FreshU-GAN: A Deep Learning model for determining beef freshness

With the ascendance of Generative Adversarial Networks (GAN), promising prospects have arisen from the abilities of machines to learn and recognize patterns in 2D datasets and generate new results as an inspirational tool in architectural design. Insofar as the majority of ML experiments in architecture are conducted with imagery based on readily available 2D data, architects and designers are faced with the challenge of transforming machine-generated images into 3D. On the other hand, GAN-generated images are found to be able to learn the 3D information out of 2D perspectival images. To facilitate such transformation from 2D and 3D data in the framework of deep learning in architecture, this paper explores making new architectural forms from flat GAN images by employing traditional tools of projective geometry. The experiments draw on Brook Taylor’s 19th- century theorem of inverse projection system for creating architectural form from perspectival information learned from GAN images of Swiss alpine architecture. The research develops a parametric tool that automates the dimensional transformation of 2D images into 3D architectural forms. This research identifies potential synergic interactions between traditional tools and techniques of architects and deep learning algorithms to achieve collective intelligence in designing and representing creative architecture forms between humans and machines.

With the widespread adoption of Generative Adversarial Networks (GANs) in image generation and processing, enhancing their generation quality and visualization capabilities has become a prominent research focus. This study introduces a deep learning framework that integrates multi-scale information chart generation with visualization enhancement to improve the performance of GAN-based image generation models across various domains. Based on the Self-Attention Generative Adversarial Network (SAGAN), which leverages self-attention mechanisms to capture long-range dependencies in images, the proposed approach significantly enhances image quality and detail representation. The framework incorporates a multi-scale feature extraction method to optimize the feature maps at each layer of the generative network. Experimental results demonstrate that SAGAN outperforms traditional GAN models in terms of image clarity, detail preservation, and visual effects. The proposed model achieves notable improvements in diversity and generalization, with a mutual information content of 0.91, clustering uniformity of 0.89, and inter-cluster dissimilarity of 0.92 on the CelebA dataset. Furthermore, in terms of image quality, SAGAN attains a Structural Similarity Index Measure (SSIM) of 0.94 and a Peak Signal-to-Noise Ratio (PSNR) of 30.1, surpassing traditional GANs by a significant margin.

Accurate detection and characterisation of defects in high-density polyethylene (HDPE) materials are important for the safety of industrially critical structures. Ultrasonic non-destructive evaluation (UNDE) has proven to be a powerful tool for detecting and characterising defects in engineered materials. However, efficient and high-precision defect imaging in these highly attenuating materials remains a significant challenge for UNDE. Least-squares reverse time migration (LSRTM) offers the potential to reconstruct high-precision images of reflectivity. Yet, the conventional LSRTM iteratively updates the reflectivity model by minimising the data residuals, making it computationally expensive. In this paper, an efficient ultrasonic LSRTM algorithm within a deep learning framework is proposed. Building upon this, a generative adversarial network (GAN) is integrated to further enhance the reconstruction results by reducing artefacts in the images. Simulation and experimental results show that the proposed ultrasonic LSRTM-GAN can generate high-quality images, effectively enabling precise defect detection in HDPE.

Characterization of hydration and dry shrinkage behavior of cement emulsified asphalt composites using deep learning

This study was performed in order to select volatile compounds to predict the off-odour and overall assessment of raw beef's freshness Maronesa breed, using multivariate analysis. M. longissimus dorsi packed in vacuum and MAP (70% O2/20% CO2/10%N2) stored at 4 ºC were examined for off-odour perception as well as the overall assessment of freshness at 10 and 21days post mortem. The results achieved in this study demonstrated that the selected volatile compounds could be considered as volatile indicators of beef spoilage, enclosing information for discrimination of Maronesa beef samples in sensory classes of odour corresponding to unspoiled and spoiled levels. Fifty-four volatile compounds were detected. A significant increase of aldehydes, ketones and alcohols were observed during storage in MAP. 2 and 3-methylbutanal, 2 and 3-methylbutanol, 1-pentanol, 1-hexanol, 2,3-octanedione, 3,5-octanedione, octanal and nonanal were suggested as indicators of beef spoilage. 3-methylpentane was considered as a marker in the first stages of spoilage in beef, decreasing during storage. Data were examined using PCR and PLSR models for different optimal subsets of volatile compounds. The simplicity and usefulness of the technique in using 0/1 data in preserving high levels of accuracy was also prevalent. The powerful analytical methodologies for reducing variables and the choice of optimal subsets could be advantageous in both basic research and the routine quality control of chilled beef.

