Search

Detecting and analyzing the local environment is crucial for investigating the dynamical processes of crystal nucleation and shape colloidal particle self-assembly. Recent developments in machine learning provide a promising avenue for better order parameters in complex systems that are challenging to study using traditional approaches. However, the application of machine learning to self-assembly on systems of particle shapes is still underexplored. To address this gap, we propose a simple, physics-agnostic, yet powerful approach that involves training a multilayer perceptron (MLP) as a local environment classifier for systems of particle shapes, using input features such as particle distances and orientations. Our MLP classifier is trained in a supervised manner with a shape symmetry-encoded data augmentation technique without the need for any conventional roto-translations invariant symmetry functions. We evaluate the performance of our classifiers on four different scenarios involving self-assembly of cubic structures, two-dimensional and three-dimensional patchy particle shape systems, hexagonal bipyramids with varying aspect ratios, and truncated shapes with different degrees of truncation. The proposed training process and data augmentation technique are both straightforward and flexible, enabling easy application of the classifier to other processes involving particle orientations. Our work thus presents a valuable tool for investigating self-assembly processes on systems of particle shapes, with potential applications in structure identification of any particle-based or molecular system where orientations can be defined. [ABSTRACT FROM AUTHOR]

Passive acoustic monitoring (PAM) is a powerful tool for studying ecosystems. However, its effective application in tropical environments, particularly for insects, poses distinct challenges. Neotropical katydids produce complex species-specific calls, spanning mere milliseconds to seconds and spread across broad audible and ultrasonic frequencies. However, subtle differences in inter-pulse intervals or central frequencies are often the only discriminatory traits. These extremities, coupled with low source levels and susceptibility to masking by ambient noise, challenge species identification in PAM recordings. This study aimed to develop a deep learning-based solution to automate the recognition of 31 katydid species of interest in a biodiverse Panamanian forest with over 80 katydid species. Besides the innate challenges, our efforts were also encumbered by a limited and imbalanced initial training dataset comprising domain-mismatched recordings. To overcome these, we applied rigorous data engineering, improving input variance through controlled playback re-recordings and by employing physics-based data augmentation techniques, and tuning signal-processing, model and training parameters to produce a custom well-fit solution. Methods developed here are incorporated into Koogu, an open-source Python-based toolbox for developing deep learning-based bioacoustic analysis solutions. The parametric implementations offer a valuable resource, enhancing the capabilities of PAM for studying insects in tropical ecosystems. This article is part of the theme issue 'Towards a toolkit for global insect biodiversity monitoring'. [ABSTRACT FROM AUTHOR]

This article introduces a novel 2D shape data augmentation approach based on intra-class shape space exploration. The proposed method relies on a geodesic interpolation between shapes, leveraging invariant-based morphing techniques. By blending a 2D shape pair belonging to a given class, we are able to generate nonlinear augmentations, hence covering more variations within the shape space. In particular, we formulate data augmentation as an optimization problem that minimizes the deformations between two shapes using the Generalized Finite Fourier Invariant Descriptor. The proposed augmentation technique is evaluated using numerous Convolution Neural Network architectures for 2D shape classification. The results indicate the superiority of the proposed method as compared to state-of-the-art techniques when considering small-scale datasets. [ABSTRACT FROM AUTHOR]

To solve the recommendation accuracy degradation problem encountered in sequential recommendation cases caused by data sparsity—such as short historical user behaviour sequences and limited information—this paper proposes a sequential recommendation model based on multipair contrastive learning with informative augmentation (IA-MPCL). The model aims to better learn user preference representations. Initially, a self-attention network is utilized to maintain the intrinsic relevance of the original sequences and introduce virtual interaction items for short sequences to achieve informative enhancement. Subsequently, multiple positive samples are generated by data augmentation methods to form multiple pairs of positive and negative samples. A multipair contrastive loss is constructed to eliminate the negative impact of fake positive and negative samples on the training process of the self-attention network. Finally, an adaptive loss weighting mechanism is proposed to dynamically regulate the role of the contrastive loss during multitask training. Through comparison experiments involving baseline methods and experiments conducted on datasets with different sparsity levels, the results show that IA-MPCL achieves significant improvements in terms of both recommendation accuracy and data sparsity resistance. [ABSTRACT FROM AUTHOR]

Bayesian shrinkage methods have been widely employed to perform variable selection with high-dimensional data. However, the presence of missing data hinders the implementation of these methods. Since complete case analyses can lead to biased estimates, applicable and efficient methods of variable selection with imputation are needed to obtain valid results. In order to address this issue, we propose an algorithm that employs the horseshoe shrinkage prior for shrinkage and multiple imputation for missing data in high-dimensional settings with a practical suggestion on model selection decision strategy. Simulation studies and real data analyses are presented and compared with those of other possible approaches. The simulation results suggest that the proposed algorithm can be considered as a general strategy for model selection of incomplete continuous data. [ABSTRACT FROM AUTHOR]

Sample selection model is a solution to eliminate the nonresponse bias. In some applications nonresponse is a multilevel variable with respect to its reasons of occurring. In these cases, the sample selection model can be extended such that a model to be considered for each of the nonresponse reasons. Also, in many cases, the reasons for nonresponse have priority over each other. In other words, it is not possible to observe all of the nonresponse reasons simultaneously. For example, in a survey with two noncontact and refusal reasons, noncontact has priority over refusal and refusal can be observed if the contact to the respondent can be established. For analyzing such extended model, a Bayesian inference approach with multiple selection rules using multivariate normal, inverse gamma and LKJ distributions as prior distributions for parameters and possibility of priority for nonresponse reasons is presented. Simulation studies are performed and an establishment survey data set is analyzed to demonstrate the performance of the proposed method. For sensitivity analysis of nonresponse on the parameters of interest, posterior displacement is applied. [ABSTRACT FROM AUTHOR]

Objective. Modern medical imaging plays a vital role in clinical practice, enabling non-invasive visualization of anatomical structures. Dynamic contrast enhancement (DCE) imaging is a technique that uses contrast agents to visualize blood flow dynamics in a time-resolved manner. It can be applied to different modalities, such as computed tomography (CT) and electrical impedance tomography (EIT). This study aims to develop a common theoretical and practical hemodynamic extraction basis for DCE modelling across modalities, based on the gamma-variate function. Approach. The study introduces a framework to generate time-intensity curves for multiple DCE imaging modalities from user-defined hemodynamic parameters. Thus, extensive datasets were simulated for both DCE-CT and EIT, representing different hemodynamic scenarios. Additionally, gamma-variate extensions to account for several physiological effects were detailed in a modality-agnostic manner, and three corresponding fitting strategies, namely nonlinear, linear, and a novel hybrid approach, were implemented and compared on the basis of accuracy of parameter estimation, first pass reconstruction, speed of computation, and failure rate. Main results. As a result, we found the linear method to be the most modality-dependent, exhibiting the greatest bias, variance and failure rates, although remaining the fastest alternative. The hybrid method at least matches the state-of-the-art nonlinear method's accuracy, while improving its robustness and speed by 10 times. Significance. Our research suggests that the hybrid method may bring noteworthy accuracy and efficiency improvements in handling the high-dimensionality of DCE imaging in general, being a step towards real-time processing. Moreover, our generative model presents a potential asset to produce benchmarking and data augmentation datasets across modalities. [ABSTRACT FROM AUTHOR]

