al., 2020; Zhan et al., 2021).”
Bone tumor necrosis rate detection in few-shot X-rays based on deep learning
Xu, Z.-Y., Niu, K., Tang, S., Song, T., Rong, Y., Guo, W., & He, Z.-Q. (2022) · Comput. Medical Imaging Graph. · doi:10.1016/j.compmedimag.2022.102141 · Cited by 13 · see citing works
Chan, K., Maralani, P., Moody, A., & Khademi, A. (2023) · Frontiers Neuroinformatics · doi:10.3389/fninf.2023.1197330 · Cited by 12 · see citing works
“Recent studies successfully used generative models for translation between brain MRI modalities to synthesize mainly between T1 and T2-weighted images (Kazuhiro et al., 2018; Plassard et al., 2018; Welander et al., 2018; Yang et al., 2018; Chong and Ho, 2021; Osman and Tamam, 2022; Zhan et al., 2022; Zhang et al., 202…”
MU-Diff: a mutual learning diffusion model for synthetic MRI with Application for brain lesions
Dayarathna, S., Wu, Y., Cai, J., Wong, T.-T., Law, M., Islam, K. T., Peiris, H., & Chen, Z. (2025) · npj Artificial Intelligence · doi:10.1038/s44387-025-00016-8 · Cited by 11 · see citing works
“For instance, some approaches 15 – 17 combine latent-level features from multiple contrasts using pixel-level operations.”
QGFormer: Queries-guided transformer for flexible medical image synthesis with domain missing
Hao, H., Xue, J., Huang, P., Ren, L., & Li, D.-W. (2024) · Expert systems with applications · doi:10.1016/j.eswa.2024.123318 · Cited by 10 · see citing works
Self-citationTouati, R., & Kadoury, S. (2023) · Physics in Medicine and Biology · doi:10.1088/1361-6560/acda78 · Cited by 9 · see citing works
“The introduced adversarialmodel in Zhan et al (2021) relies on a gatemergingmechanism formultimodal features fusion, and uses a combination of adversarial, pixel-wise loss and gradient differential losses for the network training.”
MethodAutomatic GAN-based MRI volume synthesis from US volumes: a proof of concept investigation
Vukovic, D., Ruvinov, I., Antico, M., Steffens, M., & Fontanarosa, D. (2023) · Scientific Reports · doi:10.1038/s41598-023-48595-3 · Cited by 9 · see citing works
Xiao, X.-J., Hu, Q. V., & Wang, G. (2025) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2024.3516574 · Cited by 8 · see citing works
“Recently, GAN-based methods were demonstrated to outperform other DL-based methods in many-to-one tasks [10], [14], [51], [52], [53], [54], [55].”
Generative learning-based lightweight MRI brain tumor segmentation with missing modalities
Zhang, X.-L., Chen, Q., He, H., Zhu, L., Xie, Z.-H., Lu, Y.-Y., & Cheng, F. (2024) · Expert systems with applications · doi:10.1016/j.eswa.2024.125478 · Cited by 8 · see citing works
Dayarathna, S., Islam, K. T., Zhuang, B.-H., Yang, G., Cai, J.-F., Law, M., & Chen, Z. (2024) · IEEE Workshop/Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00075 · Cited by 8 · see citing works
“Consequently, multi-contrast MRI has become integral in various clinical applications, including tasks like medical image segmentation and registration [4].”
Hameed, M., Raja, M. A. Z., Zameer, A., Dar, H., Alluhaidan, A., & Aziz, R. (2025) · Scientific Reports · doi:10.1038/s41598-025-17891-5 · Cited by 7 · see citing works
“Iterative denoising using the GAN architecture refines the images provided as inputs by adding high-quality and detailed images, which allows the model to better capture fine-grained AML-specific features 61,62 .”
DBGAN: Dual Branch Generative Adversarial Network for Multi-Modal MRI Translation
Lyu, J., Yan, S., & Hossain, M. S. (2024) · ACM Trans. Multim. Comput. Commun. Appl. · doi:10.1145/3657298 · Cited by 7 · see citing works
“Various pulse sequences and parameters are configured to generate distinct structural and tissue contrast information, a concept referred to as multi-modal [48].”
Bi, Y.-D., Abrol, A., Sui, J., & Calhoun, V. (2023) · IEEE International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp48485.2024.10446450 · Cited by 7 · see citing works
“The use of modality translation tasks in medical imaging is especially notable [4].”
BackgroundTouati, R., & Kadoury, S. (2023) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-023-02916-z · Cited by 7
“The introduced model in [12] is based on a gate mergence (GM) mechanism for features fusion.”