Food analysis and consumer protection

Paediatric pneumonia is a leading cause of morbidity and mortality worldwide, necessitating accurate and timely diagnosis. This study explores the application of Generative AI for categorising paediatric pneumonia using chest radiographs. Leveraging deep learning techniques, including Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), we enhance image quality, generate synthetic training data, and improve model generalizability. The proposed framework integrates AI-driven feature extraction, convolutional neural networks (CNNs), and attention mechanisms to improve diagnostic accuracy. The results demonstrate significant improvements in classification performance compared to traditional methods, with a focus on interpretability and clinical usability.

High-performance neural networks for synthetic aperture radar (SAR) automatic target recognition (ATR) often encounter the challenge of data scarcity. The lack of sufficient labeled SAR image datasets leads to the consideration of using simulated data to supplement the dataset. On the one hand, electromagnetic computation simulations provide high amplitude accuracy but are inefficient for large-scale datasets due to their complex computations and physical models. On the other hand, ray tracing simulations offer high geometric accuracy and computational efficiency but struggle with low amplitude correctness, hindering accurate numerical feature extraction. Furthermore, the emergence of generative adversarial networks (GANs) provides a way to generate simulated datasets, trying to balance computational efficiency with image quality. Nevertheless, the simulated SAR images generated based on random noise lack constraints, and it is also difficult to generate images that exceed the parameter conditions of the real image’s training set. Hence, it is essential to integrate physics-based simulation techniques into GANs to enhance the generalization ability of the imaging parameters. In this paper, we present the SingleScene-SAR Simulator, an efficient framework for SAR image simulation that operates under limited real SAR data. This simulator integrates rasterized shooting and bouncing rays (SBR) with cycle GAN, effectively achieving both amplitude correctness and geometric accuracy. The simulated images are appropriate for augmenting datasets in target recognition networks. Firstly, the SingleScene-SAR Simulator employs a rasterized SBR algorithm to generate radar cross section (RCS) images of target models. Secondly, a specific training pattern for cycle GAN is established to translate noisy RCS images into simulated SAR images that closely resemble real ones. Finally, these simulated images are utilized for data augmentation. Experimental results based on the constructed dataset show that with only one scene SAR image containing 30 target chips, the SingleScene-SAR Simulator can efficiently produce simulated SAR images that exhibit high similarity in both spatial and statistical distributions compared with real images. By employing simulated SAR images for data augmentation, the accuracy of target recognition networks can be consistently and significantly enhanced.

Discriminating lung nodules as malignant or benign is still an underlying challenge. To address this challenge, radiologists need computer aided diagnosis (CAD) systems which can assist in learning discriminative imaging features corresponding to malignant and benign nodules. However, learning highly discriminative imaging features is an open problem. In this paper, our aim is to learn the most discriminative features pertaining to lung nodules by using an adversarial learning methodology. Specifically, we propose to use un-supervised learning with Deep Convolutional-Generative Adversarial Networks (DC-GANs) to generate lung nodule samples realistically. We hypothesize that imaging features of lung nodules will be discriminative if it is hard to differentiate them (fake) from real (true) nodules. To test this hypothesis, we present Visual Turing tests to two radiologists in order to evaluate the quality of the generated (fake) nodules. Extensive comparisons are performed in discerning real, generated, benign, and malignant nodules. This experimental set up allows us to validate the overall quality of the generated nodules, which can then be used to (1) improve diagnostic decisions by mining highly discriminative imaging features, (2) train radiologists for educational purposes, and (3) generate realistic samples to train deep networks with big data.

The generalizability of machine learning (ML) models for wearable monitoring in stroke rehabilitation is often constrained by the limited scale and heterogeneity of available data. Data augmentation addresses this challenge by adding computationally derived data to real data to enrich the variability represented in the training set. Traditional augmentation methods, such as rotation, permutation, and time-warping, have shown some benefits in improving classifier performance, but often fail to produce realistic training examples. This study employs Conditional Generative Adversarial Networks (cGANs) to create synthetic kinematic data from a publicly available dataset, closely mimicking the experimentally measured reaching movements of stroke survivors. This approach not only captures the complex temporal dynamics and common movement patterns after stroke, but also significantly enhances the training dataset. By training deep learning models on both synthetic and experimental data, we enhanced task classification accuracy: models incorporating synthetic data attained an overall accuracy of 80.0%, significantly higher than the 66.1% seen in models trained solely with real data. These improvements allow for more precise task classification, offering clinicians the potential to monitor patient progress more accurately and tailor rehabilitation interventions more effectively.

