Data Generation Method Research Articles

Appearance-Based Gaze Estimation as a Benchmark for Eye Image Data Generation Methods

Data augmentation is commonly utilized to increase the size and diversity of training sets for deep learning tasks. In this study, we propose a novel application of an existing image generation approach in the domain of realistic eye images that leverages data collected from 40 subjects. This hybrid method combines the benefits of precise control over the image content provided by 3D rendering, while introducing the previously lacking photorealism and diversity into synthetic images through neural style transfer. We prove its general efficacy as a data augmentation tool for appearance-based gaze estimation when generated data are mixed with a sparse train set of real images. It improved the results for 39 out of 40 subjects, with an 11.22% mean and a 19.75% maximum decrease in gaze estimation error, achieving similar metrics for train and held-out subjects. We release our data repository of eye images with gaze labels used in this work for public access.

Exploring charge sharing compensation using inter-pixel coincidence counters for photon counting detectors by deep-learning from local information

Photon counting detectors (PCDs) have well-acknowledged advantages in computed tomography (CT) imaging. However, charge sharing and other problems prevent PCDs from fully realizing the anticipated potential in diagnostic CT. PCDs with multi-energy inter-pixel coincidence counters (MEICC) have been proposed to provide particular information about charge sharing, thereby achieving lower Cramér-Rao Lower Bound (CRLB) than conventional PCDs when assessing its performance by estimating material thickness or virtual monochromatic attenuation integrals (VMAIs). This work explores charge sharing compensation using local spatial coincidence counter information for MEICC detectors through a deep-learning method.
Approach: By analyzing the impact of charge sharing on photon count detection, we designed our network with a focus on individual pixels. Employing MEICC data of patches centered on POIs as input, we utilized local information for effective charge sharing compensation. The output was VMAI at different energies to address real detector issues without knowledge of primary counts. To achieve data diversity, a fast and online data generation method was proposed to provide adequate training data. A new loss function was introduced to reduce bias for training with high-noise data. The proposed method was validated by Monte Carlo (MC) simulation data for MEICC detectors that were compared with conventional PCDs. 
Main-Results: For conventional data as a reference, networks trained on low-noise data yielded results with a minimal bias (about 0.7%) compared with > 3% for the polynomial fitting method. The results of networks trained on high-noise data exhibited a slightly increased bias (about 1.3%) but a significantly reduced standard deviation (STD) and normalized root mean square error (NRMSE). The simulation study of the MEICC detector demonstrated superior compared to the conventional detector across all the metrics. Specifically, for both networks trained on high-noise and low-noise data, their biases were reduced to about 1% and 0.6%, respectively. Meanwhile, the results from a MEICC detector were of about 10% lower noise than a conventional detector. Moreover, an ablation study showed that the additional loss function on bias was beneficial for training on high-noise data.
Significance: We demonstrated that a network-based method could utilize local information in PCDs effectively by patch-based learning to reduce the impact of charge sharing. MEICC detectors provide very valuable local spatial information by additional coincidence counters. Compared with MEICC detectors, conventional PCDs only have limited local spatial information for charge sharing compensation, resulting in higher bias and standard deviation in VMAI estimation with the same patch strategy.&#xD.

To be or not to be, when synthetic data meet clinical pharmacology: A focused study on pharmacogenetics.

The use of synthetic data in pharmacology research has gained significant attention due to its potential to address privacy concerns and promote open science. In this study, we implemented and compared three synthetic data generation methods, CT-GAN, TVAE, and a simplified implementation of Avatar, for a previously published pharmacogenetic dataset of 253 patients with one measurement per patient (non-longitudinal). The aim of this study was to evaluate the performance of these methods in terms of data utility and privacy trade off. Our results showed that CT-GAN and Avatar used with k = 10 (number of patients used to create the local model of generation) had the best overall performance in terms of data utility and privacy preservation. However, the TVAE method showed a relatively lower level of performance in these aspects. In terms of Hazard ratio estimation, Avatar with k = 10 produced HR estimates closest to the original data, whereas CT-GAN slightly underestimated the HR and TVAE showed the most significant deviation from the original HR. We also investigated the effect of applying the algorithms multiple times to improve results stability in terms of HR estimation. Our findings suggested that this approach could be beneficial, especially in the case of small datasets, to achieve more reliable and robust results. In conclusion, our study provides valuable insights into the performance of CT-GAN, TVAE, and Avatar methods for synthetic data generation in pharmacogenetic research. The application to other type of data and analyses (data driven) used in pharmacology should be further investigated.