Anomaly detection is a critical issue that needs to be addressed in large-scale industrial manufacturing. DeSTSeg integrates a pre-trained teacher network, a denoising student encoder–decoder, and a segmentation network into a single framework. However, this network faces two main challenges. Firstly, the supervision between the teacher and student networks is only focused on the same layers, lacking additional efficient information exchange, making it difficult for the student to effectively learn knowledge. Secondly, simple image-level perturbations struggle to cope with complex industrial scenes. To address these issues, we propose a high-performance anomaly detection network that combines knowledge review and feature-level perturbation, named KR-FP. To facilitate more efficient learning from the teacher network by the student, we introduce a knowledge review (KR) module to enhance the learning pathways of the teacher–student networks. This module comprises an attention-based fusion (ABF) module and a hierarchical context loss (HCL) module. The ABF module generates multi-scale fused features by integrating features from different levels. Then, the HCL module transfers knowledge into different levels of contextual information for interaction, thereby enhancing learning outcomes. Additionally, due to the limitations of image-level data augmentation methods in complex industrial environments, we introduce a Feature Perturbation (FP) module to enhance the model's ability to adapt to diverse and complex industrial scenes, significantly improving the model's stability and robustness. Experiments on the MVTec AD dataset have proven the effectiveness of our approach, achieving competitive results with a pixel-level average accuracy of 78.0% and an instance-level average accuracy of 78.8%. Source code and pretrained models are available at https://github.com/chongyouxiaolu/KR-FP. [ABSTRACT FROM AUTHOR]

Objective: To provide a brief overview of artificial intelligence (AI) application within the field of eating disorders (EDs) and propose focused solutions for research. Method: An overview and summary of AI application pertinent to EDs with focus on AI's ability to address issues relating to data sharing and pooling (and associated privacy concerns), data augmentation, as well as bias within datasets is provided. Results: In addition to clinical applications, AI can utilize useful tools to help combat commonly encountered challenges in ED research, including issues relating to low prevalence of specific subpopulations of patients, small overall sample sizes, and bias within datasets. Discussion: There is tremendous potential to embed and utilize various facets of artificial intelligence (AI) to help improve our understanding of EDs and further evaluate and investigate questions that ultimately seek to improve outcomes. Beyond the technology, issues relating to regulation of AI, establishing ethical guidelines for its application, and the trust of providers and patients are all needed for ultimate adoption and acceptance into ED practice. Public Significance: Artificial intelligence (AI) offers a promise of significant potential within the realm of eating disorders (EDs) and encompasses a broad set of techniques that offer utility in various facets of ED research and by extension delivery of clinical care. Beyond the technology, issues relating to regulation, establishing ethical guidelines for application, and the trust of providers and patients are needed for the ultimate adoption and acceptance of AI into ED practice. [ABSTRACT FROM AUTHOR]

Software defect prediction (SDP) aims to distinguish between defective and nondefective instances, but the imbalance between these two classes often leads to reduced prediction performance. Conventional SDP approaches use oversampling techniques, such as synthetic oversampling, to tackle the problem of imbalanced data. However, these methods merely synthesize new instances based on traditional code features without considering actual defects at the code level. To address the issue of data imbalance while preserving semantic features of code samples, a mutation‐based data augmentation approach in SDP is proposed. The method utilizes the mutation operator to generate mutants that mutate nondefective instances and create new defective instances. Six projects from the PROMISE dataset are used to evaluate the approach, employing four traditional and two deep classifiers. The experimental results demonstrate the effectiveness of this method in improving defect prediction performance for both traditional and deep classifiers compared with other data augmentation methods. A novel mutation‐based data augmentation method is proposed, in which data are increased at the code level while preserving its semantic features. The method utilizes the mutation operator for generating mutants to mutate against nondefective instances and then generates new defective instances. Experimental results demonstrate that this approach achieves the best average overall performance. [ABSTRACT FROM AUTHOR]

Background: Cone‐beam computed tomography (CBCT) images provide high‐resolution insights into the underlying craniofacial anomaly in patients with cleft lip and palate (CLP), requiring non‐negligible annotation costs to measure the cleft defect for the guidance of the clinical secondary alveolar bone graft procedures. Considering the cumbersome volumetric image acquisition, there is a lack of paired CLP CBCTs and normal CBCTs for learning‐based anatomical structure restoration models. Nowadays, the registration‐based method relieves the annotation burden, though one‐shot registration and the regular mask are limited to handling fine‐grained shape variations and harmony between restored bony tissues and the defected maxilla. Purpose: This study aimed to design and evaluate a novel method for deformable partial registration of the CLP CBCTs and normal CBCTs, enabling personalized maxilla completion and cleft defect volume prediction from CLP CBCTs. Methods: We proposed an adaptable deep registration framework for personalized maxilla completion and cleft defect volume prediction from CLP CBCTs. The key ingredient was a cascaded partial registration to exploit the maxillary morphology prior and attribute transfer. Cascaded registration with coarse‐to‐fine registration fields handled morphological variations of cleft defects and fine‐grained maxillary restoration. We designed an adaptable cleft defect mask and volumetric Boolean operators for reliable voxel filling of the defected maxilla. A total of 36 clinically obtained CLP CBCTs were used to train and validate the proposed model, among which 22 CLP CBCTs were used to generate a training dataset with 440 synthetic CBCTs by B‐spline deformation‐based data augmentation and the remaining for testing. The proposed model was evaluated on maxilla completion and cleft defect volume prediction from clinically obtained unilateral and bilateral CLP CBCTs. Results: Extensive experiments demonstrated the effectiveness of the adaptable cleft defect mask and the cascaded partial registration on maxilla completion and cleft defect volume prediction. The proposed method achieved state‐of‐the‐art performances with the Dice similarity coefficient of 0.90 ±$\pm$ 0.02 on the restored maxilla and 0.84 ±$\pm$ 0.04 on the estimated cleft defect, respectively. The average Hausdorff distance between the estimated cleft defect and the manually annotated ground truth was 0.30 ±$\pm$ 0.08 mm. The relative volume error of the cleft defect was 0.09±$0.09\pm$0.08. The proposed model allowed for the prediction of cleft defect maps that were in line with the ground truth in the challenging unilateral and bilateral CLP CBCTs. Conclusions: The results suggest that the proposed adaptable deep registration model enables patient‐specific maxilla completion and automatic annotation of cleft defects, relieving tedious voxel‐wise annotation and image acquisition burdens. [ABSTRACT FROM AUTHOR]

Currently, traffic sign recognition techniques have been brought into the assistive driving of automobiles. However, small traffic sign recognition in real scenes is still a challenging task due to the class imbalance issue and the size limit of the traffic signs. To address the above issues, a feature‐enhanced hybrid attention network is proposed based on YOLOv5s for a small, fast, and accurate traffic sign detector. First, a series of online data augmentation strategies are designed in the preprocessing module for the model training. Second, the hybrid channel and spatial attention module CSAM are integrated into the backbone for a better feature extraction ability. Third, the channel attention module CAM is used in the detection head for a more efficient feature fusion ability. To validate the approach, extensive experiments are conducted based on the Tsinghua‐Tencent 100K dataset. It is found that the novel method achieves state‐of‐the‐art performance with only negligible increases in the model parameter and computational overhead. Specifically, the mAP@0.5$mAP@0.5$, parameters, and FLOPs are 85.8%, 7.13 M, and 16.1 G, respectively. [ABSTRACT FROM AUTHOR]

