Generative Adversarial Network Model Research Articles

Controllable Game Level Generation: Assessing the Effect of Negative Examples in GAN Models

Generative Adversarial Networks (GANs) are unsupervised models designed to learn and replicate a target distribution. The vanilla versions of these models can be extended to more controllable models. Conditional Generative Adversarial Networks (CGANs) extend vanilla GANs by conditioning both the generator and discriminator on some additional information (labels). Controllable models based on complementary learning, such as Rumi-GAN, have been introduced. Rumi-GANs leverage negative examples to enhance the generator's ability to learn positive examples. We evaluate the performance of two controllable GAN variants, CGAN and Rumi-GAN, in generating game levels targeting specific constraints of interest: playability and controllability. This evaluation is conducted under two scenarios: with and without the inclusion of negative examples. The goal is to determine whether incorporating negative examples helps the GAN models avoid generating undesirable outputs. Our findings highlight the strengths and weaknesses of each method in enforcing the generation of specific conditions when generating outputs based on given positive and negative examples.

Fault diagnosis method for elevator carriages based on temporal generative federated distillation

Addressing the ‘data silo’ issue among different elevator operating units and the temporal correlations in elevator vibration signals, a novel small-sample fault diagnosis method for elevator carriages based on temporal generative federated distillation is proposed. This method incorporates a temporal generative adversarial network into Federated Distillation via Generative Learning (FedGen). FedGen combines federated learning, knowledge distillation, and generative models to enhance model aggregation efficiency while mitigating data heterogeneity. However, the original generative model struggles to maintain dynamic correlations between signals when extracting temporal features. Therefore, an improved Time Series Generative Adversarial Networks (TimeGAN) model is introduced, substituting the initial logarithmic loss function with a least squares error function, thereby enhancing training stability and data quality. This approach eliminates the need for proxy datasets in knowledge distillation, avoiding the loss of temporal information during central server feature extraction. Simulation results demonstrate that this method enables data sharing while protecting data privacy, and enhances model generalization capabilities.

Style recommendation and simulation for handmade artworks using generative adversarial networks.

Today, artificial intelligence (AI) is used in the design and production of artworks as a powerful tool that helps artists and designers to create more creative and attractive artworks. AI as a new tool has created new opportunities to create and improve artworks. Therefore, in this study, a new model for recommending and simulating the style of handmade artworks using generative adversarial networks (GANs) is presented. This model operates in two distinct phases: style recommendation and style simulation. The reason for using these two separate phases is to increase the accuracy and efficiency of the model. In the first phase, a pre-trained GAN is used to suggest different styles to the artist. In the second phase, a new GAN specifically designed to produce realistic images of handmade artwork is used to simulate the artist's selected style. Also, in order to improve the accuracy, the genetic algorithm (GA) was used to adjust the activity configuration of Self-Attention (SA) modules. The results of a case study on two types of data, including samples without and with the initial background pattern, GA-SAGAN model has been able to significantly reduce the validation error. As a result, adjusting the activity configuration of SA modules using GA can be effective in producing realistic and accurate artworks. The outputs produced by the GA-SAGAN model have a loss rate close to zero. Also, the Entropy values and Precision and Recall indices are higher than GAN and SAGAN models, it shows the higher diversity of production output patterns and the superiority of the proposed model.

Reconstruction of 30 m Land Cover in the Qilian Mountains from 1980 to 1990 Based on Super-Resolution Generative Adversarial Networks