BackgroundMethodSolar PV Generation Prediction Based on Multisource Data Using ROI and Surrounding Area
Kim, B., & Suh, D.-J. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3392337 · Cited by 7 · see citing works
“To address these limitations, MGM-GAN proposed a gate mergence (GM) module to synthesize a multimodal MRI effectively [59].”
MethodDeepCGAN: early Alzheimer's detection with deep convolutional generative adversarial networks
Ali, I., Saleem, N., Alhussein, M. A., Zohra, B., Aurangzeb, K., & Haq, Q. M. U. (2024) · Frontiers in Medicine · doi:10.3389/fmed.2024.1443151 · Cited by 6 · see citing works
“In the field of medical imaging, GANs have found success in tasks like MRI and CT reconstruction and unconditional image synthesis (25, 26).”
Meng, R.-Q., Zhang, X., Huang, S., Gu, Y., Liu, G.-Q., Wu, G., Wang, N., Sun, K., & Shen, D.-G. (2024) · AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i5.28215 · Cited by 5 · see citing works
“Several methods (Zhan et al. 2021; Xing et al. 2022) introduce a gate mergence mechanism to automatically learn the weights of different modalities to enhance the task-related information.”
MethodMeng, R.-Q., Chen, J.-L., Sun, K., Chen, Q., Zhang, X., Dai, L., Gu, Y., Wu, G., & Shen, D. (2025) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2025.3562766 · Cited by 4 · see citing works
“Various methodologies [22], [23] have implemented a gate mergence mechanism to dynamically assign weights to different modalities, emphasizing information pertinent to the taskathand.”
Song, T., Wu, Y., Hu, M., Luo, X., Wei, L., Wang, G., Guo, Y., Xu, F., & Zhang, S. (2026) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2026.3654249
“Fusion strategies like Multi-Scale Gate Mergence [12], attention-based fusion [13], and Confidence-Guided Aggregation [14] further have been explored.”
Lee, S., Jung, J.-H., Choi, Y., Seok, E., Jung, J., Lim, H., Kim, D., & Yun, M.-J. (2024) · Clinical Nuclear Medicine · doi:10.1097/rlu.0000000000005441 · Cited by 4 · see citing works
Guest Editorial Generative Adversarial Networks in Biomedical Image Computing
Fu, H., Zhou, T., Li, S., & Frangi, A. F. (2022) · IEEE journal of biomedical and health informatics · doi:10.1109/jbhi.2021.3134004 · Cited by 4 · see citing works
Jia, N.-N., Zhang, Z., & Jia, T. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.108128 · Cited by 3 · see citing works
ReeGAN: MRI image edge-preserving synthesis based on GANs trained with misaligned data
Lu, X.-J., Liang, X.-S., Liu, W.-J., Miao, X.-X., & Guan, X. (2024) · Medical and Biological Engineering and Computing · doi:10.1007/s11517-024-03035-w · Cited by 3 · see citing works
“In medical imaging, many synthesis methods based on GAN learning [4, 5] have been proposed, which have achieved good performance.”
A layer-wise fusion network incorporating self-supervised learning for multimodal MR image synthesis
Zhou, Q.-Y., & Zou, H. (2022) · Frontiers in Genetics · doi:10.3389/fgene.2022.937042 · Cited by 3 · see citing works
“Feature-level fusion (Zhan et al., 2021; Zuo et al., 2021; Gao et al., 2022b; Roy et al., 2022) extracts modality-specific features of each modality and then fuses them to use the complementary information.”
BackgroundMultiscale Vector-Quantized Variational Autoencoder for Endoscopic Image Synthesis
Diamantis, D. E., & Iakovidis, D. (2025) · International Symposium on Telecommunications · doi:10.1109/ist66504.2025.11268400 · Cited by 2 · see citing works
“Progressive Growing GAN (PGGAN) created dermoscopic images [20], and a multiscale GAN fused MRI modalities for enhanced diagnostic value [21].”
Tao, J., Liu, Z., Ma, M., Qin, J., Liu, F., & Hao, X. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.114587 · Cited by 2 · see citing works
Parker, J. G., Servati, M., Diller, E. E., Cao, S., Ho, C. Y., Lober, R. M., & Cohen‐Gadol, A. (2022) · Medical Physics · doi:10.1002/mp.16059 · Cited by 1 · see citing works
“…images of human brain tumors, 153 improving musculoskeletal mpMR image quality, 154 reconstruction of raw mpMR data, 155 syntheti-zation of additional MR sequence images from already acquired MR data, 156 automatic segmentation of the prostate, 157 and super resolution mpMR images of the brain.”
