Ea-GANs: Edge-Aware Generative Adversarial Networks for Cross-Modality MR Image Synthesis

Lightweight cross-resolution coarse-to-fine network for efficient deformable medical image registration

BACKGROUND

Accurate and efficient deformable medical image registration is crucial in medical image analysis. While recent deep learning-based registration methods have achieved state-of-the-art accuracy, they often suffer from extensive network parameters and slow inference times, leading to inefficiency. Efforts to reduce model size and input resolution can improve computational efficiency but frequently result in suboptimal accuracy.

PURPOSE

To address the trade-off between high accuracy and efficiency, we propose a Lightweight Cross-Resolution Coarse-to-Fine registration framework, termed LightCRCF.

METHODS

Our method is built on an ultra-lightweight U-Net architecture with only 0.1 million parameters, offering remarkable efficiency. To mitigate accuracy degradation resulting from fewer parameters while preserving the lightweight nature of the networks, LightCRCF introduces three key innovations as follows: (1) selecting an efficient cross-resolution coarse-to-fine (C2F) registration strategy and integrating it into the lightweight network to progressively decompose the deformation fields into multiresolution subfields to capture fine-grained deformations; (2) a Texture-aware Reparameterization (TaRep) module that integrates Sobel and Laplacian operators to extract rich textural information; (3) a Group-flow Reparameterization (GfRep) module that captures diverse deformation modes by decomposing the deformation field into multiple groups. Furthermore, we introduce a structural reparameterization technique that enhances training accuracy through multibranch structures of the TaRep and GfRep modules, while maintaining efficient inference by equivalently transforming these multibranch structures into single-branch standard convolutions.

RESULTS

We evaluate LightCRCF against various methods on the three public MRI datasets (LPBA, OASIS, and ACDC) and one CT dataset (abdomen CT). Following the previous data division methods, the LPBA dataset comprises 30 training image pairs and nine testing image pairs. For the OASIS dataset, the training, validation, and testing data consist of 1275, 110, and 660 image pairs, respectively. Similarly, for the ACDC dataset, the training, validation, and testing data include 180, 20, and 100 image pairs, respectively. For intersubject registration of the abdomen CT dataset, there are 380 training pairs, six validation pairs, and 42 testing pairs. Compared to state-of-the-art C2F methods, LightCRCF achieves comparable accuracy scores (DSC, HD95, and MSE), while demonstrating significantly superior performance across all efficiency metrics (Params, VRAM, FLOPs, and inference time). Relative to efficiency-first approaches, LightCRCF significantly outperforms these methods in accuracy metrics.

CONCLUSIONS

Our LightCRCF method offers a favorable trade-off between accuracy and efficiency, maintaining high accuracy while achieving superior efficiency, thereby highlighting its potential for clinical applications. The code will be available at https://github.com/PerceptionComputingLab/LightCRCF.

M2OCNN: Many-to-One Collaboration Neural Networks for simultaneously multi-modal medical image synthesis and fusion

BACKGROUND AND OBJECTIVE

Acquiring comprehensive information from multi-modal medical images remains a challenge in clinical diagnostics and treatment, due to complex inter-modal dependencies and missing modalities. While cross-modal medical image synthesis (CMIS) and multi-modal medical image fusion (MMIF) address certain issues, existing methods typically treat these as separate tasks, lacking a unified framework that can generate both synthesized and fused images in the presence of missing modalities.

METHODS

In this paper, we propose the Many-to-One Collaboration Neural Network (M2OCNN), a unified model designed to simultaneously address CMIS and MMIF. Unlike traditional approaches, M2OCNN treats fusion as a specific form of synthesis and provides a comprehensive solution even when modalities are missing. The network consists of three modules: the Parallel Untangling Hybrid Network, Comprehensive Feature Router, and Series Omni-modal Hybrid Network. Additionally, we introduce a mixed-resolution attention mechanism and two transformer variants, Coarsormer and ReCoarsormer, to suppress high-frequency interference and enhance model performance. M2OCNN outperformed state-of-the-art methods on three multi-modal medical imaging datasets, achieving an average PSNR improvement of 2.4 dB in synthesis tasks and producing high-quality fusion images despite missing modalities. The source code is available at https://github.com/zjno108/M2OCNN.

CONCLUSION

M2OCNN offers a novel solution by unifying CMIS and MMIF tasks in a single framework, enabling the generation of both synthesized and fused images from a single modality. This approach sets a new direction for research in multi-modal medical imaging, with implications for improving clinical diagnosis and treatment.

Generative T2*-weighted images as a substitute for true T2*-weighted images on brain MRI in patients with acute stroke

PURPOSE

The purpose of this study was to validate a deep learning algorithm that generates T2*-weighted images from diffusion-weighted (DW) images and to compare its performance with that of true T2*-weighted images for hemorrhage detection on MRI in patients with acute stroke.

MATERIALS AND METHODS

This single-center, retrospective study included DW- and T2*-weighted images obtained less than 48 hours after symptom onset in consecutive patients admitted for acute stroke. Datasets were divided into training (60 %), validation (20 %), and test (20 %) sets, with stratification by stroke type (hemorrhagic/ischemic). A generative adversarial network was trained to produce generative T2*-weighted images using DW images. Concordance between true T2*-weighted images and generative T2*-weighted images for hemorrhage detection was independently graded by two readers into three categories (parenchymal hematoma, hemorrhagic infarct or no hemorrhage), and discordances were resolved by consensus reading. Sensitivity, specificity and accuracy of generative T2*-weighted images were estimated using true T2*-weighted images as the standard of reference.

RESULTS

A total of 1491 MRI sets from 939 patients (487 women, 452 men) with a median age of 71 years (first quartile, 57; third quartile, 81; range: 21-101) were included. In the test set (n = 300), there were no differences between true T2*-weighted images and generative T2*-weighted images for intraobserver reproducibility (κ = 0.97 [95 % CI: 0.95-0.99] vs. 0.95 [95 % CI: 0.92-0.97]; P = 0.27) and interobserver reproducibility (κ = 0.93 [95 % CI: 0.90-0.97] vs. 0.92 [95 % CI: 0.88-0.96]; P = 0.64). After consensus reading, concordance between true T2*-weighted images and generative T2*-weighted images was excellent (κ = 0.92; 95 % CI: 0.91-0.96). Generative T2*-weighted images achieved 90 % sensitivity (73/81; 95 % CI: 81-96), 97 % specificity (213/219; 95 % CI: 94-99) and 95 % accuracy (286/300; 95 % CI: 92-97) for the diagnosis of any cerebral hemorrhage (hemorrhagic infarct or parenchymal hemorrhage).

CONCLUSION

Generative T2*-weighted images and true T2*-weighted images have non-different diagnostic performances for hemorrhage detection in patients with acute stroke and may be used to shorten MRI protocols.

RTF: Recursive TransFusion for Multi-Modal Image Synthesis

Multi-modal image synthesis is crucial for obtaining complete modalities due to the imaging restrictions in reality. Current methods, primarily CNN-based models, find it challenging to extract global representations because of local inductive bias, leading to synthetic structure deformation or color distortion. Despite the significant global representation ability of transformer in capturing long-range dependencies, its huge parameter size requires considerable training data. Multi-modal synthesis solely based on one of the two structures makes it hard to extract comprehensive information from each modality with limited data. To tackle this dilemma, we propose a simple yet effective Recursive TransFusion (RTF) framework for multi-modal image synthesis. Specifically, we develop a TransFusion unit to integrate local knowledge extracted from the individual modality by connecting a CNN-based local representation block (LRB) and a transformer-based global fusion block (GFB) via a feature translating gate (FTG). Considering the numerous parameters introduced by the transformer, we further unfold a TransFusion unit with recursive constraint repeatedly, forming recursive TransFusion (RTF), which progressively extracts multi-modal information at different depths. Our RTF remarkably reduces network parameters while maintaining superior performance. Extensive experiments validate our superiority against the competing methods on multiple benchmarks. The source code will be available at https://github.com/guoliangq/RTF.

3D Wasserstein Generative Adversarial Network with Dense U-Net-Based Discriminator for Preclinical fMRI Denoising

Functional magnetic resonance imaging (fMRI) is extensively used in clinical and preclinical settings to study brain function; however, fMRI data is inherently noisy due to physiological processes, hardware, and external noise. Denoising is one of the main preprocessing steps in any fMRI analysis pipeline. This process is challenging in preclinical data in comparison to clinical data due to variations in brain geometry, image resolution, and low signal-to-noise ratios. In this paper, we propose a structure-preserved algorithm based on a 3D Wasserstein generative adversarial network with a 3D dense U-net-based discriminator called 3D U-WGAN. We apply a 4D data configuration to effectively denoise temporal and spatial information in analyzing preclinical fMRI data. GAN-based denoising methods often utilize a discriminator to identify significant differences between denoised and noise-free images, focusing on global or local features. To refine the fMRI denoising model, our method employs a 3D dense U-Net discriminator to learn both global and local distinctions. To tackle potential oversmoothing, we introduce an adversarial loss and enhance perceptual similarity by measuring feature space distances. Experiments illustrate that 3D U-WGAN significantly improves image quality in resting-state and task preclinical fMRI data, enhancing signal-to-noise ratio without introducing excessive structural changes in existing methods. The proposed method outperforms state-of-the-art methods when applied to simulated and real data in a fMRI analysis pipeline.