Long time series of annual land cover with fine spatio-temporal resolutions play a crucial role in studying environmental climate change, biophysical modeling, carbon cycling models, and land management. Despite a strong consistency exhibited by several publicly available medium to fine resolution global land cover datasets, significant discrepancies exist at the regional scale; moreover, only every 5/10 year land cover were available. Consequently, high-quality annual land cover datasets before 2000 are unavailable in China. In this study, we proposed a deep learning-based method by integrating multiple remote sensing data from different platforms with historical high spatial resolution land cover datasets (CNLUCC) to derive the 30 m annual land cover maps from 1980 to 1990 for Qilian Mountain. First, the super-resolution generative adversarial network models for upscaling the 5.5 km AVHRR NDVI to 250 m were established by employing the AVHRR and MODIS NDVI data with the same year as input, and the early time series AVHRR NDVI data were subsequently upscaled to 250 m through the above models. Second, the breaks for the additive seasonal and trend (BFAST) change detection algorithm was applied to the upscaled time series NDVI data to detect the change time of different land cover types. Third, the CNLUCC data in 1980 and 1990 were updated to annual land cover datasets from 1980 to 1990 and the annual mapping results provided insights into the dynamic processes of urbanization, deforestation, water bodies, and farmland from 1980 to 1990. Finally, comprehensive analysis and validation were carried out for evaluation and an overall accuracy of 77.26% for the land cover product in 1986 was achieved.

Multi-Model Style-Aware Diffusion Learning for Semantic Image Synthesis

Semantic image synthesis aims to generate images from given semantic layouts, which is a challenging task that requires training models to capture the relationship between layouts and images. Previous works are usually based on Generative Adversarial Networks (GAN) or autoregressive (AR) models. However, the GAN model's training process is unstable, and the AR model’s performance is seriously affected by the independent image encoder and the unidirectional generation bias. Due to the above limitations, these methods tend to synthesize unrealistic, poorly aligned images and only consider single-style image generation. In this paper, we propose a Multi-model Style-aware Diffusion Learning (MSDL) framework for semantic image synthesis, including a training module and a sampling module. In the training module, a layout-to-image model is introduced to transfer the learned knowledge from a model pretrained with massive weak correlated text-image pairs data, making the training process more efficient. In the sampling module, we designed a map-guidance technique and creatively designed a multi-model style-guidance strategy for creating images in multiple styles, e.g., oil painting, Disney Cartoon, and pixel style. We evaluate our method on Cityscapes, ADE20K, and COCO-Stuff, making visual comparisons and computing with multiple metrics such as FID, LPIPS, etc. Experimental results demonstrate that our model is highly competitive, especially in terms of fidelity and diversity.

Film image processing and production based on high-performance calculation

In film and television production, efficient and precise image processing is vital for achieving realistic visual effects. Therefore, exploring and applying advanced image processing technologies has become an essential method for elevating the production quality of film and television projects. This work investigates the application of artificial intelligence (AI) technology in the processing and production of animated images in film and television scenarios. By comparing the performance of standard Generative Adversarial Network (GAN), DenseNet, and CycleGAN models under different noise conditions, it is found that CycleGAN performs the best in image denoising and detail restoration. Experimental results demonstrate that CycleGAN achieves a Peak Signal-to-noise Ratio (PSNR) of 30.1dB and a Structural Similarity Index Measure (SSIM) of 0.88 under Gaussian noise conditions. Moreover, CycleGAN achieves a PSNR of 29.5dB and an SSIM of 0.85 under salt-and-pepper noise conditions. It outperforms the other models in both conditions. Additionally, CycleGAN’s mean absolute error is significantly lower than that of the other models. This work demonstrates that CycleGAN can more effectively handle complex noise and generate high-quality images under unsupervised learning conditions. These findings provide new directions for future image processing research and offer important references for model selection in practical applications. This work not only offers new perspectives on the development of animation image processing technology but also establishes a theoretical foundation for applying advanced AI techniques in film and television production. Through comparative analysis of various deep learning models, this work highlights the superior performance of CycleGAN under complex noise conditions. This advancement not only drives progress in image processing technology but also provides effective solutions for efficient production and quality enhancement of future film and television works.