Method for Enhancing AI Accuracy in Pressure Injury Detection Using Real and Synthetic Datasets

Pressure injuries pose significant health risks, especially for the elderly, immobile individuals, and those with sensory impairments. These injuries can rapidly become chronic, making initial diagnosis important. Due to the difficulty of transporting patients from local health facilities to higher-level general hospitals for treatment, it is essential to utilize telemedicine tools, such as chatbots, to ensure rapid initial diagnosis. Recent advances in artificial intelligence have demonstrated potential for medical imaging and disease classification. Ongoing research in the field of dermatological diseases focuses on disease classification. However, the assessment accuracy of artificial intelligence is often limited by unequal class distributions and insufficient dataset quantities. In this study, we aim to enhance the accuracy of artificial intelligence models by generating synthetic datasets. Specifically, we focused on training models for Pressure Injury assessment using both real and synthetic datasets. We used PI data at a domestic medical university. As part of our supplementary research, we established a chatbot system to facilitate the assessment of pressure injuries. Using both constructed and synthetic data, we achieved a top-1 accuracy of 92.03%. The experimental results demonstrate that combining real and synthetic data significantly improves model accuracy. These findings suggest that synthetic datasets can be effectively utilized to address the limitations of small-scale datasets in medical applications. Future research should explore the use of diverse synthetic data generation methods and validate model performance on a variety of datasets to enhance the generalization and robustness of AI models for Pressure Injury assessment.

Analytical ab initio hessian from a deep learning potential for transition state optimization

Identifying transition states—saddle points on the potential energy surface connecting reactant and product minima—is central to predicting kinetic barriers and understanding chemical reaction mechanisms. In this work, we train a fully differentiable equivariant neural network potential, NewtonNet, on thousands of organic reactions and derive the analytical Hessians. By reducing the computational cost by several orders of magnitude relative to the density functional theory (DFT) ab initio source, we can afford to use the learned Hessians at every step for the saddle point optimizations. We show that the full machine learned (ML) Hessian robustly finds the transition states of 240 unseen organic reactions, even when the quality of the initial guess structures are degraded, while reducing the number of optimization steps to convergence by 2–3× compared to the quasi-Newton DFT and ML methods. All data generation, NewtonNet model, and ML transition state finding methods are available in an automated workflow.

A Martingale Posterior-Based Fault Detection and Estimation Method for Electrical Systems of Industry

The improvement of information sciences promotes the utilization of data for process monitoring. As the core of modern automation, time-stamped signals are used to estimate the system state and construct the data-driven model. Many recent studies claimed that the effectiveness of data-driven methods relies greatly on data quality. Considering the complexity of the operating environment, process data will inevitably be affected. This poses big challenges to estimating faults from data and delivers feasible strategies for electrical systems of industry. This paper addresses the missing data problem commonly in traction systems by designing a martingale posterior-based data generation method for the state-space model. Then, a data-driven approach is proposed for fault detection and estimation via the subspace identification technique. It is a general scheme using the Bayesian framework, in which the Dirichlet process plays a crucial role. The data-driven method is applied to a pilot-scale traction motor platform. Experimental results show that the method has good estimation performance.