BreathVisionNet: A pulmonary-function-guided CNN-transformer hybrid model for expiratory CT image synthesis

BACKGROUND AND OBJECTIVE

Chronic obstructive pulmonary disease (COPD) has high heterogeneity in etiologies and clinical manifestations. Expiratory Computed tomography (CT) can effectively assess air trapping, aiding in disease diagnosis. However, due to concerns about radiation exposure and cost, expiratory CT is not routinely performed. Recent work on synthesizing expiratory CT has primarily focused on imaging features while neglecting patient-specific pulmonary function.

METHODS

To address these issues, we developed a novel model named BreathVisionNet that incorporates pulmonary function data to guide the synthesis of expiratory CT from inspiratory CT. An architecture combining a convolutional neural network and transformer is introduced to leverage the irregular phenotypic distribution in COPD patients. The model can better understand the long-range and global contexts by incorporating global information into the encoder. The utilization of edge information and multi-view data further enhances the quality of the synthesized CT. Parametric response mapping (PRM) can be estimated by using synthesized expiratory CT and inspiratory CT to quantify COPD phenotypes of the normal, emphysema, and functional small airway disease (fSAD), including their percentages, spatial distributions, and voxel distribution maps.

RESULTS

BreathVisionNet outperforms other generative models in terms of synthesized image quality. It achieves a mean absolute error, normalized mean square error, structural similarity index and peak signal-to-noise ratio of 78.207 HU, 0.643, 0.847 and 25.828 dB, respectively. Comparing the predicted and real PRM, the Dice coefficient can reach 0.732 (emphysema) and 0.560 (fSAD). The mean of differences between true and predicted fSAD percentage is 4.42 for the development dataset (low radiation dose CT scans), and 9.05 for an independent external validation dataset (routine dose), indicating that model has great generalizability. A classifier trained on voxel distribution maps can achieve an accuracy of 0.891 in predicting the presence of COPD.

CONCLUSIONS

BreathVisionNet can accurately synthesize expiratory CT images from inspiratory CT and predict their voxel distribution. The estimated PRM can help to quantify COPD phenotypes of the normal, emphysema, and fSAD. This capability provides additional insights into COPD diversity while only inspiratory CT images are available.

NCCT-to-CECT synthesis with contrast-enhanced knowledge and anatomical perception for multi-organ segmentation in non-contrast CT images

Contrast-enhanced computed tomography (CECT) is constantly used for delineating organs-at-risk (OARs) in radiation therapy planning. The delineated OARs are needed to transfer from CECT to non-contrast CT (NCCT) for dose calculation. Yet, the use of iodinated contrast agents (CA) in CECT and the dose calculation errors caused by the spatial misalignment between NCCT and CECT images pose risks of adverse side effects. A promising solution is synthesizing CECT images from NCCT scans, which can improve the visibility of organs and abnormalities for more effective multi-organ segmentation in NCCT images. However, existing methods neglect the difference between tissues induced by CA and lack the ability to synthesize the details of organ edges and blood vessels. To address these issues, we propose a contrast-enhanced knowledge and anatomical perception network (CKAP-Net) for NCCT-to-CECT synthesis. CKAP-Net leverages a contrast-enhanced knowledge learning network to capture both similarities and dissimilarities in domain characteristics attributable to CA. Specifically, a CA-based perceptual loss function is introduced to enhance the synthesis of CA details. Furthermore, we design a multi-scale anatomical perception transformer that utilizes multi-scale anatomical information from NCCT images, enabling the precise synthesis of tissue details. Our CKAP-Net is evaluated on a multi-center abdominal NCCT-CECT dataset, a head an neck NCCT-CECT dataset, and an NCMRI-CEMRI dataset. It achieves a MAE of 25.96 ± 2.64, a SSIM of 0.855 ± 0.017, and a PSNR of 32.60 ± 0.02 for CECT synthesis, and a DSC of 81.21 ± 4.44 for segmentation on the internal dataset. Extensive experiments demonstrate that CKAP-Net outperforms state-of-the-art CA synthesis methods and has better generalizability across different datasets.

Bridging MRI Cross-Modality Synthesis and Multi-Contrast Super-Resolution by Fine-Grained Difference Learning

In multi-modal magnetic resonance imaging (MRI), the tasks of imputing or reconstructing the target modality share a common obstacle: the accurate modeling of fine-grained inter-modal differences, which has been sparingly addressed in current literature. These differences stem from two sources: 1) spatial misalignment remaining after coarse registration and 2) structural distinction arising from modality-specific signal manifestations. This paper integrates the previously separate research trajectories of cross-modality synthesis (CMS) and multi-contrast super-resolution (MCSR) to address this pervasive challenge within a unified framework. Connected through generalized down-sampling ratios, this unification not only emphasizes their common goal in reducing structural differences, but also identifies the key task distinguishing MCSR from CMS: modeling the structural distinctions using the limited information from the misaligned target input. Specifically, we propose a composite network architecture with several key components: a label correction module to align the coordinates of multi-modal training pairs, a CMS module serving as the base model, an SR branch to handle target inputs, and a difference projection discriminator for structural distinction-centered adversarial training. When training the SR branch as the generator, the adversarial learning is enhanced with distinction-aware incremental modulation to ensure better-controlled generation. Moreover, the SR branch integrates deformable convolutions to address cross-modal spatial misalignment at the feature level. Experiments conducted on three public datasets demonstrate that our approach effectively balances structural accuracy and realism, exhibiting overall superiority in comprehensive evaluations for both tasks over current state-of-the-art approaches. The code is available at https://github.com/papshare/FGDL.

Texture-preserving diffusion model for CBCT-to-CT synthesis

Cone beam computed tomography (CBCT) serves as a vital imaging modality in diverse clinical applications, but is constrained by inherent limitations such as reduced image quality and increased noise. In contrast, computed tomography (CT) offers superior resolution and tissue contrast. Bridging the gap between these modalities through CBCT-to-CT synthesis becomes imperative. Deep learning techniques have enhanced this synthesis, yet challenges with generative adversarial networks persist. Denoising Diffusion Probabilistic Models have emerged as a promising alternative in image synthesis. In this study, we propose a novel texture-preserving diffusion model for CBCT-to-CT synthesis that incorporates adaptive high-frequency optimization and a dual-mode feature fusion module. Our method aims to enhance high-frequency details, effectively fuse cross-modality features, and preserve fine image structures. Extensive validation demonstrates superior performance over existing methods, showcasing better generalization. The proposed model offers a transformative pathway to augment diagnostic accuracy and refine treatment planning across various clinical settings. This work represents a pivotal step toward non-invasive, safer, and high-quality CBCT-to-CT synthesis, advancing personalized diagnostic imaging practices.

Prior Knowledge-Guided Triple-Domain Transformer-GAN for Direct PET Reconstruction From Low-Count Sinograms

To obtain high-quality positron emission tomography (PET) images while minimizing radiation exposure, numerous methods have been dedicated to acquiring standard-count PET (SPET) from low-count PET (LPET). However, current methods have failed to take full advantage of the different emphasized information from multiple domains, i.e., the sinogram, image, and frequency domains, resulting in the loss of crucial details. Meanwhile, they overlook the unique inner-structure of the sinograms, thereby failing to fully capture its structural characteristics and relationships. To alleviate these problems, in this paper, we proposed a prior knowledge-guided transformer-GAN that unites triple domains of sinogram, image, and frequency to directly reconstruct SPET images from LPET sinograms, namely PK-TriDo. Our PK-TriDo consists of a Sinogram Inner-Structure-based Denoising Transformer (SISD-Former) to denoise the input LPET sinogram, a Frequency-adapted Image Reconstruction Transformer (FaIR-Former) to reconstruct high-quality SPET images from the denoised sinograms guided by the image domain prior knowledge, and an Adversarial Network (AdvNet) to further enhance the reconstruction quality via adversarial training. Specifically tailored for the PET imaging mechanism, we injected a sinogram embedding module that partitions the sinograms by rows and columns to obtain 1D sequences of angles and distances to faithfully preserve the inner-structure of the sinograms. Moreover, to mitigate high-frequency distortions and enhance reconstruction details, we integrated global-local frequency parsers (GLFPs) into FaIR-Former to calibrate the distributions and proportions of different frequency bands, thus compelling the network to preserve high-frequency details. Evaluations on three datasets with different dose levels and imaging scenarios demonstrated that our PK-TriDo outperforms the state-of-the-art methods.

AI-ADC: Channel and Spatial Attention-Based Contrastive Learning to Generate ADC Maps from T2W MRI for Prostate Cancer Detection

BACKGROUND/OBJECTIVES

Apparent Diffusion Coefficient (ADC) maps in prostate MRI can reveal tumor characteristics, but their accuracy can be compromised by artifacts related with patient motion or rectal gas associated distortions. To address these challenges, we propose a novel approach that utilizes a Generative Adversarial Network to synthesize ADC maps from T2-weighted magnetic resonance images (T2W MRI).

METHODS

By leveraging contrastive learning, our model accurately maps axial T2W MRI to ADC maps within the cropped region of the prostate organ boundary, capturing subtle variations and intricate structural details by learning similar and dissimilar pairs from two imaging modalities. We trained our model on a comprehensive dataset of unpaired T2-weighted images and ADC maps from 506 patients. In evaluating our model, named AI-ADC, we compared it against three state-of-the-art methods: CycleGAN, CUT, and StyTr2.