Kidneys are most easily segmented by convolutional neural networks (CNN) on contrast enhanced CT (CECT) images, but their segmentation accuracy may be reduced when only non-contrast CT (NCCT) images are available. The purpose of this work was to investigate the improvement in segmentation accuracy when implementing a generative adversarial network (GAN) to create virtual contrast enhanced (vCECT) images from non-contrast inputs. A 2D cycleGAN model, incorporating an additional idempotent loss function to restrict the GAN from making unnecessary modifications to data already in the translated domain, was trained to generate virtual contrast enhanced images on 286 paired non-contrast and contrast enhanced inputs. A 3D CNN trained on contrast enhanced images was applied to segment the kidneys in a test set of 20 paired non-contrast and contrast enhanced images. The non-contrast images were converted to virtual contrast enhanced images, then kidneys in both image conditions were segmented by the CNN. Segmentation results were compared to analyst annotations on non-contrast images visually and by Dice Coefficient (DC). Segmentation on virtual contrast enhanced images were more complete with fewer extraneous detections compared to non-contrast images in 16/20 cases. Mean(±SD) DC was 0.88(±0.80), 0.90(±0.03), and 0.95(±0.05) for non-contrast, virtual contrast enhanced, and real contrast enhanced, respectively. Virtual contrast enhancement visually improved segmentation quality, poor performing cases had their performance improved resulting in an overall reduction in DC variation, and the minimum DC increased from 0.65 to 0.85. This work provides preliminary results demonstrating the potential effectiveness of using a GAN for virtual contrast enhancement to improve CNN-based kidney segmentation on non-contrast images.

Credit card fraud detection is highly important to maintain financial security. However, it is challenging to train suitable models due to the class imbalance in credit card transaction data. To address this issue, this work proposes a novel deep learning framework, gated attention-based generative adversarial networks (GA-GAN) for credit card fraud detection in class-imbalanced data. GA-GAN integrates GAN and the gated attention mechanism to generate high-quality synthetic data that realistically simulates fraudulent behaviors. Experimental results on two public credit card datasets demonstrate that GA-GAN outperforms state-of-the-art methods on credit card fraud detection tasks in class-imbalanced data, indicating the advantage of GA-GAN. The code is publicly available at https://github.com/Gejiangmeng/gagan/tree/main.

Inferring the topology of a network from network dynamics is a significant problem with both theoretical research significance and practical value. This paper considers how to reconstruct the network topology according to the continuous-time data on the network. Inspired by the generative adversarial network(GAN), we design a deep learning framework based on network continuous-time data. The framework predicts the edge connection probability between network nodes by learning the correlation between network node state vectors. To verify the accuracy and adaptability of our method, we conducted extensive experiments on scale-free networks and small-world networks at different network scales using three different dynamics: heat diffusion dynamics, mutualistic interaction dynamics, and gene regulation dynamics. Experimental results show that our method significantly outperforms the other five traditional correlation indices, which demonstrates that our method can reconstruct the topology of different scale networks well under different network dynamics.

Brain source localization from electroencephalogram (EEG) signals is an challenging problem for noninvasively localizing the brain activity. Conventional methods use handcrafted regularization terms based on neural-physiological assumptions by exploiting the spatial-temporal structure on the source signals. In recent years, deep learning frameworks have demonstrated superior performance for solving the inverse problems in the natural and medical imaging field. This study proposes a novel unsupervised learning training-free framework based on Generative Adversarial Networks and deep image prior (GANs-DIP) as a generative model simulating spatially structured source signal. The proposed framework can faithfully recover extended source patches activation patterns of the brain in an unsupervised manner. Numerical experiments on a realistic brain model are performed under different levels of signal-to-noise ratio (SNR). The proposed model shows satisfactory performance in recovering the underlying source activation.

A deep learning model for FaceSwap and face-reenactment deepfakes detection