ADGAN: Adaptive Domain Medical Image Synthesis Based on Generative Adversarial Networks
Xu, L.-M., Lei, Y., Zheng, B.-C., Lv, J.-C., & Li, W. (2024) · CAAI Artificial Intelligence Research · doi:10.26599/air.2024.9150035 · Cited by 2 · see citing works
Diagnostically faithful fMRI-dMRI synthesis using disease-informed diffusion priors
Shen, X., Wang, J., Zhong, Y., Song, Z.-X., Zhao, L.-L., Anter, A., Wei, Y.-C., Liang, L.-Y., Deng, D., Lei, B.-Y., et al. (2026) · Information Fusion · doi:10.1016/j.inffus.2026.104301 · Cited by 1
Cyclic Contrastive Representation Learning for Incomplete Multi-Modal Medical Image Segmentation
He, S., Yang, Z., Zhao, J., Zhang, H., & Wang, R.-X. (2025) · IEEE journal of biomedical and health informatics · doi:10.1109/jbhi.2025.3637570 · Cited by 1 · see citing works
“Subsequently, various generative models [15], [16], [17] have been successively proposed to synthesize missing modalities.”
Liu, Y., Du, S., Xu, Z., Zhang, Z., Wei, Z., Wong, C., Liang, Y., Liu, K., Xu, P., Zhang, L., et al. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.130567 · Cited by 1 · see citing works
Self-citationHigh-resolution conditional MR image synthesis through the PACGAN framework
Lai, M., Marzi, C., Citi, L., & Diciotti, S. (2025) · Scientific Reports · doi:10.1038/s41598-025-16257-1
Wang, C.-X., Zhang, S., Xu, J., Wang, H., Wang, Q.-Z., Zhu, Y.-P., Xing, X.-Y., Hao, D., & Lang, N. (2025) · Academic Radiology · doi:10.1016/j.acra.2025.02.024 · Cited by 1 · see citing works
Liu, X.-F., Prince, J. L., Xing, F.-X., Zhuo, J., Reese, T., Stone, M., Fakhri, G. E., & Woo, J. (2023) · arXiv.org · doi:10.48550/arxiv.2305.14589 · Cited by 1
“Multi-modal neuroimaging is another area, where image modality translation is widely used to overcome the missing modality issue caused by motion during the acquisition process (Chartsias et al., 2017; Gu et al., 2019; Armanious et al., 2020; Zhou et al., 2020; Zhan et al., 2021).”
Missing Information Reconstruction of Land Surface Temperature Data Based on LPRN
Xue, C., Wu, T., Huang, X., & Ashrafzadeh, A. H. (2021) · Mathematical Problems in Engineering · doi:10.1155/2021/4046083 · Cited by 1 · see citing works
Lei, K.-X., Luo, Z.-X., Tang, W.-J., Guo, Y., Kong, Q., Liu, Z.-W., Ding, H.-J., Wang, J.-H., & Huang, B.-S. (2026) · Expert systems with applications · doi:10.1016/j.eswa.2026.134710
Multimodal MRI feature fusion with missing image generation for Alzheimer's disease diagnosis
He, P., Chen, J.-B., Cui, Y.-P., Wu, D., Wang, R.-Y., & Shi, Z. (2026) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2026.110691
Ow-Mamba: modality-adaptive 3D brain tumor segmentation for single- and partial-modality MRI
Zhou, M., Li, J.-B., & Guo, Y.-H. (2026) · The Visual Computer · doi:10.1007/s00371-026-04677-7
A Review of Deep Learning Methods for Multimodal Medical Image Fusion
Shi, Y., & Rigou, M. (2026) · Sensors · doi:10.3390/s26144632
“MGM-GAN [50] constructs its generator based on multiscale convolutional modules and a gating mechanism to integrate more information into the fused image.”
Qiu, D.-F., Wong, K. K. L., Cheng, Y.-H., Yi, Z., & Wang, X.-S. (2026) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2025.3631689
“With the creation of generative adversarial networks (GANs), [28].”
Pham, Q. N., Nguyen, H., Nguyen, H., Le, T.-H.-H., Nguyen, H., & Nguyen, L. (2026) · Imaging Science Journal · doi:10.1080/13682199.2026.2635216
Xu, S.-S., Zhong, W.-L., Wang, Y., & Kuzhati, X. (2025) · 2025 5th International Conference on Mobile Networks and Wireless Communications (ICMNWC) · doi:10.1109/icmnwc66779.2025.11354170
“Then, the feature maps of all image blocks are aggregated into an overall discriminant vector, as shown in Equation (6): ω k is the adaptive weight coefficient [16].”