RESULTS

Our model demonstrated a higher mean Structural Similarity Index (SSIM) of 0.863 on a test dataset of 3240 2D MRI slices from 195 patients, compared to values of 0.855, 0.797, and 0.824 for CycleGAN, CUT, and StyTr2, respectively. Similarly, our model achieved a significantly lower Fréchet Inception Distance (FID) value of 31.992, compared to values of 43.458, 179.983, and 58.784 for the other three models, indicating its superior performance in generating ADC maps. Furthermore, we evaluated our model on 147 patients from the publicly available ProstateX dataset, where it demonstrated a higher SSIM of 0.647 and a lower FID of 113.876 compared to the other three models.

CONCLUSIONS

These results highlight the efficacy of our proposed model in generating ADC maps from T2W MRI, showcasing its potential for enhancing clinical diagnostics and radiological workflows.

PST-Diff: Achieving High-Consistency Stain Transfer by Diffusion Models With Pathological and Structural Constraints

Histopathological examinations heavily rely on hematoxylin and eosin (HE) and immunohistochemistry (IHC) staining. IHC staining can offer more accurate diagnostic details but it brings significant financial and time costs. Furthermore, either re-staining HE-stained slides or using adjacent slides for IHC may compromise the accuracy of pathological diagnosis due to information loss. To address these challenges, we develop PST-Diff, a method for generating virtual IHC images from HE images based on diffusion models, which allows pathologists to simultaneously view multiple staining results from the same tissue slide. To maintain the pathological consistency of the stain transfer, we propose the asymmetric attention mechanism (AAM) and latent transfer (LT) module in PST-Diff. Specifically, the AAM can retain more local pathological information of the source domain images, while ensuring the model's flexibility in generating virtual stained images that highly confirm to the target domain. Subsequently, the LT module transfers the implicit representations across different domains, effectively alleviating the bias introduced by direct connection and further enhancing the pathological consistency of PST-Diff. Furthermore, to maintain the structural consistency of the stain transfer, the conditional frequency guidance (CFG) module is proposed to precisely control image generation and preserve structural details according to the frequency recovery process. To conclude, the pathological and structural consistency constraints provide PST-Diff with effectiveness and superior generalization in generating stable and functionally pathological IHC images with the best evaluation score. In general, PST-Diff offers prospective application in clinical virtual staining and pathological image analysis.

Integrated diagnosis of glioma based on magnetic resonance images with incomplete ground truth labels

BACKGROUND

Since the 2016 WHO guidelines, glioma diagnosis has entered an era of integrated diagnosis, combining tissue pathology and molecular pathology. The WHO has focused on promoting the application of molecular diagnosis in the classification of central nervous system tumors. Genetic information such as IDH1 and 1p/19q are important molecular markers, and pathological grading is also a key clinical indicator. However, obtaining genetic pathology labels is more costly than conventional MRI images, resulting in a large number of missing labels in realistic modeling.

METHOD

We propose a training strategy based on label encoding and a corresponding loss function to enable the model to effectively utilize data with missing labels. Additionally, we integrate a graph model with genes and pathology-related clinical prior knowledge into the ResNet backbone to further improve the efficacy of diagnosis. Ten-fold cross-validation experiments were conducted on a large dataset of 1072 patients.

RESULTS

The classification area under the curve (AUC) values are 0.93, 0.91, and 0.90 for IDH1, 1p/19q status, and grade (LGG/HGG), respectively. When the label miss rate reached 59.3 %, the method improved the AUC by 0.09, 0.10, and 0.04 for IDH1, 1p/19q, and pathological grade, respectively, compared to the same backbone without the missing label strategy.

CONCLUSIONS

Our method effectively utilizes data with missing labels and integrates clinical prior knowledge, resulting in improved diagnostic performance for glioma genetic and pathological markers, even with high rates of missing labels.

LungViT: Ensembling Cascade of Texture Sensitive Hierarchical Vision Transformers for Cross-Volume Chest CT Image-to-Image Translation

Chest computed tomography (CT) at inspiration is often complemented by an expiratory CT to identify peripheral airways disease. Additionally, co-registered inspiratory-expiratory volumes can be used to derive various markers of lung function. Expiratory CT scans, however, may not be acquired due to dose or scan time considerations or may be inadequate due to motion or insufficient exhale; leading to a missed opportunity to evaluate underlying small airways disease. Here, we propose LungViT- a generative adversarial learning approach using hierarchical vision transformers for translating inspiratory CT intensities to corresponding expiratory CT intensities. LungViT addresses several limitations of the traditional generative models including slicewise discontinuities, limited size of generated volumes, and their inability to model texture transfer at volumetric level. We propose a shifted-window hierarchical vision transformer architecture with squeeze-and-excitation decoder blocks for modeling dependencies between features. We also propose a multiview texture similarity distance metric for texture and style transfer in 3D. To incorporate global information into the training process and refine the output of our model, we use ensemble cascading. LungViT is able to generate large 3D volumes of size 320×320×320 . We train and validate our model using a diverse cohort of 1500 subjects with varying disease severity. To assess model generalizability beyond the development set biases, we evaluate our model on an out-of-distribution external validation set of 200 subjects. Clinical validation on internal and external testing sets shows that synthetic volumes could be reliably adopted for deriving clinical endpoints of chronic obstructive pulmonary disease.

Multi-Modal Modality-Masked Diffusion Network for Brain MRI Synthesis With Random Modality Missing

Synthesis of unavailable imaging modalities from available ones can generate modality-specific complementary information and enable multi-modality based medical images diagnosis or treatment. Existing generative methods for medical image synthesis are usually based on cross-modal translation between acquired and missing modalities. These methods are usually dedicated to specific missing modality and perform synthesis in one shot, which cannot deal with varying number of missing modalities flexibly and construct the mapping across modalities effectively. To address the above issues, in this paper, we propose a unified Multi-modal Modality-masked Diffusion Network (M2DN), tackling multi-modal synthesis from the perspective of "progressive whole-modality inpainting", instead of "cross-modal translation". Specifically, our M2DN considers the missing modalities as random noise and takes all the modalities as a unity in each reverse diffusion step. The proposed joint synthesis scheme performs synthesis for the missing modalities and self-reconstruction for the available ones, which not only enables synthesis for arbitrary missing scenarios, but also facilitates the construction of common latent space and enhances the model representation ability. Besides, we introduce a modality-mask scheme to encode availability status of each incoming modality explicitly in a binary mask, which is adopted as condition for the diffusion model to further enhance the synthesis performance of our M2DN for arbitrary missing scenarios. We carry out experiments on two public brain MRI datasets for synthesis and downstream segmentation tasks. Experimental results demonstrate that our M2DN outperforms the state-of-the-art models significantly and shows great generalizability for arbitrary missing modalities.

ReeGAN: MRI image edge-preserving synthesis based on GANs trained with misaligned data

As a crucial medical examination technique, different modalities of magnetic resonance imaging (MRI) complement each other, offering multi-angle and multi-dimensional insights into the body's internal information. Therefore, research on MRI cross-modality conversion is of great significance, and many innovative techniques have been explored. However, most methods are trained on well-aligned data, and the impact of misaligned data has not received sufficient attention. Additionally, many methods focus on transforming the entire image and ignore crucial edge information. To address these challenges, we propose a generative adversarial network based on multi-feature fusion, which effectively preserves edge information while training on noisy data. Notably, we consider images with limited range random transformations as noisy labels and use an additional small auxiliary registration network to help the generator adapt to the noise distribution. Moreover, we inject auxiliary edge information to improve the quality of synthesized target modality images. Our goal is to find the best solution for cross-modality conversion. Comprehensive experiments and ablation studies demonstrate the effectiveness of the proposed method.

Deformable registration of preoperative MR and intraoperative long-length tomosynthesis images for guidance of spine surgery via image synthesis

PURPOSE

Improved integration and use of preoperative imaging during surgery hold significant potential for enhancing treatment planning and instrument guidance through surgical navigation. Despite its prevalent use in diagnostic settings, MR imaging is rarely used for navigation in spine surgery. This study aims to leverage MR imaging for intraoperative visualization of spine anatomy, particularly in cases where CT imaging is unavailable or when minimizing radiation exposure is essential, such as in pediatric surgery.

METHODS

This work presents a method for deformable 3D-2D registration of preoperative MR images with a novel intraoperative long-length tomosynthesis imaging modality (viz., Long-Film [LF]). A conditional generative adversarial network is used to translate MR images to an intermediate bone image suitable for registration, followed by a model-based 3D-2D registration algorithm to deformably map the synthesized images to LF images. The algorithm's performance was evaluated on cadaveric specimens with implanted markers and controlled deformation, and in clinical images of patients undergoing spine surgery as part of a large-scale clinical study on LF imaging.

RESULTS

The proposed method yielded a median 2D projection distance error of 2.0 mm (interquartile range [IQR]: 1.1-3.3 mm) and a 3D target registration error of 1.5 mm (IQR: 0.8-2.1 mm) in cadaver studies. Notably, the multi-scale approach exhibited significantly higher accuracy compared to rigid solutions and effectively managed the challenges posed by piecewise rigid spine deformation. The robustness and consistency of the method were evaluated on clinical images, yielding no outliers on vertebrae without surgical instrumentation and 3% outliers on vertebrae with instrumentation.