Perception systems in autonomous vehicles need to accurately detect and classify objects within their surrounding environments. Numerous types of sensors are deployed on these vehicles, and the combination of such multimodal data streams can significantly boost performance. The authors introduce a novel sensor fusion framework using deep convolutional neural networks. The framework employs both camera and LiDAR sensors in a multimodal, multiview configuration. The authors leverage both data types by introducing two new innovative fusion mechanisms: element‐wise multiplication and multimodal factorised bilinear pooling. The methods improve the bird's eye view moderate average precision score by +4.97% and +8.35% on the KITTI dataset when compared to traditional fusion operators like element‐wise addition and feature map concatenation. An in‐depth analysis of key design choices impacting performance, such as data augmentation, multi‐task learning, and convolutional architecture design is offered. The study aims to pave the way for the development of more robust multimodal machine vision systems. The authors conclude the paper with qualitative results, discussing both successful and problematic cases, along with potential ways to mitigate the latter. [ABSTRACT FROM AUTHOR]

This paper proposes a method called Meta Generative Data Augmentation Optimization (MGDAO) to overcome limited types of operations for the policy-based automatic data augmentation method. While traditional data augmentation methods have relied on expert intuition to determine effective transformations, recent approaches have attempted to generate data augmentation strategies automatically. However, these automatic methods can still suffer from limited operation sets and difficulty training conditional generative models. To address these issues, MGDAO replaces the limited operations space in the AutoAugment series with a deep-style generator and replaces the discriminator in a generative adversarial model with the validation loss of the target model. These replacements released fixed image operations and made MGDAO useful for sequential data. The generator learns to transform the data from the training domain to the validation data domain. It is further used to generate augmented samples to train the target model and reduce the validation loss. Experiments on classification benchmarks of few-shot image and text-based datasets show that MGDAO achieves competitive results compared to data auto-augmentation methods. [ABSTRACT FROM AUTHOR]

This study aims to improve the recognition rate of a recognizer for detecting cracks on concrete surfaces from image data. When developing a recognizer, a clean image is commonly used as training data. Therefore, the recognition rate decreases when blurred images not taken cleanly are used as test data. To improve the recognition rate, the training data were mixed with processed images. In experiment, a simple convolutional neural network was used to confirm the effect of data augmentation. Experimental results showed that adding blurred images to the training data improved the recognition rate for blurred images, regardless of how the images were blurred. For training data without blurred images, the recognition rate for strongly blurred images is about 75%. On the other hand, the recognition rate improves to about 95% for training data that includes strongly blurred images. Thus, including images in which strong blur filters are used for data augmentation is effective in improving the recognition rate of blurred images. [ABSTRACT FROM AUTHOR]

Data augmentation (DA) is an effective way to improve the performance of deep networks. Unfortunately, current methods are mostly developed for high-level vision tasks (eg, image classification) and few are studied for low-level (eg, image restoration). In this paper, we provide a comprehensive analysis of the existing DAs in the frequency domain. We find that the methods that largely manipulate the spatial information can hinder the image restoration process and hurt the performance. Based on our analyses, we propose CutBlur and mixture-of-augmentation (MoA). CutBlur cuts a low-quality patch and pastes it to the corresponding high-quality image region, or vice versa. The key intuition is to provide enough DA effect while keeping the pixel distribution intact. This characteristic of CutBlur enables a model to learn not only "how" but also "where" to reconstruct an image. Eventually, the model understands "how much" to restore given pixels, which allows it to generalize better to unseen data distributions. We further improve the restoration performance by MoA that incorporates the curated list of DAs. We demonstrate the effectiveness of our methods by conducting extensive experiments on several low-level vision tasks on both single or a mixture of distortion tasks. Our results show that CutBlur and MoA consistently and significantly improve the performance especially when the model size is big and the data is collected under real-world environments. Our code is available at https://github.com/clovaai/cutblur. [ABSTRACT FROM AUTHOR]

The breakthrough brought by generative adversarial networks (GANs) in computer vision (CV) applications has gained a lot of attention in different fields due to their ability to capture the distribution of a dataset and generate high-quality similar images. From one side, this technology has been rapidly adopted as an alternative to traditional applications and introduced novel perspectives in data augmentation, domain transfer, image expansion, image restoration, image segmentation, and super-resolution. From another side, we found that due to the lack of industrial datasets and the limitation for acquiring and accurately annotating new images, GANs form an exciting solution to generate new industrial image datasets or to restore and augment existing ones. Therefore, we introduce a review of the latest trend in GANs applications and project them in industrial use cases. We conducted our experiments with synthetic images and analyzed most of GAN's failures and image artifacts to provide training's best practices. [ABSTRACT FROM AUTHOR]

Research on the data-driven health state estimation of lithium-ion batteries has gained significant attention in recent years. However, the practical implementation of obtaining one data point in one cycle has resulted in poor data quality, leading to low accuracy and prediction instability. To overcome this challenge, a two-step approach is proposed. First, available data are enriched using Akima spline curve interpolation, and the overall degradation trend of the battery is extracted as a Sigmoid function, enhancing the data quality. Second, a parallel network model that combines the strengths of the convolutional neural network (CNN) and the long short-term memory network (LSTM) is introduced. This model leverages the ability of one-dimensional convolutional neural network (1DCNN) to effectively capture local features and proficiency of the LSTM in capturing long-term dependencies. By employing this hybrid model, a better understanding and prediction of the remaining battery life is achieved. Finally, based on the NASA public battery dataset, expanded and decomposed data are trained and predicted by the parallel network model. Experimental results demonstrate that the proposed method exhibits high accuracy and strong generalization capability. [ABSTRACT FROM AUTHOR]

Timely and accurate estimates of sugarcane yield provide valuable information for food management, bio-energy production, (inter)national trade, industry planning and government policy. Remote sensing and machine learning approaches can improve sugarcane yield estimation. Previous attempts have however often suffered from too few training samples due to the fact that field data collection is expensive and time-consuming. Our study demonstrates that unmanned aerial vehicle (UAV) data can be used to generate field-level yield data using only a limited number of field measurements. Plant height obtained from RGB UAV-images was used to train a model to derive intra-field yield maps based on 41 field sample plots spread over 20 sugarcane fields in the Udon Thani Province, Thailand. The yield maps were subsequently used as reference data to train another model to estimate yield from multi-spectral Sentinel-2 (S2) imagery. The integrated UAV yield and S2 data was found efficient with RMSE of 6.88 t/ha (per 10 m × 10 m pixel), for average yields of about 58 t/ha. The expansion of the sugarcane yield mapping across the entire region of 11,730 km2 was in line with the official statistical yield data and highlighted the high spatial variability of yields, both between and within fields. The presented method is a cost-effective and high-quality yield mapping approach which provides useful information for sustainable sugarcane yield management and decision-making. [ABSTRACT FROM AUTHOR]