Self-citationPattern-Aware Diffusion Synthesis of fMRI/dMRI with Tissue and Microstructural Refinement
Shen, X., Wang, J.-Q., Zhong, Y., Song, Z.-X., Zhao, L.-L., Wei, Y.-C., Liang, L.-Y., Wang, S., Lei, B.-Y., Deng, D., et al. (2025) · arXiv.org · doi:10.48550/arxiv.2511.04963
GAN-Driven Generation of Brain MR Images
Vashisth, A., Bisht, A., Singh, C., & Tiwari, D. (2025) · 2025 12th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) · doi:10.1109/icrito66076.2025.11241380
“GANs work by training two neural networks – a Generator and a Discriminator – in an adversarial setup [2].”
D2Diff: A Dual-Domain Diffusion Model for Accurate Multi-Contrast MRI Synthesis
Dayarathna, S., Peiris, H., Islam, K. T., Wong, T.-T., & Chen, Z.-L. (2025) · International Conference on Medical Image Computing and Computer-Assisted Intervention · doi:10.1007/978-3-032-04937-7_13
“Most of the existing methods [5,9,8] focus on fusing features from multiple contrasts, leveraging latent-level operations or hierarchical representations to model cross-contrast dependencies.”
Dual‐way magnetic resonance image translation with transformer‐based adversarial network
Li, W., Xia, J., Gao, W.-L., Hu, Z., Nie, S.-D., & Li, Y. (2025) · Medical Physics (Lancaster) · doi:10.1002/mp.17837
“18,19 The principle of medical image conversion is to preserve the essential content of the source domain while incorporating stylistic features from the target domain.”
D2GAN: A Dual-Domain Generative Adversarial Network for High-Quality PET Image Reconstruction
Jiang, L., Cui, J., Xu, Y., Yan, B.-Y., Zhou, J.-L., Wang, Y., & Shen, D. (2024) · IEEE International Joint Conference on Neural Network · doi:10.1109/ijcnn60899.2024.10650280
“Utilizing multi-scale receptive fields has been demonstrated effective, particularly in capturing a diverse hierarchy of features, which significantly boosts the network performance [18].”
BackgroundSelf-citationRani, P., Ghosh, Halder, S., Mowla, H., & Abdullah, P. D. A. B. M. (2023) · Journal of Primeasia · doi:10.25163/primeasia.4140047
Explainable Multimodal Brain MRI Synthesis Using Deep Generative Models with Grad-CAM
K., A., Saddala, R. R., Singh, T., Kumar, R., & Naik, G. R. (2026) · Procedia Computer Science · doi:10.1016/j.procs.2026.06.417
Zhang, D., Chen, B., Li, Y.-L., Yang, Z.-P., & Chen, C.-Y. (2026) · IEEE Access · doi:10.1109/access.2026.3711764
“Several approaches have been proposed to handle missing modalities, including modality synthesis and shared-representation learning [6].”
GAN-CMA: Causal Mediation Analysis Based on Generative Adversarial Network
Zhang, H.-Y., Li, Q., Yang, Z., Gao, M., Shang, L., & Tan, Z.-J. (2025) · IEEE Access · doi:10.1109/access.2025.3633643
“Zhan et al. [25] enhanced the robustness of missing MRI sequences using a multi-scale gating mechanism.”
- Authors
- Zhan, B., Li, D., Wu, X., Zhou, J., & Wang, Y.
- Affiliations
- Sichuan University (CN)
- Chengdu University of Information Technology (CN)
- Venue
- IEEE Journal of Biomedical and Health Informatics
- Published
- 2022-01
- Volume
- 26
- Issue
- 1
- Pages
- 17-26
- ISSN
- 2168-2194 (ISSN-L)
- 2168-2208
- OpenAlex
- W3166504651
- Cited by
- 90
- References
- 36
- Citations by year
- 2026: 10 · 2025: 23 · 2024: 28 · 2023: 17 · 2022: 10 · 2021: 2
- FWCI
- 5.13
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Journal of Biomedical and Health Informatics (publishedVersion)
- PubMed (publishedVersion)
- Funders
- National Natural Science Foundation of China
- National Stem Cell Foundation (62071314)
- Sichuan Science and Technology Program (2020YFG0079, 2021YFG0326)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Advanced MRI Techniques and Applications
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- multimodal MRI, MRI image synthesis, generative adversarial networks, modal fusion
- MeSH terms
- Humans, Image Processing, Computer-Assisted, Magnetic Resonance Imaging
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