CONCLUSIONS

This work constitutes the first reported approach for deformable MR to LF registration based on deep image synthesis. The proposed framework provides access to the preoperative annotations and planning information during surgery and enables surgical navigation within the context of MR images and/or dual-plane LF images.

Multi-sequence generative adversarial network: better generation for enhanced magnetic resonance imaging images

Introduction

MRI is one of the commonly used diagnostic methods in clinical practice, especially in brain diseases. There are many sequences in MRI, but T1CE images can only be obtained by using contrast agents. Many patients (such as cancer patients) must undergo alignment of multiple MRI sequences for diagnosis, especially the contrast-enhanced magnetic resonance sequence. However, some patients such as pregnant women, children, etc. find it difficult to use contrast agents to obtain enhanced sequences, and contrast agents have many adverse reactions, which can pose a significant risk. With the continuous development of deep learning, the emergence of generative adversarial networks makes it possible to extract features from one type of image to generate another type of image.

Methods

We propose a generative adversarial network model with multimodal inputs and end-to-end decoding based on the pix2pix model. For the pix2pix model, we used four evaluation metrics: NMSE, RMSE, SSIM, and PNSR to assess the effectiveness of our generated model.

Results

Through statistical analysis, we compared our proposed new model with pix2pix and found significant differences between the two. Our model outperformed pix2pix, with higher SSIM and PNSR, lower NMSE and RMSE. We also found that the input of T1W images and T2W images had better effects than other combinations, providing new ideas for subsequent work on generating magnetic resonance enhancement sequence images. By using our model, it is possible to generate magnetic resonance enhanced sequence images based on magnetic resonance non-enhanced sequence images.

Discussion

This has significant implications as it can greatly reduce the use of contrast agents to protect populations such as pregnant women and children who are contraindicated for contrast agents. Additionally, contrast agents are relatively expensive, and this generation method may bring about substantial economic benefits.

One model to unite them all: Personalized federated learning of multi-contrast MRI synthesis

Curation of large, diverse MRI datasets via multi-institutional collaborations can help improve learning of generalizable synthesis models that reliably translate source- onto target-contrast images. To facilitate collaborations, federated learning (FL) adopts decentralized model training while mitigating privacy concerns by avoiding sharing of imaging data. However, conventional FL methods can be impaired by the inherent heterogeneity in the data distribution, with domain shifts evident within and across imaging sites. Here we introduce the first personalized FL method for MRI Synthesis (pFLSynth) that improves reliability against data heterogeneity via model specialization to individual sites and synthesis tasks (i.e., source-target contrasts). To do this, pFLSynth leverages an adversarial model equipped with novel personalization blocks that control the statistics of generated feature maps across the spatial/channel dimensions, given latent variables specific to sites and tasks. To further promote communication efficiency and site specialization, partial network aggregation is employed over later generator stages while earlier generator stages and the discriminator are trained locally. As such, pFLSynth enables multi-task training of multi-site synthesis models with high generalization performance across sites and tasks. Comprehensive experiments demonstrate the superior performance and reliability of pFLSynth in MRI synthesis against prior federated methods.

Cross2SynNet: cross-device-cross-modal synthesis of routine brain MRI sequences from CT with brain lesion

OBJECTIVES

CT and MR are often needed to determine the location and extent of brain lesions collectively to improve diagnosis. However, patients with acute brain diseases cannot complete the MRI examination within a short time. The aim of the study is to devise a cross-device and cross-modal medical image synthesis (MIS) method Cross2SynNet for synthesizing routine brain MRI sequences of T1WI, T2WI, FLAIR, and DWI from CT with stroke and brain tumors.

MATERIALS AND METHODS

For the retrospective study, the participants covered four different diseases of cerebral ischemic stroke (CIS-cohort), cerebral hemorrhage (CH-cohort), meningioma (M-cohort), glioma (G-cohort). The MIS model Cross2SynNet was established on the basic architecture of conditional generative adversarial network (CGAN), of which, the fully convolutional Transformer (FCT) module was adopted into generator to capture the short- and long-range dependencies between healthy and pathological tissues, and the edge loss function was to minimize the difference in gradient magnitude between synthetic image and ground truth. Three metrics of mean square error (MSE), peak signal-to-noise ratio (PSNR), and structure similarity index measure (SSIM) were used for evaluation.

RESULTS

A total of 230 participants (mean patient age, 59.77 years ± 13.63 [standard deviation]; 163 men [71%] and 67 women [29%]) were included, including CIS-cohort (95 participants between Dec 2019 and Feb 2022), CH-cohort (69 participants between Jan 2020 and Dec 2021), M-cohort (40 participants between Sep 2018 and Dec 2021), and G-cohort (26 participants between Sep 2019 and Dec 2021). The Cross2SynNet achieved averaged values of MSE = 0.008, PSNR = 21.728, and SSIM = 0.758 when synthesizing MRIs from CT, outperforming the CycleGAN, pix2pix, RegGAN, Pix2PixHD, and ResViT. The Cross2SynNet could synthesize the brain lesion on pseudo DWI even if the CT image did not exhibit clear signal in the acute ischemic stroke patients.

CONCLUSIONS

Cross2SynNet could achieve routine brain MRI synthesis of T1WI, T2WI, FLAIR, and DWI from CT with promising performance given the brain lesion of stroke and brain tumor.

Contrast-enhanced to non-contrast-enhanced image translation to exploit a clinical data warehouse of T1-weighted brain MRI

BACKGROUND

Clinical data warehouses provide access to massive amounts of medical images, but these images are often heterogeneous. They can for instance include images acquired both with or without the injection of a gadolinium-based contrast agent. Harmonizing such data sets is thus fundamental to guarantee unbiased results, for example when performing differential diagnosis. Furthermore, classical neuroimaging software tools for feature extraction are typically applied only to images without gadolinium. The objective of this work is to evaluate how image translation can be useful to exploit a highly heterogeneous data set containing both contrast-enhanced and non-contrast-enhanced images from a clinical data warehouse.

METHODS

We propose and compare different 3D U-Net and conditional GAN models to convert contrast-enhanced T1-weighted (T1ce) into non-contrast-enhanced (T1nce) brain MRI. These models were trained using 230 image pairs and tested on 77 image pairs from the clinical data warehouse of the Greater Paris area.

RESULTS

Validation using standard image similarity measures demonstrated that the similarity between real and synthetic T1nce images was higher than between real T1nce and T1ce images for all the models compared. The best performing models were further validated on a segmentation task. We showed that tissue volumes extracted from synthetic T1nce images were closer to those of real T1nce images than volumes extracted from T1ce images.

CONCLUSION

We showed that deep learning models initially developed with research quality data could synthesize T1nce from T1ce images of clinical quality and that reliable features could be extracted from the synthetic images, thus demonstrating the ability of such methods to help exploit a data set coming from a clinical data warehouse.

Scalable Swin Transformer network for brain tumor segmentation from incomplete MRI modalities

BACKGROUND

Deep learning methods have shown great potential in processing multi-modal Magnetic Resonance Imaging (MRI) data, enabling improved accuracy in brain tumor segmentation. However, the performance of these methods can suffer when dealing with incomplete modalities, which is a common issue in clinical practice. Existing solutions, such as missing modality synthesis, knowledge distillation, and architecture-based methods, suffer from drawbacks such as long training times, high model complexity, and poor scalability.

METHOD

This paper proposes IMS2Trans, a novel lightweight scalable Swin Transformer network by utilizing a single encoder to extract latent feature maps from all available modalities. This unified feature extraction process enables efficient information sharing and fusion among the modalities, resulting in efficiency without compromising segmentation performance even in the presence of missing modalities.

RESULTS

Two datasets, BraTS 2018 and BraTS 2020, containing incomplete modalities for brain tumor segmentation are evaluated against popular benchmarks. On the BraTS 2018 dataset, our model achieved higher average Dice similarity coefficient (DSC) scores for the whole tumor, tumor core, and enhancing tumor regions (86.57, 75.67, and 58.28, respectively), in comparison with a state-of-the-art model, i.e. mmFormer (86.45, 75.51, and 57.79, respectively). Similarly, on the BraTS 2020 dataset, our model scored higher DSC scores in these three brain tumor regions (87.33, 79.09, and 62.11, respectively) compared to mmFormer (86.17, 78.34, and 60.36, respectively). We also conducted a Wilcoxon test on the experimental results, and the generated p-value confirmed that our model's performance was statistically significant. Moreover, our model exhibits significantly reduced complexity with only 4.47 M parameters, 121.89G FLOPs, and a model size of 77.13 MB, whereas mmFormer comprises 34.96 M parameters, 265.79 G FLOPs, and a model size of 559.74 MB. These indicate our model, being light-weighted with significantly reduced parameters, is still able to achieve better performance than a state-of-the-art model.

CONCLUSION

By leveraging a single encoder for processing the available modalities, IMS2Trans offers notable scalability advantages over methods that rely on multiple encoders. This streamlined approach eliminates the need for maintaining separate encoders for each modality, resulting in a lightweight and scalable network architecture. The source code of IMS2Trans and the associated weights are both publicly available at https://github.com/hudscomdz/IMS2Trans.