NIMG-33. HIGH-RESOLUTION SODIUM MRI OF HUMAN GLIOMAS AT 3T USING PHYSICS-BASED GENERATIVE AI

Abstract BACKGROUND Sodium neuroimaging is a promising technique for diagnosing and monitoring brain tumors, providing insights into microenvironment and metabolism. However, at 3T, it is limited by low signal-to-noise ratio and resolution, resulting in long acquisition times and low-quality images. To address these limitations, we developed a physics-informed generative adversarial network (GAN) approach for high-resolution sodium neuroimaging of brain tumors at 3T. METHODS 5,078 anatomical sequences from 1,330 brain tumor patients undergoing routine proton MRI were used to create a synthetic dataset with physics-based simulated artifacts, including B0 magnetic susceptibility artifacts, chemical shift artifacts, Nyquist aliasing artifacts, Gibbs artifacts, and low-resolution images, with Rician noise added during training to augment the dataset. By using multiple proton MRI contrasts, we hypothesized that the GAN model would learn to correct artifacts while preserving the inherent contrast information. We employed a transfer learning approach with a modified Pix2PixGAN architecture and an Attention-R2UNet generator. Twenty glioma patients consented to participate in this IRB-approved prospective study. Sodium and proton MRIs were acquired in the same sessions using a 3T Siemens scanner using a dual-tuned head coil, with a 3D spoiled gradient-echo sequence optimized for short TE. Sodium-proton exchanger (NHE1) expression was tested with immunohistochemistry from image-guided surgical biopsies. RESULTS High-resolution synthetic-sodium MR images were generated using the GAN model. The proposed method successfully reconstructed high-resolution synthetic-sodium MR images with improved SNR compared to traditional sodium images. Synthetic and native measurements were strongly correlated (R2=0.8539, P<0.0001). A significant correlation was seen between synthetic-sodium MR measurements and relative NHE1 tissue expression (ρ=0.5817, P=0.0036). CONCLUSION Results suggest that high-resolution synthetic-sodium MRI retains much of the inherent information from native sodium MR while accurately representing NHE1 expression. This approach could make high-resolution sodium neuroimaging at 3T feasible in the clinical environment, potentially improving diagnosis, monitoring, and treatment of brain tumors.

Damage detection for railway bridges using time‐frequency decomposition and conditional generative model

AbstractA novel damage detection model, which utilizes the spatiotemporal characteristics of the acceleration data, is proposed to assess the structural integrity of railway bridges. For this, the measured acceleration data are decomposed into several intrinsic mode functions (IMFs) using the sparse random mode decomposition model. The generated IMFs are subsequently integrated into the enhanced time series conditional generative adversarial network model to identify possible damage in bridges across various frequency bands. The influence of environmental and operational variables (EOVs), particularly temperature fluctuations, was also investigated. The proposed model was verified using both numerical and experimental data from a plate girder bridge. Further validation was conducted using the Z24 bridge dataset, and damage cases under the influence of EOVs were successfully predicted. Throughout the validation process, various anomaly metrics were introduced to establish a threshold value, and a covariance‐based time domain metric was proven to be the most effective in our cases.

NIMG-65. SYNTHETIC MRI GENERATION TO FACILITATE AUTOSEGMENTATION OF PEDIATRIC BRAIN TUMORS IN LIMITED DATA SCENARIOS

Abstract BACKGROUND The availability of standard multiparametric MRI sequences, including pre-contrast T1w (T1w-pre), post-contrast T1w (T1w-post), T2w, and FLAIR images, is essential for accurate segmentation of tumor subregions and subsequent response assessment in pediatric brain tumors (PBTs). However, clinical MRI sets are often incomplete due to imaging artifacts or inconsistent acquisition protocols across different centers. This study leverages Generative Adversarial Networks (GANs) to synthesize missing FLAIR and T1w-pre images from T2w and T1w-post images, respectively, to facilitate tumor segmentation in PBTs. METHODS We developed two GAN models based on the pix2pix framework, trained, validated, and tested on FLAIR-T2w and T1w-post-T1w-pre pairs from a multi-histology and multi-institutional cohort of 476 subjects with PBTs from the Children’s Brain Tumor Network (CBTN). The perceptual quality of the synthesized T1w-pre and FLAIR images was evaluated using the Structural Similarity Index (SSI; ideal value = 1). The robustness of the synthesized sequences to replace missing MRI sequences was evaluated by comparing the segmentation performance (Dice score) of our pre-trained autosegmentation model using all real modalities versus replacing real sequences with their synthesized counterparts. RESULTS The T1w-pre and FLAIR synthesis GANs achieved mean SSI values of 0.95 and 0.93, respectively. Median Dice scores for segmentation of whole tumor (WT), enhancing tumor (ET), and tumor core (TC) were calculated for real sequences: 0.90, 0.83, 0.89, replacing real FLAIR with synthetic: 0.86, 0.81, 0.86, and replacing real T1w-pre with synthetic: 0.90, 0.75, 0.89. No statistical significance (p>0.05) was observed in segmentation performance for WT, TC, ET when replacing real T1w-pre with synthetic, and for ET when replacing real with synthetic FLAIR sequences. CONCLUSION This method achieves robust segmentation performance in limited data scenarios by synthesizing missing MRI sequences. Future work includes improving FLAIR synthesis and segmentation performance for WT and ET by implementing and comparing other GAN architectures.