Data augmentation methods are crucial to improve the accuracy of densely occluded object recognition in the scene where the quantity and diversity of training images are insufficient. However, the current methods that use regional dropping and mixing strategies suffer from the problem of missing foreground objects and redundant background features, which can lead to densely occluded object recognition issues in classification or detection tasks. Herein, saliency information and mosaic based data augmentation method for densely occluded object recognition is proposed, which utilizes saliency information as prior knowledge to supervise the mosaic process of training images containing densely occluded objects. And the method uses fogging processing and class label mixing to construct new augmented images, in order to improve the accuracy of image classification and object recognition tasks by augmenting the quantity and diversity of training images. Extensive experiments on different classification datasets with various CNN architectures prove the effectiveness of our method. [ABSTRACT FROM AUTHOR]

This work presents Spacecraft Pose Network v2 (SPNv2), a Convolutional Neural Network (CNN) for pose estimation of noncooperative spacecraft across domain gap. SPNv2 is a multi-scale, multi-task CNN which consists of a shared multi-scale feature encoder and multiple prediction heads that perform different tasks on a shared feature output. These tasks are all related to detection and pose estimation of a target spacecraft from an image, such as prediction of pre-defined satellite keypoints, direct pose regression, and binary segmentation of the satellite foreground. It is shown that by jointly training on different yet related tasks with extensive data augmentations on synthetic images only, the shared encoder learns features that are common across image domains that have fundamentally different visual characteristics compared to synthetic images. This work also introduces Online Domain Refinement (ODR) which refines the parameters of the normalization layers of SPNv2 on the target domain images online at deployment. Specifically, ODR performs self-supervised entropy minimization of the predicted satellite foreground, thereby improving the CNN's performance on the target domain images without their pose labels and with minimal computational efforts. The GitHub repository for SPNv2 is available at https://github.com/tpark94/spnv2. [ABSTRACT FROM AUTHOR]

Graph semi-supervised learning (GSSL) plays an important role in semi-supervised classification by leveraging the similarity of graph topology and convex optimization with Laplacian-based regularization. Current approaches mainly focus on data representation in the Euclidean space and solely rely on a single graph structure, ignoring the underlined manifold-valued structure of the data. To this end, we propose a novel manifold-based multi-graph embedding (M2GE), which explicitly learns representations on both Riemannian manifold and Euclidean space in the embedding learning process. Incorporating multi-graph structure, the rich and varied features are used by M2GE to more fully capture categorical information, where the heterogeneous spatial representations from manifold are complementary. Extensive experimental results on three popular benchmarks verify demonstrate the superiority of our method over state-of-the art competitors. • A novel manifold-based multi-graph embedding (M2GE) is proposed for semi-supervised classification. • Multi-sampling augmentation is employed to improve the generalization of the model. • M2GE effectively improves the accuracy rate of the semi-supervised classification. [ABSTRACT FROM AUTHOR]

The navigation safety of polar ships depends on accurate monitoring of ice conditions along the route. Complex polar environments but scarce ice datasets pose challenges for deep learning‐based object detection methods. To address this issue, we propose an ice monitoring method based on polar environment adaptive image enhancement. This approach adaptively augments the ice dataset by simulating polar environments such as snowflakes and aurora to improve the training precision of ice detection models in harsh polar environments and meet the needs of safe navigation of polar ships. First, an image augmentation method based on randomly generated snowflake textures is proposed to address the difficulty of covering polar storm scenarios in existing ice datasets, which leads to weak generalization of ice monitoring models. Second, targeting the problem that the boundary between ice and background is confused by aurora phenomena, which severely reduces the ice monitoring accuracy, we propose an image augmentation method based on aurora simulations to improve the ice monitoring performance of trained models under aurora. Finally, an adaptive image augmentation strategy based on the brightness‐contrast threshold is proposed to optimize the snowflake density and aurora intensity in the augmented images to improve the discrimination of ice boundaries. Experimental results demonstrate the effectiveness of the proposed method, which improves the ice state detection accuracy from 0.640 to 0.950, thus providing a reliable guarantee for the safe navigation of polar ships. [ABSTRACT FROM AUTHOR]

Many longitudinal studies are designed to monitor participants for major events related to the progression of diseases. Data arising from such longitudinal studies are usually subject to interval censoring since the events are only known to occur between two monitoring visits. In this work, we propose a new method to handle interval‐censored multistate data within a proportional hazards model framework where the hazard rate of events is modeled by a nonparametric function of time and the covariates affect the hazard rate proportionally. The main idea of this method is to simplify the likelihood functions of a discrete‐time multistate model through an approximation and the application of data augmentation techniques, where the assumed presence of censored information facilitates a simpler parameterization. Then the expectation‐maximization algorithm is used to estimate the parameters in the model. The performance of the proposed method is evaluated by numerical studies. Finally, the method is employed to analyze a dataset on tracking the advancement of coronary allograft vasculopathy following heart transplantation. [ABSTRACT FROM AUTHOR]

Establishing a reliable credit card fraud detection model has become a primary focus for academia and the financial industry. The existing anti-fraud methods face challenges related to low recall rates, inaccurate results, and insufficient causal modeling ability. This paper proposes a credit card fraud detection model based on counterfactual data enhancement of the triplet network. Firstly, we convert the problem of generating optimal counterfactual explanations (CFs) into a policy optimization of agents in the discrete–continuous mixed action space, thereby ensuring the stable generation of optimal CFs. The triplet network then utilizes the feature similarity and label difference of positive example samples and CFs to enhance the learning of the causal relationship between features and labels. Experimental results demonstrate that the proposed method improves the accuracy and robustness of the credit card fraud detection model, outperforming existing methods. The research outcomes are of significant value for both credit card anti-fraud research and practice while providing a novel approach to causal modeling issues across other fields. [ABSTRACT FROM AUTHOR]

Creep-feed grinding is a high-efficiency, high-precision grinding process widely used in the manufacturing of aviation engines. However, the workpiece burn and other quality issues caused by high processing temperature limit the yield of grinding. Therefore, the accurate model of grinding temperature has become the key to improving processing efficiency and quality. Different from the traditional physical or data-driven models, this paper attempts to combine both perspectives based on Physics-Guided Neural Networks (PGNN) to accurately predict grinding temperature with a small number of experiments. At the level of data acquisition, real grinding experiment data was obtained and a data augmentation method had been proposed. At the level of neural network structure, optimisation processes were implemented to enhance prediction performance, and a physics-guided loss function was inserted to guide network training. The experiment results shows that PGNN had better prediction accuracy than the physical model, while also mitigating the limitations of data-driven models on small sample sets. PGNN also performed better with noisy data and predictions out of the training data range, this reveals the benefits of PGNN for small sample problems in processing scenarios. [ABSTRACT FROM AUTHOR]