MTFN: multi-temporal feature fusing network with co-attention for DCE-MRI synthesis

BACKGROUND

Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) plays an important role in the diagnosis and treatment of breast cancer. However, obtaining complete eight temporal images of DCE-MRI requires a long scanning time, which causes patients' discomfort in the scanning process. Therefore, to reduce the time, the multi temporal feature fusing neural network with Co-attention (MTFN) is proposed to generate the eighth temporal images of DCE-MRI, which enables the acquisition of DCE-MRI images without scanning. In order to reduce the time, multi-temporal feature fusion cooperative attention mechanism neural network (MTFN) is proposed to generate the eighth temporal images of DCE-MRI, which enables DCE-MRI image acquisition without scanning.

METHODS

In this paper, we propose multi temporal feature fusing neural network with Co-attention (MTFN) for DCE-MRI Synthesis, in which the Co-attention module can fully fuse the features of the first and third temporal image to obtain the hybrid features. The Co-attention explore long-range dependencies, not just relationships between pixels. Therefore, the hybrid features are more helpful to generate the eighth temporal images.

RESULTS

We conduct experiments on the private breast DCE-MRI dataset from hospitals and the multi modal Brain Tumor Segmentation Challenge2018 dataset (BraTs2018). Compared with existing methods, the experimental results of our method show the improvement and our method can generate more realistic images. In the meanwhile, we also use synthetic images to classify the molecular typing of breast cancer that the accuracy on the original eighth time-series images and the generated images are 89.53% and 92.46%, which have been improved by about 3%, and the classification results verify the practicability of the synthetic images.

CONCLUSIONS

The results of subjective evaluation and objective image quality evaluation indicators show the effectiveness of our method, which can obtain comprehensive and useful information. The improvement of classification accuracy proves that the images generated by our method are practical.

Contrastive image adaptation for acquisition shift reduction in medical imaging

The domain shift, or acquisition shift in medical imaging, is responsible for potentially harmful differences between development and deployment conditions of medical image analysis techniques. There is a growing need in the community for advanced methods that could mitigate this issue better than conventional approaches. In this paper, we consider configurations in which we can expose a learning-based pixel level adaptor to a large variability of unlabeled images during its training, i.e. sufficient to span the acquisition shift expected during the training or testing of a downstream task model. We leverage the ability of convolutional architectures to efficiently learn domain-agnostic features and train a many-to-one unsupervised mapping between a source collection of heterogeneous images from multiple unknown domains subjected to the acquisition shift and a homogeneous subset of this source set of lower cardinality, potentially constituted of a single image. To this end, we propose a new cycle-free image-to-image architecture based on a combination of three loss functions : a contrastive PatchNCE loss, an adversarial loss and an edge preserving loss allowing for rich domain adaptation to the target image even under strong domain imbalance and low data regimes. Experiments support the interest of the proposed contrastive image adaptation approach for the regularization of downstream deep supervised segmentation and cross-modality synthesis models.

Deep learning based synthesis of MRI, CT and PET: Review and analysis

Medical image synthesis represents a critical area of research in clinical decision-making, aiming to overcome the challenges associated with acquiring multiple image modalities for an accurate clinical workflow. This approach proves beneficial in estimating an image of a desired modality from a given source modality among the most common medical imaging contrasts, such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and Positron Emission Tomography (PET). However, translating between two image modalities presents difficulties due to the complex and non-linear domain mappings. Deep learning-based generative modelling has exhibited superior performance in synthetic image contrast applications compared to conventional image synthesis methods. This survey comprehensively reviews deep learning-based medical imaging translation from 2018 to 2023 on pseudo-CT, synthetic MR, and synthetic PET. We provide an overview of synthetic contrasts in medical imaging and the most frequently employed deep learning networks for medical image synthesis. Additionally, we conduct a detailed analysis of each synthesis method, focusing on their diverse model designs based on input domains and network architectures. We also analyse novel network architectures, ranging from conventional CNNs to the recent Transformer and Diffusion models. This analysis includes comparing loss functions, available datasets and anatomical regions, and image quality assessments and performance in other downstream tasks. Finally, we discuss the challenges and identify solutions within the literature, suggesting possible future directions. We hope that the insights offered in this survey paper will serve as a valuable roadmap for researchers in the field of medical image synthesis.

Correcting synthetic MRI contrast-weighted images using deep learning

Synthetic magnetic resonance imaging (MRI) offers a scanning paradigm where a fast multi-contrast sequence can be used to estimate underlying quantitative tissue parameter maps, which are then used to synthesize any desirable clinical contrast by retrospectively changing scan parameters in silico. Two benefits of this approach are the reduced exam time and the ability to generate arbitrary contrasts offline. However, synthetically generated contrasts are known to deviate from the contrast of experimental scans. The reason for contrast mismatch is the necessary exclusion of some unmodeled physical effects such as partial voluming, diffusion, flow, susceptibility, magnetization transfer, and more. The inclusion of these effects in signal encoding would improve the synthetic images, but would make the quantitative imaging protocol impractical due to long scan times. Therefore, in this work, we propose a novel deep learning approach that generates a multiplicative correction term to capture unmodeled effects and correct the synthetic contrast images to better match experimental contrasts for arbitrary scan parameters. The physics inspired deep learning model implicitly accounts for some unmodeled physical effects occurring during the scan. As a proof of principle, we validate our approach on synthesizing arbitrary inversion recovery fast spin-echo scans using a commercially available 2D multi-contrast sequence. We observe that the proposed correction visually and numerically reduces the mismatch with experimentally collected contrasts compared to conventional synthetic MRI. Finally, we show results of a preliminary reader study and find that the proposed method statistically significantly improves in contrast and SNR as compared to synthetic MR images.

3D multi-modality Transformer-GAN for high-quality PET reconstruction

Positron emission tomography (PET) scans can reveal abnormal metabolic activities of cells and provide favorable information for clinical patient diagnosis. Generally, standard-dose PET (SPET) images contain more diagnostic information than low-dose PET (LPET) images but higher-dose scans can also bring higher potential radiation risks. To reduce the radiation risk while acquiring high-quality PET images, in this paper, we propose a 3D multi-modality edge-aware Transformer-GAN for high-quality SPET reconstruction using the corresponding LPET images and T1 acquisitions from magnetic resonance imaging (T1-MRI). Specifically, to fully excavate the metabolic distributions in LPET and anatomical structural information in T1-MRI, we first use two separate CNN-based encoders to extract local spatial features from the two modalities, respectively, and design a multimodal feature integration module to effectively integrate the two kinds of features given the diverse contributions of features at different locations. Then, as CNNs can describe local spatial information well but have difficulty in modeling long-range dependencies in images, we further apply a Transformer-based encoder to extract global semantic information in the input images and use a CNN decoder to transform the encoded features into SPET images. Finally, a patch-based discriminator is applied to ensure the similarity of patch-wise data distribution between the reconstructed and real images. Considering the importance of edge information in anatomical structures for clinical disease diagnosis, besides voxel-level estimation error and adversarial loss, we also introduce an edge-aware loss to retain more edge detail information in the reconstructed SPET images. Experiments on the phantom dataset and clinical dataset validate that our proposed method can effectively reconstruct high-quality SPET images and outperform current state-of-the-art methods in terms of qualitative and quantitative metrics.

Unsupervised Medical Image Translation With Adversarial Diffusion Models

Imputation of missing images via source-to-target modality translation can improve diversity in medical imaging protocols. A pervasive approach for synthesizing target images involves one-shot mapping through generative adversarial networks (GAN). Yet, GAN models that implicitly characterize the image distribution can suffer from limited sample fidelity. Here, we propose a novel method based on adversarial diffusion modeling, SynDiff, for improved performance in medical image translation. To capture a direct correlate of the image distribution, SynDiff leverages a conditional diffusion process that progressively maps noise and source images onto the target image. For fast and accurate image sampling during inference, large diffusion steps are taken with adversarial projections in the reverse diffusion direction. To enable training on unpaired datasets, a cycle-consistent architecture is devised with coupled diffusive and non-diffusive modules that bilaterally translate between two modalities. Extensive assessments are reported on the utility of SynDiff against competing GAN and diffusion models in multi-contrast MRI and MRI-CT translation. Our demonstrations indicate that SynDiff offers quantitatively and qualitatively superior performance against competing baselines.

Deformation equivariant cross-modality image synthesis with paired non-aligned training data

Cross-modality image synthesis is an active research topic with multiple medical clinically relevant applications. Recently, methods allowing training with paired but misaligned data have started to emerge. However, no robust and well-performing methods applicable to a wide range of real world data sets exist. In this work, we propose a generic solution to the problem of cross-modality image synthesis with paired but non-aligned data by introducing new deformation equivariance encouraging loss functions. The method consists of joint training of an image synthesis network together with separate registration networks and allows adversarial training conditioned on the input even with misaligned data. The work lowers the bar for new clinical applications by allowing effortless training of cross-modality image synthesis networks for more difficult data sets.

High-fidelity direct contrast synthesis from magnetic resonance fingerprinting

PURPOSE

This work was aimed at proposing a supervised learning-based method that directly synthesizes contrast-weighted images from the Magnetic Resonance Fingerprinting (MRF) data without performing quantitative mapping and spin-dynamics simulations.