Hydrological Response to Climate Change: McGAN for Multi-Site Scenario Weather Series Generation and LSTM for Streamflow Modeling

This study focuses on the impacts of climate change on hydrological processes in watersheds and proposes an integrated approach combining a weather generator with a multi-site conditional generative adversarial network (McGAN) model. The weather generator incorporates ensemble GCM predictions to generate regional average synthetic weather series, while McGAN transforms these regional averages into spatially consistent multi-site data. By addressing the spatial consistency problem in generating multi-site synthetic weather series, this approach tackles a key challenge in site-scale climate change impact assessment. Applied to the Jinghe River Basin in west-central China, the approach generated synthetic daily temperature and precipitation data for four stations under different shared socioeconomic pathways (SSP1-26, SSP2-45, SSP3-70, SSP5-85) up to 2100. These data were then used with a long short-term memory (LSTM) network, trained on historical data, to simulate daily river flow from 2021 to 2100. The results show that (1) the approach effectively addresses the spatial correlation problem in multi-site weather data generation; (2) future climate change is likely to increase river flow, particularly under high-emission scenarios; and (3) while the frequency of extreme events may increase, proactive climate policies can mitigate flood and drought risks. This approach offers a new tool for hydrologic–climatic impact assessment in climate change studies.

Fiber-Optic Sensor Spectrum Noise Reduction Based on a Generative Adversarial Network

In the field of fiber-optic sensing, effectively reducing the noise of sensing spectra and achieving a high signal-to-noise ratio (SNR) has consistently been a focal point of research. This study proposes a deep-learning-based denoising method for fiber-optic sensors, which involves pre-processing the sensor spectrum into a 2D image and training with a cycle-consistent generative adversarial network (Cycle-GAN) model. The pre-trained algorithm demonstrates the ability to effectively denoise various spectrum types and noise profiles. This study evaluates the denoising performance of simulated spectra obtained from four different types of fiber-optic sensors: fiber Fabry–Perot interferometer (FPI), regular fiber Bragg grating (FBG), chirped FBG, and FBG pair. Compared to traditional denoising algorithms such as wavelet transform (WT) and empirical mode decomposition (EMD), the proposed method achieves an SNR improvement of up to 13.71 dB, an RMSE that is up to three times smaller, and a minimum correlation coefficient (R2) of no less than 99.70% with the original high-SNR signals. Additionally, the proposed algorithm was tested for multimode noise reduction, demonstrating an excellent linearity in temperature response with a R2 of 99.95% for its linear fitting and 99.74% for the temperature response obtained from single-mode fiber sensors. The proposed denoising approach effectively reduces the impact of various noises from the sensing system, enhancing the practicality of fiber-optic sensing, especially for specialized fiber applications in research and industrial domains.

Generative adversarial network model to classify human induced pluripotent stem cell-cardiomyocytes based on maturation level

Our study develops a generative adversarial network (GAN)-based method that generates faithful synthetic image data of human cardiomyocytes at varying stages in their maturation process, as a tool to significantly enhance the classification accuracy of cells and ultimately assist the throughput of computational analysis of cellular structure and functions. Human induced pluripotent stem cell derived cardiomyocytes (hiPSC-CMs) were cultured on micropatterned collagen coated hydrogels of physiological stiffnesses to facilitate maturation and optical measurements were performed for their structural and functional analyses. Control groups were cultured on collagen coated glass well plates. These image recordings were used as the real data to train the GAN model. The results show the GAN approach is able to replicate true features from the real data, and inclusion of such synthetic data significantly improves the classification accuracy compared to usage of only real experimental data that is often limited in scale and diversity. The proposed model outperformed four conventional machine learning algorithms with respect to improved data generalization ability and data classification by incorporating synthetic data. This work demonstrates the importance of integrating synthetic data in situations where there are limited sample sizes and thus, effectively addresses the challenges imposed by data availability.