Hydrogen production for clean energy is gaining a foothold, notably through the gasification of biomass. Machine learning aids in its accurate production predictions, yet its opaque nature limits explanation. This study looked into machine learning models' transparency and evaluated the impact of data augmentation on hydrogen production prediction accuracy using a publicly available numerical dataset. Fifteen (15) machine learning regression models including linear and tree-based were evaluated leading to the introduction of a hybrid model. The proposed hybrid model is based on ensemble and averaging techniques in mathematics. The deployed model's prediction was evaluated using the Mean Absolute Error (MAE), the Mean Square Error (MSE), the Root Mean Squared Error (RMSE), the Root Mean Squared Log Error (RMSLE), and R-squared (R2) based on K-Fold cross-validation. Furthermore, the effect of the grid search, random search, and Bayesian optimization techniques for model hyperparameter optimization were carried out. The evaluated models demonstrate superior predictions and explore data augmentation's potential in refining machine learning accuracy with the proposed hybrid model recording an MAE of 2.187, MSE of 9.788, RMSE of 3.129, R2 of 0.250, MSLE of 0.005, and RMSLE of 0.070 against the original data result of MAE of 2.665, MSE of 11.877, RMSE of 3.446, R2 of 0.112, MSLE of 0.006 and RMSLE of 0.076. However, when the proposed model hyperparameter training setting was optimized, the yielded result improved further with an MAE of 0.044, MSE of 0.040, RMSE of 0.215, R2 of 0.996, MSLE of 0.000 and RMSLE of 0.005. By unraveling the inner workings of black box models using the SHAP, LIME, and ELI-5 interpretable AI models, this study advances hydrogen production prediction using biomass gasification. Insights gained in this study foster the development of more transparent and effective models, aiming for the widespread adoption of sustainable hydrogen energy. [Display omitted] • An in-depth analysis of 15 machine-learning (ML) regression models for hydrogen production prediction (HPP) is presented. • An ensemble model was developed to enhance the accuracy and robustness of hydrogen production prediction. • A comprehensive evaluation of the impact of data augmentation techniques on ML models for HPP. • A comprehensive evaluation of the impact of Hyperparameter optimization techniques on ML models for HPP. • HPP influential factors and insights were presented and analyzed using Explainable AI. [ABSTRACT FROM AUTHOR]

In this study, we propose a generative data augmentation technique to overcome the challenges of severely limited data when designing a deep learning-based automated strabismus diagnosis system. We implement a generative model based on the StyleGAN2-ADA model for system design and assess strabismus classification performance using two classifiers. We evaluate the capability of our proposed method against traditional data augmentation techniques and confirm a substantial enhancement in performance. Furthermore, we conduct experiments to explore the relationship between the diagnosis agreement among ophthalmologists and the generation performance of the generative model. Beyond FID, we validate the generative samples on the classifier to establish their practicality. Through these experiments, we demonstrate that the generative model-based data augmentation improves overall quantitative performance in scenarios of extreme data scarcity and effectively mitigates overfitting issues during deep learning model training. [ABSTRACT FROM AUTHOR]

Recent studies have highlighted the significance of cellular metabolism in the initiation of clonal expansion and effector differentiation of T cells. Upon exposure to antigens, naïve CD4+ T cells undergo metabolic reprogramming to meet their metabolic requirements. However, only few studies have simultaneously evaluated the changes in protein and metabolite levels during T cell differentiation. Our research seeks to fill the gap by conducting a comprehensive analysis of changes in levels of metabolites, including sugars, amino acids, intermediates of the TCA cycle, fatty acids, and lipids. By integrating metabolomics and proteomics data, we discovered that the quantity and composition of cellular lipids underwent significant changes in different effector Th cell subsets. Especially, we found that the sphingolipid biosynthesis pathway was commonly activated in Th1, Th2, Th17, and iTreg cells and that inhibition of this pathway led to the suppression of Th17 and iTreg cells differentiation. Additionally, we discovered that Th17 and iTreg cells enhance glycosphingolipid metabolism, and inhibition of this pathway also results in the suppression of Th17 and iTreg cell generation. These findings demonstrate that the utility of our combined metabolomics and proteomics analysis in furthering the understanding of metabolic transition during Th cell differentiation. Using metabolomics and proteomics data, the quantity and composition of cellular lipid underwent notable changes in different eff Th cell subsets. Especially, Th17 and iTreg cells enhance glycosphingolipid metabolism to sustain their differentiation. [ABSTRACT FROM AUTHOR]

This article, published in the journal JCO Precision Oncology, discusses the use of machine learning models and cautions for analyzing censored time-to-event outcomes in clinical research. The authors highlight common errors and provide advice for potential authors. They emphasize the difference between machine learning models and conventional statistical models, the importance of using the most appropriate method rather than the fanciest or most accessible, and the need to define comparison groups and time zero in survival analysis. The article also discusses the limitations of deep learning models for tabular data and the importance of reporting model performance measures with corresponding confidence intervals. The authors conclude by offering general recommendations for data analysis in research. [Extracted from the article]

The prediction of depression is a crucial area of research which makes it one of the top priorities in mental health research as it enables early intervention and can lead to higher success rates in treatment. Self-reported feelings by patients represent a valuable biomarker for predicting depression as they can be expressed in a lower-dimensional network form, offering an advantage in visualizing the interactive characteristics of depression-related feelings. Furthermore, the network form of data expresses high-dimensional data in a compact form, making the data easy to use as input for the machine learning processes. In this study, we applied the graph convolutional network (GCN) algorithm, an effective machine learning tool for handling network data, to predict depression-prone patients using the network form of self-reported log data as the input. We took a data augmentation step to expand the initially small dataset and fed the resulting data into the GCN algorithm, which achieved a high level of accuracy from 86–97% and an F1 (harmonic mean of precision and recall) score of 0.83–0.94 through three experimental cases. In these cases, the ratio of depressive cases varied, and high accuracy and F1 scores were observed in all three cases. This study not only demonstrates the potential for predicting depression-prone patients using self-reported logs as a biomarker in advance, but also shows promise in handling small data sets in the prediction, which is critical given the challenge of obtaining large datasets for biomarker research. The combination of self-reported logs and the GCN algorithm is a promising approach for predicting depression and warrants further investigation. [ABSTRACT FROM AUTHOR]

Self-supervised learning is a new machine learning method that does not rely on manually labeled data, and learns from rich unlabeled data itself by designing agent tasks using the input data as supervision to obtain a more generalized representation for application in downstream tasks. However, the current self-supervised learning suffers from the problem of relying on the selection and number of negative samples and the problem of sample bias phenomenon after graph data augmentation. In this paper, we investigate the above problems and propose a corresponding solution, proposing a graph contrastive learning algorithm without negative samples. The model uses matrix sketching in the implicit space for feature augmentation to reduce sample bias and iteratively trains the mutual correlation matrix of two viewpoints by drawing closer to the distance of the constant matrix as the objective function. This method does not require techniques such as negative samples, gradient stopping, and momentum updating to prevent self-supervised model collapse. This method is compared with 10 graph representation learning algorithms on four datasets for node classification tasks, and the experimental results show that the algorithm proposed in this paper achieves good results. [ABSTRACT FROM AUTHOR]