METHODS

To implement our direct contrast synthesis (DCS) method, we deploy a conditional generative adversarial network (GAN) framework with a multi-branch U-Net as the generator and a multilayer CNN (PatchGAN) as the discriminator. We refer to our proposed approach as N-DCSNet. The input MRF data are used to directly synthesize T1-weighted, T2-weighted, and fluid-attenuated inversion recovery (FLAIR) images through supervised training on paired MRF and target spin echo-based contrast-weighted scans. The performance of our proposed method is demonstrated on in vivo MRF scans from healthy volunteers. Quantitative metrics, including normalized root mean square error (nRMSE), peak signal-to-noise ratio (PSNR), structural similarity (SSIM), learned perceptual image patch similarity (LPIPS), and Fréchet inception distance (FID), were used to evaluate the performance of the proposed method and compare it with others.

RESULTS

In-vivo experiments demonstrated excellent image quality with respect to that of simulation-based contrast synthesis and previous DCS methods, both visually and according to quantitative metrics. We also demonstrate cases in which our trained model is able to mitigate the in-flow and spiral off-resonance artifacts typically seen in MRF reconstructions, and thus more faithfully represent conventional spin echo-based contrast-weighted images.

CONCLUSION

We present N-DCSNet to directly synthesize high-fidelity multicontrast MR images from a single MRF acquisition. This method can significantly decrease examination time. By directly training a network to generate contrast-weighted images, our method does not require any model-based simulation and therefore can avoid reconstruction errors due to dictionary matching and contrast simulation (code available at:https://github.com/mikgroup/DCSNet).

Multi-Scale Transformer Network With Edge-Aware Pre-Training for Cross-Modality MR Image Synthesis

Cross-modality magnetic resonance (MR) image synthesis can be used to generate missing modalities from given ones. Existing (supervised learning) methods often require a large number of paired multi-modal data to train an effective synthesis model. However, it is often challenging to obtain sufficient paired data for supervised training. In reality, we often have a small number of paired data while a large number of unpaired data. To take advantage of both paired and unpaired data, in this paper, we propose a Multi-scale Transformer Network (MT-Net) with edge-aware pre-training for cross-modality MR image synthesis. Specifically, an Edge-preserving Masked AutoEncoder (Edge-MAE) is first pre-trained in a self-supervised manner to simultaneously perform 1) image imputation for randomly masked patches in each image and 2) whole edge map estimation, which effectively learns both contextual and structural information. Besides, a novel patch-wise loss is proposed to enhance the performance of Edge-MAE by treating different masked patches differently according to the difficulties of their respective imputations. Based on this proposed pre-training, in the subsequent fine-tuning stage, a Dual-scale Selective Fusion (DSF) module is designed (in our MT-Net) to synthesize missing-modality images by integrating multi-scale features extracted from the encoder of the pre-trained Edge-MAE. Furthermore, this pre-trained encoder is also employed to extract high-level features from the synthesized image and corresponding ground-truth image, which are required to be similar (consistent) in the training. Experimental results show that our MT-Net achieves comparable performance to the competing methods even using 70% of all available paired data. Our code will be released at https://github.com/lyhkevin/MT-Net.

DC-cycleGAN: Bidirectional CT-to-MR synthesis from unpaired data

Magnetic resonance (MR) and computer tomography (CT) images are two typical types of medical images that provide mutually-complementary information for accurate clinical diagnosis and treatment. However, obtaining both images may be limited due to some considerations such as cost, radiation dose and modality missing. Recently, medical image synthesis has aroused gaining research interest to cope with this limitation. In this paper, we propose a bidirectional learning model, denoted as dual contrast cycleGAN (DC-cycleGAN), to synthesize medical images from unpaired data. Specifically, a dual contrast loss is introduced into the discriminators to indirectly build constraints between real source and synthetic images by taking advantage of samples from the source domain as negative samples and enforce the synthetic images to fall far away from the source domain. In addition, cross-entropy and structural similarity index (SSIM) are integrated into the DC-cycleGAN in order to consider both the luminance and structure of samples when synthesizing images. The experimental results indicate that DC-cycleGAN is able to produce promising results as compared with other cycleGAN-based medical image synthesis methods such as cycleGAN, RegGAN, DualGAN, and NiceGAN. Code is available at https://github.com/JiayuanWang-JW/DC-cycleGAN.

Generative AI in Medical Imaging: Applications, Challenges, and Ethics

Medical imaging is playing an important role in diagnosis and treatment of diseases. Generative artificial intelligence (AI) have shown great potential in enhancing medical imaging tasks such as data augmentation, image synthesis, image-to-image translation, and radiology report generation. This commentary aims to provide an overview of generative AI in medical imaging, discussing applications, challenges, and ethical considerations, while highlighting future research directions in this rapidly evolving field.

Reduction of scan duration and radiation dose in cerebral CT perfusion imaging of acute stroke using a recurrent neural network

Objective. Cerebral CT perfusion (CTP) imaging is most commonly used to diagnose acute ischaemic stroke and support treatment decisions. Shortening CTP scan duration is desirable to reduce the accumulated radiation dose and the risk of patient head movement. In this study, we present a novel application of a stochastic adversarial video prediction approach to reduce CTP imaging acquisition time.Approach. A variational autoencoder and generative adversarial network (VAE-GAN) were implemented in a recurrent framework in three scenarios: to predict the last 8 (24 s), 13 (31.5 s) and 18 (39 s) image frames of the CTP acquisition from the first 25 (36 s), 20 (28.5 s) and 15 (21 s) acquired frames, respectively. The model was trained using 65 stroke cases and tested on 10 unseen cases. Predicted frames were assessed against ground-truth in terms of image quality and haemodynamic maps, bolus shape characteristics and volumetric analysis of lesions.Main results. In all three prediction scenarios, the mean percentage error between the area, full-width-at-half-maximum and maximum enhancement of the predicted and ground-truth bolus curve was less than 4 ± 4%. The best peak signal-to-noise ratio and structural similarity of predicted haemodynamic maps was obtained for cerebral blood volume followed (in order) by cerebral blood flow, mean transit time and time to peak. For the 3 prediction scenarios, average volumetric error of the lesion was overestimated by 7%-15%, 11%-28% and 7%-22% for the infarct, penumbra and hypo-perfused regions, respectively, and the corresponding spatial agreement for these regions was 67%-76%, 76%-86% and 83%-92%.Significance. This study suggests that a recurrent VAE-GAN could potentially be used to predict a portion of CTP frames from truncated acquisitions, preserving the majority of clinical content in the images, and potentially reducing the scan duration and radiation dose simultaneously by 65% and 54.5%, respectively.

Contrast-Enhanced Liver Magnetic Resonance Image Synthesis Using Gradient Regularized Multi-Modal Multi-Discrimination Sparse Attention Fusion GAN

PURPOSES

To provide abdominal contrast-enhanced MR image synthesis, we developed an gradient regularized multi-modal multi-discrimination sparse attention fusion generative adversarial network (GRMM-GAN) to avoid repeated contrast injections to patients and facilitate adaptive monitoring.

METHODS

With IRB approval, 165 abdominal MR studies from 61 liver cancer patients were retrospectively solicited from our institutional database. Each study included T2, T1 pre-contrast (T1pre), and T1 contrast-enhanced (T1ce) images. The GRMM-GAN synthesis pipeline consists of a sparse attention fusion network, an image gradient regularizer (GR), and a generative adversarial network with multi-discrimination. The studies were randomly divided into 115 for training, 20 for validation, and 30 for testing. The two pre-contrast MR modalities, T2 and T1pre images, were adopted as inputs in the training phase. The T1ce image at the portal venous phase was used as an output. The synthesized T1ce images were compared with the ground truth T1ce images. The evaluation metrics include peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and mean squared error (MSE). A Turing test and experts' contours evaluated the image synthesis quality.

RESULTS

The proposed GRMM-GAN model achieved a PSNR of 28.56, an SSIM of 0.869, and an MSE of 83.27. The proposed model showed statistically significant improvements in all metrics tested with p-values < 0.05 over the state-of-the-art model comparisons. The average Turing test score was 52.33%, which is close to random guessing, supporting the model's effectiveness for clinical application. In the tumor-specific region analysis, the average tumor contrast-to-noise ratio (CNR) of the synthesized MR images was not statistically significant from the real MR images. The average DICE from real vs. synthetic images was 0.90 compared to the inter-operator DICE of 0.91.

CONCLUSION

We demonstrated the function of a novel multi-modal MR image synthesis neural network GRMM-GAN for T1ce MR synthesis based on pre-contrast T1 and T2 MR images. GRMM-GAN shows promise for avoiding repeated contrast injections during radiation therapy treatment.

Generative AI for brain image computing and brain network computing: a review

Recent years have witnessed a significant advancement in brain imaging techniques that offer a non-invasive approach to mapping the structure and function of the brain. Concurrently, generative artificial intelligence (AI) has experienced substantial growth, involving using existing data to create new content with a similar underlying pattern to real-world data. The integration of these two domains, generative AI in neuroimaging, presents a promising avenue for exploring various fields of brain imaging and brain network computing, particularly in the areas of extracting spatiotemporal brain features and reconstructing the topological connectivity of brain networks. Therefore, this study reviewed the advanced models, tasks, challenges, and prospects of brain imaging and brain network computing techniques and intends to provide a comprehensive picture of current generative AI techniques in brain imaging. This review is focused on novel methodological approaches and applications of related new methods. It discussed fundamental theories and algorithms of four classic generative models and provided a systematic survey and categorization of tasks, including co-registration, super-resolution, enhancement, classification, segmentation, cross-modality, brain network analysis, and brain decoding. This paper also highlighted the challenges and future directions of the latest work with the expectation that future research can be beneficial.