A novel generative approach to the parametric design and multi-objective optimization of horizontal axis tidal turbines

As the demand for ocean energy continues to grow, the development of efficient design and optimization methods for tidal current turbines is crucial. Traditional approaches, often based on parameterized models, face challenges in fully capturing the intricate geometric features of turbine blades, limiting the optimization space and affecting convergence efficiency. In response, this study introduces a novel design methodology for horizontal axis tidal turbines (HATTs) using a variational autoencoder generative adversarial network (VAEGAN) model. This approach uses unsupervised learning from a custom dataset to generate new HATT designs, with the VAEGAN model encoding distinct geometric features of turbine blades within a compact latent space, enabling more efficient design space exploration and facilitating the discovery of innovative shapes. Furthermore, in the multi-objective optimization process targeting both hydrodynamic performance and structural strength, the reduced dimensionality of the design variables accelerates convergence while maintaining a broad and meaningful design space. The proposed methodology demonstrates the VAEGAN model's ability to generate diverse and effective turbine blade designs, highlighting its potential as a powerful tool in advancing HATT technology.

UNVEILING THE DECISION MAKING PROCESS IN ALZHEIMER’S DISEASE DIAGNOSIS: A CASE-BASED COUNTERFACTUAL METHODOLOGY FOR EXPLAINABLE DEEP LEARNING

BackgroundThe field of Alzheimer's disease (AD) diagnosis is undergoing significant transformation due to the application of deep learning (DL) models. While DL surpasses traditional machine learning in disease prediction from structural magnetic resonance imaging (sMRI), the lack of explainability limits clinical adoption. Counterfactual inference offers a way to integrate causal explanations into these models, enhancing their robustness and transparency. New methodThis study develops a novel methodology combining U-Net and generative adversarial network (GAN) models to create comprehensive counterfactual diagnostic maps for AD. The proposed methodology uses case-based counterfactual reasoning for robust decision classification for counterfactual maps to understand how changes in specific features affect the model's predictions. Comparison with existing methodsThe proposed methodology is compared with state-of-the-art visual explanation techniques across the ADNI dataset. The proposed methodology is also benchmarked against other gradient-based and generative models for its ability to generate comprehensive counterfactual maps. ResultsThe results demonstrate that the proposed methodology significantly outperforms existing methods in accuracy, sensitivity, and specificity while providing detailed counterfactual maps that visualize how slight changes in brain morphology could lead to different diagnostic outcomes. The proposed methodology achieves an accuracy of 95% and an AUC of 0.97, illustrating its superiority in capturing subtle yet crucial anatomical features. ConclusionsBy generating intuitive visual explanations, the proposed methodology improves the interpretability and robustness of AD diagnostic models, making them more reliable and accountable. The use of counterfactual inference enhances clinicians' understanding of disease progression and the impact of different interventions, fostering precision medicine in AD care.

Resolution-enhanced quantitative phase imaging of blood platelets using a generative adversarial network

We developed a new method to enhance the resolution of blood platelet aggregates imaged via quantitative phase imaging (QPI) using a Pix2Pix generative adversarial network (GAN). First, 1 µm polystyrene beads were imaged with low- and high-resolution QPI, to train the GAN model and validate its applicability. Testing on the polystyrene beads demonstrated a mean error of 4.14% in the generated high-resolution optical-path-delay values compared to the optically acquired ones. Next, blood platelets were collected with low- and high-resolution QPI, and a deep neural network was trained to predict the high-resolution platelet optical-path-delay profiles using the low-resolution profiles, achieving a mean error of 7.01% in the generated high-resolution optical-path-delay values compared to the optically acquired ones. These results highlight the potential of the method in enhancing QPI resolution of cell aggregates without the need for sophisticated optical equipment and optical system modifications for high-resolution microscopy, allowing for better understanding of platelet-related disorders and conditions such as thrombocytopenia and thrombocytosis.