Traffic sign recognition plays a crucial role in enhancing the safety and efficiency of traffic systems. However, in snowy conditions, traffic signs are often obscured by particles, leading to a severe decrease in detection accuracy. To address this challenge, we propose an improved YOLOv8-based model for traffic sign recognition. Initially, we introduce a Multi-Scale Group Convolution (MSGC) module to replace the C2f module in the YOLOv8 backbone. Data indicate that MSGC enhances detection accuracy while maintaining model lightweightness. Subsequently, to improve the recognition ability for small targets, we introduce an enhanced small target detection layer, which enhances small target detection accuracy while reducing parameters. In addition, we replaced the original BCE loss with the improved EfficientSlide loss to improve the sample imbalance problem. Finally, we integrate Deformable Attention into the model to improve the detection efficiency and performance of complex targets. The resulting fused model, named MSGC-YOLOv8, is evaluated on an enhanced dataset of snow-covered traffic signs. Experimental results show that the MSGC-YOLOv8 model is used for snow road traffic sign recognition. Compared with the YOLOv8n model mAP@0.5:0.95, mAP@0.5:0.95 is increased by 17.7% and 18.1%, respectively, greatly improving the detection accuracy. Compared with the YOLOv8s model, while the parameters are reduced by 59.6%, mAP@0.5 only loses 1.5%. Considering all aspects of the data, our proposed model shows high detection efficiency and accuracy under snowy conditions. [ABSTRACT FROM AUTHOR]

The domain of human locomotion identification through smartphone sensors is witnessing rapid expansion within the realm of research. This domain boasts significant potential across various sectors, including healthcare, sports, security systems, home automation, and real-time location tracking. Despite the considerable volume of existing research, the greater portion of it has primarily concentrated on locomotion activities. Comparatively less emphasis has been placed on the recognition of human localization patterns. In the current study, we introduce a system by facilitating the recognition of both human physical and location-based patterns. This system utilizes the capabilities of smartphone sensors to achieve its objectives. Our goal is to develop a system that can accurately identify different human physical and localization activities, such as walking, running, jumping, indoor, and outdoor activities. To achieve this, we perform preprocessing on the raw sensor data using a Butterworth filter for inertial sensors and a Median Filter for Global Positioning System (GPS) and then applying Hamming windowing techniques to segment the filtered data. We then extract features from the raw inertial and GPS sensors and select relevant features using the variance threshold feature selection method. The extrasensory dataset exhibits an imbalanced number of samples for certain activities. To address this issue, the permutation-based data augmentation technique is employed. The augmented features are optimized using the Yeo–Johnson power transformation algorithm before being sent to a multi-layer perceptron for classification. We evaluate our system using the K-fold cross-validation technique. The datasets used in this study are the Extrasensory and Sussex Huawei Locomotion (SHL), which contain both physical and localization activities. Our experiments demonstrate that our system achieves high accuracy with 96% and 94% over Extrasensory and SHL in physical activities and 94% and 91% over Extrasensory and SHL in the location-based activities, outperforming previous state-of-the-art methods in recognizing both types of activities. [ABSTRACT FROM AUTHOR]

Methane steam reforming is the foremost method for hydrogen production, and it has been studied through experiments and diverse computational models to enhance its energy efficiency. This study focuses on employing an artificial neural network as a model of the methane steam reforming process. The proposed data-driven model predicts the output mixture's composition based on reactor operating conditions, such as the temperature, steam-to-methane ratio, nitrogen-to-methane ratio, methane flow, and nickel catalyst mass. The network, a feedforward type, underwent training with a comprehensive dataset augmentation strategy that augments the primary experimental dataset through interpolation and theoretical simulations of the process, ensuring a robust model training phase. Additionally, it introduces weights to evaluate the relative significance of different data categories (experimental, interpolated, and theoretical) within the dataset. The optimal artificial neural network architecture was determined by evaluating various configurations, with the aim of minimizing the mean squared error (0.00022) and maximizing the Pearson correlation coefficient (0.97) and Spearman correlation coefficient (1.00). [ABSTRACT FROM AUTHOR]

Energy management systems allow the Smart Grids industry to track, improve, and regulate energy use. Particularly, demand-side management is regarded as a crucial component of the entire Smart Grids system. Therefore, by aligning utility offers with customer demand, anticipating future energy demands is essential for regulating consumption. An updated examination of several forecasting techniques for projecting energy short-term load forecasts is provided in this article. Each class of algorithms, including statistical techniques, Machine Learning, Deep Learning, and hybrid combinations, are comparatively evaluated and critically analyzed, based on three real consumption datasets from Spain, Germany, and the United States of America. To increase the size of tiny training datasets, this paper also proposes a data augmentation technique based on Generative Adversarial Networks. The results show that the Deep Learning-hybrid model is more accurate than traditional statistical methods and basic Machine Learning procedures. In the same direction, it is demonstrated that more comprehensive datasets assisted by complementary data, such as energy generation and weather, may significantly boost the accuracy of the models. Additionally, it is also demonstrated that Generative Adversarial Networks-based data augmentation may greatly improve algorithm accuracy. [ABSTRACT FROM AUTHOR]

Ship detection in synthetic aperture radar (SAR) images is widely applied in marine monitoring. In recent years, convolutional neural networks (CNNs) have made significant advances in SAR ship detection. However, the scene imbalance problem, i.e. much more offshore scenes than inshore scenes in SAR images, severely hampers the improvement of detection accuracy. In this article, we propose scene-aware data augmentation to address the problem. First, a novel concept, inshore rate, is introduced to estimate the land area percentage of various scenes in SAR images. We can divide image samples into inshore and offshore ones based on the inshore rate without manual category annotations. Second, to augment inshore samples, we synthesize dummy inshore images by cropping the ground truth bounding boxes (GT bboxes) from offshore images and pasting them onto inshore images. This offline data augmentation strategy can address the scene imbalance problem and improve detection accuracy. The experiments conducted on SAR Ship Detection Dataset (SSDD) and Large-Scale SAR Ship Detection Dataset (LS-SSDD) demonstrate the effectiveness of the proposed method in improving detection performance. [ABSTRACT FROM AUTHOR]

Aspect-based sentiment analysis (ABSA) constitutes a significant field within natural language processing (NLP). This study proposes a multi-strategy text data augmentation methodology to overcome challenges such as limited dataset sizes and the absence of comprehensive, high-quality corpora in aspect-level sentiment classification (ASC). Specifically, it expands the SemEval 2014 Restaurant and Laptop training datasets from 3017 to 4646 instances and from 1864 to 3693 instances, respectively. The methodology encompasses both word-level and sentence-level augmentations. Evaluations using advanced deep learning techniques, including Att-LSTM, SVM, CNN, and Bi-LSTM, were conducted on the enhanced SemEval 2014 Task 4 restaurant dataset and laptops dataset. The test dataset sizes are 920 and 668, respectively. The Att-LSTM, demonstrating superior performance, recorded F1 score improvements of 5.0% and 4.4% on the restaurant and laptop datasets, respectively, following the application of the multi-strategy augmentation methodology compared to others. This approach significantly enlarges the dataset and improves performance in ASC tasks. [ABSTRACT FROM AUTHOR]