Fan beam CT image synthesis from cone beam CT image using nested residual UNet based conditional generative adversarial network

A radiotherapy technique called Image-Guided Radiation Therapy adopts frequent imaging throughout a treatment session. Fan Beam Computed Tomography (FBCT) based planning followed by Cone Beam Computed Tomography (CBCT) based radiation delivery drastically improved the treatment accuracy. Furtherance in terms of radiation exposure and cost can be achieved if FBCT could be replaced with CBCT. This paper proposes a Conditional Generative Adversarial Network (CGAN) for CBCT-to-FBCT synthesis. Specifically, a new architecture called Nested Residual UNet (NR-UNet) is introduced as the generator of the CGAN. A composite loss function, which comprises adversarial loss, Mean Squared Error (MSE), and Gradient Difference Loss (GDL), is used with the generator. The CGAN utilises the inter-slice dependency in the input by taking three consecutive CBCT slices to generate an FBCT slice. The model is trained using Head-and-Neck (H&N) FBCT-CBCT images of 53 cancer patients. The synthetic images exhibited a Peak Signal-to-Noise Ratio of 34.04±0.93 dB, Structural Similarity Index Measure of 0.9751±0.001 and a Mean Absolute Error of 14.81±4.70 HU. On average, the proposed model guarantees an improvement in Contrast-to-Noise Ratio four times better than the input CBCT images. The model also minimised the MSE and alleviated blurriness. Compared to the CBCT-based plan, the synthetic image results in a treatment plan closer to the FBCT-based plan. The three-slice to single-slice translation captures the three-dimensional contextual information in the input. Besides, it withstands the computational complexity associated with a three-dimensional image synthesis model. Furthermore, the results demonstrate that the proposed model is superior to the state-of-the-art methods.

A literature survey of MR-based brain tumor segmentation with missing modalities

Multimodal MR brain tumor segmentation is one of the hottest issues in the community of medical image processing. However, acquiring the complete set of MR modalities is not always possible in clinical practice, due to the acquisition protocols, image corruption, scanner availability, scanning cost or allergies to certain contrast materials. The missing information can cause some restraints to brain tumor diagnosis, monitoring, treatment planning and prognosis. Thus, it is highly desirable to develop brain tumor segmentation methods to address the missing modalities problem. Based on the recent advancements, in this review, we provide a detailed analysis of the missing modality issue in MR-based brain tumor segmentation. First, we briefly introduce the biomedical background concerning brain tumor, MR imaging techniques, and the current challenges in brain tumor segmentation. Then, we provide a taxonomy of the state-of-the-art methods with five categories, namely, image synthesis-based method, latent feature space-based model, multi-source correlation-based method, knowledge distillation-based method, and domain adaptation-based method. In addition, the principles, architectures, benefits and limitations are elaborated in each method. Following that, the corresponding datasets and widely used evaluation metrics are described. Finally, we analyze the current challenges and provide a prospect for future development trends. This review aims to provide readers with a thorough knowledge of the recent contributions in the field of brain tumor segmentation with missing modalities and suggest potential future directions.

Local contrastive loss with pseudo-label based self-training for semi-supervised medical image segmentation

Supervised deep learning-based methods yield accurate results for medical image segmentation. However, they require large labeled datasets for this, and obtaining them is a laborious task that requires clinical expertise. Semi/self-supervised learning-based approaches address this limitation by exploiting unlabeled data along with limited annotated data. Recent self-supervised learning methods use contrastive loss to learn good global level representations from unlabeled images and achieve high performance in classification tasks on popular natural image datasets like ImageNet. In pixel-level prediction tasks such as segmentation, it is crucial to also learn good local level representations along with global representations to achieve better accuracy. However, the impact of the existing local contrastive loss-based methods remains limited for learning good local representations because similar and dissimilar local regions are defined based on random augmentations and spatial proximity; not based on the semantic label of local regions due to lack of large-scale expert annotations in the semi/self-supervised setting. In this paper, we propose a local contrastive loss to learn good pixel level features useful for segmentation by exploiting semantic label information obtained from pseudo-labels of unlabeled images alongside limited annotated images with ground truth (GT) labels. In particular, we define the proposed contrastive loss to encourage similar representations for the pixels that have the same pseudo-label/GT label while being dissimilar to the representation of pixels with different pseudo-label/GT label in the dataset. We perform pseudo-label based self-training and train the network by jointly optimizing the proposed contrastive loss on both labeled and unlabeled sets and segmentation loss on only the limited labeled set. We evaluated the proposed approach on three public medical datasets of cardiac and prostate anatomies, and obtain high segmentation performance with a limited labeled set of one or two 3D volumes. Extensive comparisons with the state-of-the-art semi-supervised and data augmentation methods and concurrent contrastive learning methods demonstrate the substantial improvement achieved by the proposed method. The code is made publicly available at https://github.com/krishnabits001/pseudo_label_contrastive_training.

Data synthesis and adversarial networks: A review and meta-analysis in cancer imaging

Despite technological and medical advances, the detection, interpretation, and treatment of cancer based on imaging data continue to pose significant challenges. These include inter-observer variability, class imbalance, dataset shifts, inter- and intra-tumour heterogeneity, malignancy determination, and treatment effect uncertainty. Given the recent advancements in image synthesis, Generative Adversarial Networks (GANs), and adversarial training, we assess the potential of these technologies to address a number of key challenges of cancer imaging. We categorise these challenges into (a) data scarcity and imbalance, (b) data access and privacy, (c) data annotation and segmentation, (d) cancer detection and diagnosis, and (e) tumour profiling, treatment planning and monitoring. Based on our analysis of 164 publications that apply adversarial training techniques in the context of cancer imaging, we highlight multiple underexplored solutions with research potential. We further contribute the Synthesis Study Trustworthiness Test (SynTRUST), a meta-analysis framework for assessing the validation rigour of medical image synthesis studies. SynTRUST is based on 26 concrete measures of thoroughness, reproducibility, usefulness, scalability, and tenability. Based on SynTRUST, we analyse 16 of the most promising cancer imaging challenge solutions and observe a high validation rigour in general, but also several desirable improvements. With this work, we strive to bridge the gap between the needs of the clinical cancer imaging community and the current and prospective research on data synthesis and adversarial networks in the artificial intelligence community.

3D-MedTranCSGAN: 3D Medical Image Transformation using CSGAN

Computer vision techniques are a rapidly growing area of transforming medical images for various specific medical applications. In an end-to-end application, this paper proposes a 3D Medical Image Transformation Using a CSGAN model named a 3D-MedTranCSGAN. The 3D-MedTranCSGAN model is an integration of non-adversarial loss components and the Cyclic Synthesized Generative Adversarial Networks. The proposed model utilizes PatchGAN's discriminator network, to penalize the difference between the synthesized image and the original image. The model also computes the non-adversary loss functions such as content, perception, and style transfer losses. 3DCascadeNet is a new generator architecture introduced in the paper, which is used to enhance the perceptiveness of the transformed medical image by encoding-decoding pairs. We use the 3D-MedTranCSGAN model to do various tasks without modifying specific applications: PET to CT image transformation; reconstruction of CT to PET; modification of movement artefacts in MR images; and removing noise in PET images. We found that 3D-MedTranCSGAN outperformed other transformation methods in our experiments. For the first task, the proposed model yields SSIM is 0.914, PSNR is 26.12, MSE is 255.5, VIF is 0.4862, UQI is 0.9067 and LPIPs is 0.2284. For the second task, the model yields 0.9197, 25.7, 257.56, 0.4962, 0.9027, 0.2262. For the third task, the model yields 0.8862, 24.94, 0.4071, 0.6410, 0.2196. For the final task, the model yields 0.9521, 33.67, 33.57, 0.6091, 0.9255, 0.0244. Based on the result analysis, the proposed model outperforms the other techniques.

Topologically preserved registration of 3D CT images with deep networks

Objective. Computed Tomography (CT) image registration makes fast and accurate imaging-based disease diagnosis possible. We aim to develop a framework which can perform accurate local registration of organs in 3D CT images while preserving the topology of transformation.Approach. In this framework, the Faster R-CNN method is first used to detect local areas containing organs from fixed and moving images whose results are then registered with a weakly supervised deep neural network. In this network, a novel 3D channel coordinate attention (CA) module is introduced to reduce the loss of position information. The image edge loss and the organ labelling loss are used to weakly supervise the training process of our deep network, which enables the network learning to focus on registering organs and image structures. An intuitive inverse module is also used to reduce the folding of deformation field. More specifically, the folding is suppressed directly by simultaneously maximizing forward and backward registration accuracy in the image domain rather than indirectly by measuring the consistency of forward and inverse deformation fields as usual.Main results. Our method achieves an average dice similarity coefficient (DSC) of 0.954 and an average Similarity (Sim) of 0.914 on publicly available liver datasets (LiTS for training and Sliver07 for testing) and achieves an average DSC of 0.914 and an average Sim of 0.947 on our home-built left ventricular myocardium (LVM) dataset.Significance. Experimental results show that our proposed method can significantly improve the registration accuracy of organs such as the liver and LVM. Moreover, our inverse module can intuitively improve the inherent topological preservation of transformations.