Removing cloudiness on optical space images by a generative adversarial network model using SAR images

The object of this study is the process of removing cloudiness on optical space images. Solving the cloudiness removal task is an important stage in processing data from the Earth remote probing (ERP) aimed at reconstructing the information hidden by these atmospheric disturbances. The analyzed shortcomings in the fusion of purely optical data led to the conclusion that the best solution to the cloudiness removal problem is a combination of optical and radar data. Compared to conventional methods of image processing, neural networks could provide more efficient and better performance indicators due to the ability to adapt to different conditions and types of images. As a result, a generative adversarial network (GAN) model with cyclic-sequential 7-ResNeXt block architecture was constructed for cloud removal in optical space imagery using synthetic aperture radar (SAR) imagery. The model built generates fewer artifacts when transforming the image compared to other models that process multi-temporal images. The experimental results on the SEN12MS-CR data set demonstrate the ability of the constructed model to remove dense clouds from simultaneous Sentinel-2 space images. This is confirmed by the pixel reconstruction of all multispectral channels with an average RMSE value of 2.4 %. To increase the informativeness of the neural network during model training, a SAR image with a C-band signal is used, which has a longer wavelength and thereby provides medium-resolution data about the geometric structure of the Earth's surface. Applying this model could make it possible to improve the situational awareness at all levels of control over the Armed Forces (AF) of Ukraine through the use of current space observations of the Earth from various ERP systems

Transformer-Based GAN with Multi-STFT for Rotating Machinery Vibration Data Analysis

Prognostics and health management of general rotating machinery have been studied over time to improve system stability. Recently, the excellent abnormal diagnosis performance of artificial intelligence (AI) was demonstrated, and therefore, AI-based intelligent diagnosis is now being implemented in these systems. AI models are trained using large volumes of data. Therefore, we propose a transformer-based generative adversarial network (GAN) model with a multi-resolution short-time Fourier transform (multi-STFT) loss function to augment the vibration data of rotating machinery to facilitate the successful learning of deep learning models. We constructed a model with a conditional GAN structure, which is transformer based, for learning the feature points of vibration data in the time-series domain. In addition, we applied the multi-STFT loss function to capture the frequency features of the vibration data. The generated data, which adequately captured the frequency features, were used to augment the training data to improve the performance of a deep learning classifier. Furthermore, by visualizing the generated vibration data and comparing the visualizations to those of the vibration data obtained from real machinery, we demonstrated that the generated data were indistinguishable from the actual data.

Malware Detection and Classification using Generative Adversarial Network

The Generative Adversarial Networks (GANs) are playing a crucial role in deep-learning-based malware classification to overcome the dataset imbalance and unseen malware. The Generative AI is preferably used in many applications, such as improving image resolution and generating audio, video, and text. The cybercriminals are also using the Generative AI for generating the malware and deepfake videos to harm the targeted person or device. By generating the synthetic data, it makes the deep learning model more robust to detect such types of unseen and adversarial attacks. This work utilizes GANs for generating adversarial malware samples to train a classification and detection model, improving the model’s ability to identify sophisticated malware variants. The performance of the proposed Conditional Generative Adversarial Network (CGAN) model is evaluated on a multiclass malware grayscale image dataset and a binary class malware RGB image dataset. The performance of proposed model is compared with current state-of-the-art. Results indicate a significant improvement in classification accuracy and a reduction in training time and false positives, showcasing GAN’s potential in the dynamic cybersecurity landscape.