Surface defect detection is an essential task for ensuring the quality of products. Many excellent object detectors have been employed to detect surface defects in resent years, which has achieved outstanding success. To further improve the detection performance, a defect detector based on state‐of‐the‐art YOLOv8, named improved YOLOv8 by neck, head and data (NHD‐YOLO), is proposed. Specifically, YOLOv8 from three crucial aspects including neck, head and data is improved. First, a shortcut feature pyramid network is designed to effectively fuse features from backbone by improving the information transmission. Then, an adaptive decoupled head is proposed to alleviate the feature spatial misalignment between the classification and regression tasks. Finally, to enhance the training on small objects, a data augmentation method named selective small object copy and paste is proposed. Extensive experiments are conducted on three real‐world datasets: detection dataset from Northeastern University (NEU‐DET), printed circuit boards from Peking University (PKU‐Market‐PCB) and common objects in context (COCO). According to the results, NHD‐YOLO achieves the highest detection accuracy and exhibits outstanding inference speed and generalisation performance. [ABSTRACT FROM AUTHOR]

Accurate prediction of blast loading from accidental hydrogen-air cloud explosion is critical for the planning, design, and operation of the hydrogen industry. This study proposes an efficient and accurate prediction model of explosion loads from hydrogen-air cloud explosions in vented silos. Based on the Bayesian regularization and data augmentation, an artificial neural network (ANN) model is generated and trained with the data collected from computational fluid dynamics (CFD) simulations using FLACS. The influences of various parameters including the silo dimensions (diameter and height), the ventilation size (width and length), and the ventilation panel activation pressures are considered. Analysis shows that the R2 value of the ANN model predictions and CFD simulation results exceeds 0.99. The developed ANN model could provide almost instantaneous prediction of the peak overpressure, the rise rate of overpressure, and the impulse for constructing the blast loading time history, which can be used in the design analysis and safety assessment of structures exposed to hydrogen explosions. Additionally, the model interpretability was conducted to analyse the importance and contribution of the input variables, which also verifies the reliability of the prediction model. • A comprehensive ANN model with interpretability was developed to predict Hydrogen-Air explosion loads. • Over 400 data sets for ANN training are generated via CFD simulation using FLACS Hydrogen Modular. • The performance of the developed ANN model closely aligns with CFD simulation results. • The developed model combines the advantages of simulations, empirical methods, and neural networks. [ABSTRACT FROM AUTHOR]

Tailoring a deep convolutional neural network (DCNN) is a tedious and time-consuming task in the field of medical image analysis. In this research paper, Deep-generative adversial neural network (Deep-GAN) based model is proposed using grid search optimization (GSO) technique for identification and classification of thyroid nodule. The main objective of this work is to propose a deep learning (DL) model for the identification and classification of thyroid nodules without user or specialist intervention. The proposed model has gone through four phases namely (i) data acquisition, (ii) pre-processing (iii) data augmentation using GAN technique and (iv) optimization and classification using Deep-GAN model. Two pre-trained architectures namely Alex-Net and Visual Geometry Group (VGG-16) are considered for the identification and classification of thyroid nodule in ultrasonography (USG) images. From the experiment, it is found that Alex-GAN model has shown an improvement of 2 to 4 percentage points in comparison with VGG-GAN model and reported literature on Thyroid digital image database (TDID) public and collected dataset. [ABSTRACT FROM AUTHOR]

Grapevine leaf is a commodity that is collected only once a year and has a high return on investment due to its export. However, only certain types of grapevine leaves are consumed. Therefore, it is extremely important to distinguish the types of grapevine leaves. In particular, performing this process automatically on industrial machines will reduce human errors, workload, and thus cost. In this study, a new hybrid approach based on a convolutional neural network is proposed that can automatically distinguish the types of grapevine leaves. In the proposed approach, firstly, the overfitting of network models is prevented by applying data augmentation techniques. Second, new synthetic images were created with the ESRGAN technique to obtain detailed texture information. Third, the top blocks of the MobileNetV2 and VGG19 CNN models were replaced with the newly designed top block, effectively extracting features with the data. Fourthly, the GASVM algorithm was adapted and used to create a subset of the features to eliminate the ineffective and unimportant ones from the obtained features. Finally, SVM classification was performed with the feature subset consisting of 314 features, and approximately 2% higher accuracy and MCC score were obtained compared to the approaches in the literature. [ABSTRACT FROM AUTHOR]

In recent years, self-supervised learning-based models have been widely used for electrocardiogram (ECG) representation learning. However, most of the models utilize contrastive learning that strongly depend on data augmentation. In this paper, we propose a masked self-supervised learning model based on multiview information bottleneck principle. Our method masks the ECG signal instances in the time and frequency domains at a high ratio and then uses the autoencoder to reconstruct the original input. Not only the intra-view relations within each view but also the inter-view relations between two views are exploited in ECG representation learning. Furthermore, we use the multiview information bottleneck principle to remove redundant information in the time and frequency domains, so that the representations of both views contain more task-relevant information. Our model is pre-trained on three larger ECG datasets at once and fine-tuned on each classification task. Experimental results show that our model not only outperforms state-of-the-art models with self-supervised learning, but also outperforms models with supervised learning. [ABSTRACT FROM AUTHOR]

Cardiac image synthesis and analysis are pivotal in modern healthcare for accurate diagnosis and treatment of cardiovascular conditions. This study introduces a pioneering approach—hybrid GAN with semantic resonance—for the generation and analysis of synthetic cardiac images. Drawing from a diverse dataset encompassing local and global views, acquired through various modalities like MRI, CT, and echocardiography, the critical need for both accuracy and clinical relevance in cardiac image synthesis is addressed. In the architectural setup, a hybrid GAN is defined, comprising both local and global components. The local generator excels in capturing intricate details, while the global generator contextualizes the overall cardiac structure. A fundamental innovation lies in the incorporation of pre-trained CNN classifiers specialized in recognizing cardiac pathologies, anatomical structures, and clinical features. These classifiers provide conditional guidance to the generators, ensuring that the synthesized images align semantically with clinical expectations. The training phase employs a pioneering approach, integrating real-time feedback from the classifiers to steer the image synthesis process. The generators, discriminators, and classifiers are optimized, with a dual emphasis on adversarial loss for authenticity and classification loss for diagnostic significance. The accuracy obtained by the proposed approach is 98.96%. Further, the SSIM and PSNR values attained by the proposed approach are 0.955 and 45.23, which is higher than the existing approaches like GAN, CNN, VAE, DenseNet, VGG19, DeepCardiac, Pix2Pix GAN and Cycle-GAN, respectively. Validation on an independent test dataset underscores the generalization capabilities of the proposed hybrid GAN. Furthermore, post-processing techniques refine the generated images, elevating their clinical relevance. Results visualization presents the prowess of the proposed approach, providing a holistic view of local and global perspectives within synthetic images. [ABSTRACT FROM AUTHOR]