Semi-Supervised Learning of MRI Synthesis Without Fully-Sampled Ground Truths

Learning-based translation between MRI contrasts involves supervised deep models trained using high-quality source- and target-contrast images derived from fully-sampled acquisitions, which might be difficult to collect under limitations on scan costs or time. To facilitate curation of training sets, here we introduce the first semi-supervised model for MRI contrast translation (ssGAN) that can be trained directly using undersampled k-space data. To enable semi-supervised learning on undersampled data, ssGAN introduces novel multi-coil losses in image, k-space, and adversarial domains. The multi-coil losses are selectively enforced on acquired k-space samples unlike traditional losses in single-coil synthesis models. Comprehensive experiments on retrospectively undersampled multi-contrast brain MRI datasets are provided. Our results demonstrate that ssGAN yields on par performance to a supervised model, while outperforming single-coil models trained on coil-combined magnitude images. It also outperforms cascaded reconstruction-synthesis models where a supervised synthesis model is trained following self-supervised reconstruction of undersampled data. Thus, ssGAN holds great promise to improve the feasibility of learning-based multi-contrast MRI synthesis.

Synthesis of T2-weighted images from proton density images using a generative adversarial network in a temporomandibular joint magnetic resonance imaging protocol

Purpose

This study proposed a generative adversarial network (GAN) model for T2-weighted image (WI) synthesis from proton density (PD)-WI in a temporomandibular joint (TMJ) magnetic resonance imaging (MRI) protocol.

Materials and Methods

From January to November 2019, MRI scans for TMJ were reviewed and 308 imaging sets were collected. For training, 277 pairs of PD- and T2-WI sagittal TMJ images were used. Transfer learning of the pix2pix GAN model was utilized to generate T2-WI from PD-WI. Model performance was evaluated with the structural similarity index map (SSIM) and peak signal-to-noise ratio (PSNR) indices for 31 predicted T2-WI (pT2). The disc position was clinically diagnosed as anterior disc displacement with or without reduction, and joint effusion as present or absent. The true T2-WI-based diagnosis was regarded as the gold standard, to which pT2-based diagnoses were compared using Cohen's ĸ coefficient.

Results

The mean SSIM and PSNR values were 0.4781(±0.0522) and 21.30(±1.51) dB, respectively. The pT2 protocol showed almost perfect agreement (ĸ=0.81) with the gold standard for disc position. The number of discordant cases was higher for normal disc position (17%) than for anterior displacement with reduction (2%) or without reduction (10%). The effusion diagnosis also showed almost perfect agreement (ĸ=0.88), with higher concordance for the presence (85%) than for the absence (77%) of effusion.

Conclusion

The application of pT2 images for a TMJ MRI protocol useful for diagnosis, although the image quality of pT2 was not fully satisfactory. Further research is expected to enhance pT2 quality.

A Systematic Literature Review on Applications of GAN-Synthesized Images for Brain MRI

With the advances in brain imaging, magnetic resonance imaging (MRI) is evolving as a popular radiological tool in clinical diagnosis. Deep learning (DL) methods can detect abnormalities in brain images without an extensive manual feature extraction process. Generative adversarial network (GAN)-synthesized images have many applications in this field besides augmentation, such as image translation, registration, super-resolution, denoising, motion correction, segmentation, reconstruction, and contrast enhancement. The existing literature was reviewed systematically to understand the role of GAN-synthesized dummy images in brain disease diagnosis. Web of Science and Scopus databases were extensively searched to find relevant studies from the last 6 years to write this systematic literature review (SLR). Predefined inclusion and exclusion criteria helped in filtering the search results. Data extraction is based on related research questions (RQ). This SLR identifies various loss functions used in the above applications and software to process brain MRIs. A comparative study of existing evaluation metrics for GAN-synthesized images helps choose the proper metric for an application. GAN-synthesized images will have a crucial role in the clinical sector in the coming years, and this paper gives a baseline for other researchers in the field.

Tumor-Attentive Segmentation-Guided GAN for Synthesizing Breast Contrast-Enhanced MRI Without Contrast Agents

OBJECTIVE

Breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a sensitive imaging technique critical for breast cancer diagnosis. However, the administration of contrast agents poses a potential risk. This can be avoided if contrast-enhanced MRI can be obtained without using contrast agents. Thus, we aimed to generate T1-weighted contrast-enhanced MRI (ceT1) images from pre-contrast T1 weighted MRI (preT1) images in the breast.

METHODS

We proposed a generative adversarial network to synthesize ceT1 from preT1 breast images that adopted a local discriminator and segmentation task network to focus specifically on the tumor region in addition to the whole breast. The segmentation network performed a related task of segmentation of the tumor region, which allowed important tumor-related information to be enhanced. In addition, edge maps were included to provide explicit shape and structural information. Our approach was evaluated and compared with other methods in the local (n = 306) and external validation (n = 140) cohorts. Four evaluation metrics of normalized mean squared error (NRMSE), Pearson cross-correlation coefficients (CC), peak signal-to-noise ratio (PSNR), and structural similarity index map (SSIM) for the whole breast and tumor region were measured. An ablation study was performed to evaluate the incremental benefits of various components in our approach.

RESULTS

Our approach performed the best with an NRMSE 25.65, PSNR 54.80 dB, SSIM 0.91, and CC 0.88 on average, in the local test set.

CONCLUSION

Performance gains were replicated in the validation cohort.

SIGNIFICANCE

We hope that our method will help patients avoid potentially harmful contrast agents. Clinical and Translational Impact Statement-Contrast agents are necessary to obtain DCE-MRI which is essential in breast cancer diagnosis. However, administration of contrast agents may cause side effects such as nephrogenic systemic fibrosis and risk of toxic residue deposits. Our approach can generate DCE-MRI without contrast agents using a generative deep neural network. Thus, our approach could help patients avoid potentially harmful contrast agents resulting in an improved diagnosis and treatment workflow for breast cancer.

TCGAN: a transformer-enhanced GAN for PET synthetic CT

Multimodal medical images can be used in a multifaceted approach to resolve a wide range of medical diagnostic problems. However, these images are generally difficult to obtain due to various limitations, such as cost of capture and patient safety. Medical image synthesis is used in various tasks to obtain better results. Recently, various studies have attempted to use generative adversarial networks for missing modality image synthesis, making good progress. In this study, we propose a generator based on a combination of transformer network and a convolutional neural network (CNN). The proposed method can combine the advantages of transformers and CNNs to promote a better detail effect. The network is designed for positron emission tomography (PET) to computer tomography synthesis, which can be used for PET attenuation correction. We also experimented on two datasets for magnetic resonance T1- to T2-weighted image synthesis. Based on qualitative and quantitative analyses, our proposed method outperforms the existing methods.

Functional Output Regression for Machine Learning in Materials Science

In recent years, there has been a rapid growth in the use of machine learning in material science. Conventionally, a trained predictive model describes a scalar output variable, such as thermodynamic, electronic, or mechanical properties, as a function of input descriptors that vectorize the compositional or structural features of any given material, such as molecules, chemical compositions, or crystalline systems. In machine learning of material data, on the other hand, the output variable is often given as a function. For example, when predicting the optical absorption spectrum of a molecule, the output variable is a spectral function defined in the wavelength domain. Alternatively, in predicting the microstructure of a polymer nanocomposite, the output variable is given as an image from an electron microscope, which can be represented as a two- or three-dimensional function in the image coordinate system. In this study, we consider two unified frameworks to handle such multidimensional or functional output regressions, which are applicable to a wide range of predictive analyses in material science. The first approach employs generative adversarial networks, which are known to exhibit outstanding performance in various computer vision tasks such as image generation, style transfer, and video generation. We also present another type of statistical modeling inspired by a statistical methodology referred to as functional data analysis. This is an extension of kernel regression to deal with functional outputs, and its simple mathematical structure makes it effective in modeling even with small amounts of data. We demonstrate the proposed methods through several case studies in materials science.

ResViT: Residual Vision Transformers for Multimodal Medical Image Synthesis

Generative adversarial models with convolutional neural network (CNN) backbones have recently been established as state-of-the-art in numerous medical image synthesis tasks. However, CNNs are designed to perform local processing with compact filters, and this inductive bias compromises learning of contextual features. Here, we propose a novel generative adversarial approach for medical image synthesis, ResViT, that leverages the contextual sensitivity of vision transformers along with the precision of convolution operators and realism of adversarial learning. ResViT's generator employs a central bottleneck comprising novel aggregated residual transformer (ART) blocks that synergistically combine residual convolutional and transformer modules. Residual connections in ART blocks promote diversity in captured representations, while a channel compression module distills task-relevant information. A weight sharing strategy is introduced among ART blocks to mitigate computational burden. A unified implementation is introduced to avoid the need to rebuild separate synthesis models for varying source-target modality configurations. Comprehensive demonstrations are performed for synthesizing missing sequences in multi-contrast MRI, and CT images from MRI. Our results indicate superiority of ResViT against competing CNN- and transformer-based methods in terms of qualitative observations and quantitative metrics.