A novel digital twin strategy to enable the translation of evidence from randomized control trials to new populations

Abstract Background Randomized clinical trials (RCTs) are essential to guide medical practice; however, their generalizability to a given population is often uncertain. Between the SPRINT and ACCORD trials, only SPRINT found a significant reduction in major cardiovascular events (MACE) with intensive blood pressure control, but ACCORD did not, without clear evidence of whether this reflected differences in enrolled patients. Purpose To use a novel RCT digital twin-based approach to demonstrate the translation of an RCT to a different population, leveraging the differences in populations and treatment effects in SPRINT and ACCORD. Methods We developed a statistically informed Generative Adversarial Network (GAN) model that leverages relationships between covariates and outcomes using Directed Acyclic Graphs (DAGs). This model can generate a digital twin of an RCT (RCT-Twin) conditioned on covariate distributions from a second patient population while maintaining these relationships (Fig A). Using this, we generated the (i) RCT-Twin of SPRINT, which was conditioned on the 10 most disparate covariates drawn from ACCORD, and (ii) an RCT-Twin of ACCORD conditioned on SPRINT. To demonstrate treatment effect estimates of each RCT conditioned on the other RCT’s patient population, we evaluated the cardiovascular event-free survival of 10 iterations of these cross-trained RCT-Twins against the original trials. Results Using 9361 SPRINT participants and conditioning on 4733 ACCORD participants, we generated 10 ACCORD-conditioned SPRINT-Twin datasets. We confirmed that the RCT-Twins were consistently balanced between treatment and controls across all measured covariates (mean absolute standardized mean difference (MASMD) 0.019, SD 0.018), as expected for real RCTs. We also confirmed that covariates of the ACCORD-conditioned SPRINT-Twins were closer to ACCORD than SPRINT (MASMD 0.0082 vs 0.46, SD 0.016 vs 0.20), suggestive of successful conditioning. Most importantly, across iterations, ACCORD-conditioned SPRINT-Twin datasets reproduced the overall non-significant effect size seen in ACCORD (5-year MACE hazard ratio (95% confidence interval) of 0.88 (0.73-1.06) in ACCORD vs median 0.87 (0.68-1.13) in the ACCORD-conditioned SPRINT-Twins, Fig B), while the SPRINT-conditioned ACCORD-Twin datasets reproduced the significant effect size seen in SPRINT (0.75 (0.64-0.89) vs median 0.79 (0.72-0.86) in the SPRINT-conditioned ACCORD-Twins, Fig C). Conclusions Our novel conditioned digital twin approach simulates RCT-derived effects in different patient populations by translating these effects to the covariate distributions of the patients. This key methodological advance may enable the direct translation of RCT-derived effects into disparate patient populations and may enable causal inference in real-world settings.

Deep learning approaches for image restoration in invasive coronary angiography

Abstract Background Despite being the gold standard for diagnosing coronary artery disease and guiding percutaneous coronary interventions (PCI), image restoration during invasive coronary angiography (ICA) has rarely been explored. Generative adversarial network (GAN), a deep learning model, has demonstrated its effectiveness in image restoration. Our hypothesis was that we could restore ICA images, damaged or deleted to varying degrees, using the inpainting GAN method, and we undertook efforts to test this hypothesis. Methods We extracted 22,659 ICA patch images (128 × 128 size) from 60 patients, focusing on sparse vessels for model training. To assess our inpainting GAN model, we degraded the extracted images to varying degrees: minimal, mild, moderate, severe, and completely blocked. The performance was quantitatively analyzed using metrics such as the peak signal-to-noise ratio (PSNR), structured similarity indexing methods (SSIM), and mean squared error (MSE). Results The inpainting GAN method significantly improved the image quality in all degraded images. For the blocked images, PSNR and SSIM were enhanced by 161.9% and 14.3%, respectively, with a 98% MSE reduction. In severely, moderately, mildly, and minimally damaged images, PSNR and SSIM increased by 14.9% and 14.8%, 15.7% and 14.7%, 19.3% and 15.6%, and 12.7% and 5.3%, respectively, with MSE reductions of 61.5%, 64.1%, 72.7%, and 33.9%, respectively. Conclusions Inpainting GAN can restore ICA images with varying damage levels, including deletions. Our deep-learning model can instantaneously improve ICA image quality, potentially aiding in emergency or chronic total occlusion PCI.Table 2.Quantitative Analysis of ImageFigure 1.Sample Cases Pre and Post-Rest