Background: Although electronic nose (eNose) has been intensively investigated for diagnosing lung cancer, cross-site validation remains a major obstacle to be overcome and no studies have yet been performed. Methods: Patients with lung cancer, as well as healthy control and diseased control groups, were prospectively recruited from two referral centers between 2019 and 2022. Deep learning models for detecting lung cancer with eNose breathprint were developed using training cohort from one site and then tested on cohort from the other site. Semi-Supervised Domain-Generalized (Semi-DG) Augmentation (SDA) and Noise-Shift Augmentation (NSA) methods with or without fine-tuning was applied to improve performance. Results: In this study, 231 participants were enrolled, comprising a training/validation cohort of 168 individuals (90 with lung cancer, 16 healthy controls, and 62 diseased controls) and a test cohort of 63 individuals (28 with lung cancer, 10 healthy controls, and 25 diseased controls). The model has satisfactory results in the validation cohort from the same hospital while directly applying the trained model to the test cohort yielded suboptimal results (AUC, 0.61, 95% CI: 0.47─0.76). The performance improved after applying data augmentation methods in the training cohort (SDA, AUC: 0.89 [0.81─0.97]; NSA, AUC:0.90 [0.89─1.00]). Additionally, after applying fine-tuning methods, the performance further improved (SDA plus fine-tuning, AUC:0.95 [0.89─1.00]; NSA plus fine-tuning, AUC:0.95 [0.90─1.00]). Conclusion: Our study revealed that deep learning models developed for eNose breathprint can achieve cross-site validation with data augmentation and fine-tuning. Accordingly, eNose breathprints emerge as a convenient, non-invasive, and potentially generalizable solution for lung cancer detection. Clinical trial registration: This study is not a clinical trial and was therefore not registered. [ABSTRACT FROM AUTHOR]

This paper proposes an approach to enhance the differentiation task between benign and malignant Breast Tumors (BT) using histopathology images from the BreakHis dataset. The main stages involve preprocessing, which encompasses image resizing, data partitioning (training and testing sets), followed by data augmentation techniques. Both feature extraction and classification tasks are employed by a Custom CNN. The experimental results show that the proposed approach using the Custom CNN model exhibits better performance with an accuracy of 84% than applying the same approach using other pretrained models, including MobileNetV3, EfficientNetB0, Vgg16, and ResNet50V2, that present relatively lower accuracies, ranging from 74 to 82%; these four models are used as both feature extractors and classifiers. To increase the accuracy and other performance metrics, Grey Wolf Optimization (GWO), and Modified Gorilla Troops Optimization (MGTO) metaheuristic optimizers are applied to each model separately for hyperparameter tuning. In this case, the experimental results show that the Custom CNN model, refined with MGTO optimization, reaches an exceptional accuracy of 93.13% in just 10 iterations, outperforming the other state-of-the-art methods, and the other four used pretrained models based on the BreakHis dataset. [ABSTRACT FROM AUTHOR]

BACKGROUND: Thyroid tumor is considered to be a very rare form of cancer. But recent researches and surveys highlight the fact that it is becoming prevalent these days because of various factors. OBJECTIVES: This paper proposes a novel hybrid classification system that is able to identify and classify the above said four different types of thyroid tumors using high end artificial intelligence techniques. The input data set is obtained from Digital Database of Thyroid Ultrasound Images through Kaggle repository and augmented for achieving a better classification performance using data warping mechanisms like flipping, rotation, cropping, scaling, and shifting. METHODS: The input data after augmentation goes through preprocessing with the help of bilateral filter and is contrast enhanced using dynamic histogram equalization. The ultrasound images are then segmented using SegNet algorithm of convolutional neural network. The features needed for thyroid tumor classification are obtained from two different algorithms called CapsuleNet and EfficientNetB2 and both the features are fused together. This process of feature fusion is carried out to heighten the accuracy of classification. RESULTS: A Multilayer Perceptron Classifier is used for classification and Bonobo optimizer is employed for optimizing the results produced. The classification performance of the proposed model is weighted using metrics like accuracy, sensitivity, specificity, F1-score, and Matthew's correlation coefficient. CONCLUSION: It can be observed from the results that the proposed multilayer perceptron based thyroid tumor type classification system works in an efficient manner than the existing classifiers like CANFES, Spatial Fuzzy C means, Deep Belief Networks, Thynet and Generative adversarial network and Long Short-Term memory. [ABSTRACT FROM AUTHOR]

Highlights: • Introduce a data augmentation strategy to expand the required different morphological data during the training and learning phase, and improve the algorithm's feature learning ability for complex and diverse tumor morphology CT images. • Design attention mechanisms for encoding and decoding paths to extract fine pixel level features, improve feature extraction capabilities, and achieve efficient spatial channel feature fusion. • The deep supervision layer is used to correct and decode the final image data to provide high accuracy of results. • The effectiveness of this method has been affirmed through validation on the LITS, 3DIRCADb, and SLIVER datasets. BACKGROUND: Accurately extracting liver and liver tumors from medical images is an important step in lesion localization and diagnosis, surgical planning, and postoperative monitoring. However, the limited number of radiation therapists and a great number of images make this work time-consuming. OBJECTIVE: This study designs a spatial attention deep supervised network (SADSNet) for simultaneous automatic segmentation of liver and tumors. METHOD: Firstly, self-designed spatial attention modules are introduced at each layer of the encoder and decoder to extract image features at different scales and resolutions, helping the model better capture liver tumors and fine structures. The designed spatial attention module is implemented through two gate signals related to liver and tumors, as well as changing the size of convolutional kernels; Secondly, deep supervision is added behind the three layers of the decoder to assist the backbone network in feature learning and improve gradient propagation, enhancing robustness. RESULTS: The method was testing on LITS, 3DIRCADb, and SLIVER datasets. For the liver, it obtained dice similarity coefficients of 97.03%, 96.11%, and 97.40%, surface dice of 81.98%, 82.53%, and 86.29%, 95% hausdorff distances of 8.96 mm, 8.26 mm, and 3.79 mm, and average surface distances of 1.54 mm, 1.19 mm, and 0.81 mm. Additionally, it also achieved precise tumor segmentation, which with dice scores of 87.81% and 87.50%, surface dice of 89.63% and 84.26%, 95% hausdorff distance of 12.96 mm and 16.55 mm, and average surface distances of 1.11 mm and 3.04 mm on LITS and 3DIRCADb, respectively. CONCLUSION: The experimental results show that the proposed method is effective and superior to some other methods. Therefore, this method can provide technical support for liver and liver tumor segmentation in clinical practice. [ABSTRACT FROM AUTHOR]

Deep network fault diagnosis methods heavily rely on abundant labeled data for effective model training. However, small-sized samples and imbalanced samples often lead to insufficient features, resulting in accuracy degradation and even instability in the diagnosis model. To address this challenge, this paper introduces a coupled adversarial autoencoder (CoAAE) based on the Bayesian method. This model aims to solve the issue of insufficient samples by generating fake samples and integrating them with the original ones. Within the CoAAE framework, the probability density distribution of the original data is captured using an encoder and fake samples are generated by random sampling from this distribution and decoding them. This process is the adversarial interaction between the encoder and a classifier to obtain the prior distribution of the encoder's parameters. The encoder's parameters are updated through the decoder's reconstruction process, leading to the posterior distribution. Concurrently, the decoder is trained to enhance its ability to reconstruct samples accurately. To address the imbalance in the original samples, a parallel coupled network is employed. This network shares the weights of the extraction layer in the encoder, enabling it to learn the joint distribution between fault-related and normal samples. To evaluate the effectiveness of the proposed data augmentation method, experiments were conducted on a bearing database from Case Western Reserve University using ResNet18 as the deep learning diagnosis model representative. The results demonstrate that CoAAE can effectively augment imbalanced datasets and outperform other advanced methods. [ABSTRACT FROM AUTHOR]