From shallows to depths: unveiling hybrid synthetic data modeling for enhanced learning with privacy considerations in naturally imbalanced datasets

Data serves as the foundational element that drives model development and performance in machine learning and deep learning. The learning algorithms are as good as data on which they are trained on. Data constrained environments pose serious threat to the effectiveness of learning algorithms. Data limitations stem from a range of factors, including regulatory restrictions, privacy concerns, and the inherent scarcity of relevant data. This constrained availability of data often leads to class imbalance problem in the context of classification tasks. To address the challenges of limited data, the algorithmic generated data, also known as synthetic data, is gaining significant traction as a cost-effective, readily available, and secure alternative. Synthetic data generation techniques can be employed to enhance dataset size by augmenting data samples and to address class imbalance by increasing the number of minority class instances. These techniques generally fall into two main categories: distance-based shallow models and probability estimation-based deep generative models. Shallow interpolation-based models generate new data points within the local space between existing data points, while deep density estimation based models generate new data by learning the whole distribution of data. . In the context of smaller datasets, deep generative models often struggle to accurately estimate the probability distribution of whole data. To effectively represent the global data distribution, these deep models require initial starting data samples to guide their approximation. This paper examines the potential of integration of shallow and deep generative models in the data generation pipeline for effective synthetic data augmentation which furthers enhanced learning and generalization of downstream tasks. In this work, we present a hybrid approach of tabular data generation involving mixed type data attributes (continuous, discrete) and pay special attention to data imbalance and insufficient data problems. We introduce the Hybrid Data Balancing and Augmentation Approach for Mixed Tabular Data (HDBA-MTD), specifically designed to synthesize samples for underrepresented labels of output class and address issues of insufficient data instances. This approach enhances training data diversity, thereby paying special attention to the downstream classification and generalization performance. Experiments are carried out using benchmark datasets to assess the practicality of the presented hybrid model in real-world scenarios. This work has also attempted to quantify the privacy preservability for real data concerning ethical considerations and data security circumstances. The evaluation and analysis of these experiments show that the present hybrid model performs favorably compared to other current hybrid synthetic data generation methods.

The regional development and implementation of home-based stroke rehabilitation using participatory action research

Purpose This study aims to overcome the challenges experienced in the regional development and implementation of home-based stroke rehabilitation (HBSR) and to understand the change process needed. Materials and methods Using participatory action research (PAR), participants and researchers collaboratively produced knowledge and took action to improve the offered HBSR. Different methods for data generation and analysis were used, depending on the aim of the PAR phase and the participants’ stages of change. The Consolidated Framework for Implementation Research (CFIR) was used to select implementation strategies and to evaluate the implementation process. Results Developing and implementing HBSR resulted in multiple products that promoted the implementation of a regional stroke network and affiliated work arrangements. Work arrangements were embodied in a stroke care pathway, follow-up tool, and expertise requirements. Evaluating the PAR process identified participants being able to take the lead, being facilitated by others, and making progress visible, as implementation facilitators. Collaborating within a primary care project can be challenging but is considered essential and has a positive impact on multiple levels. Also, the implementation of HBSR calls for multiple implementation strategies reflecting multiple CFIR constructs. Conclusion This study highlights the complexity and achievements of developing and implementing HBSR using PAR.

Hydrological Data Projection Using Empirical Mode Decomposition: Applications in a Changing Climate

This paper demonstrates the effectiveness and superiority of Empirical Mode Decomposition (EMD) in projecting non-stationary hydrological data. The study focuses on daily Sea Surface Temperature (SST) sequences in the Niño 3.4 region and uses EMD to forecast the probability of El Niño events. Applying the Mann–Kendall test at the 5% significance level reveals a significant increasing trend in SST changes in this region, particularly noticeable after 1980. This trend is associated with the occurrence of El Niño and La Niña events, which have a recurrence interval of approximately 8.4 years and persist for over a year. The modified Oceanic Niño Index (ONI) proposed in this study demonstrates high forecast accuracy, with 97.56% accuracy for El Niño and 89.80% for La Niña events. Additionally, the EMD of SST data results in 13 Intrinsic Mode Functions (IMFs) and a residual component. The oscillation period increases with each IMF level, with IMF7 exhibiting the largest amplitude, fluctuating between ±1 °C. The residual component shows a significant upward trend, with an average annual increase of 0.0107 °C. These findings reveal that the EMD-based data generation method overcomes the limitations of traditional hydrological models in managing non-stationary sequences, representing a notable advancement in data-driven hydrological time series modeling. Practically, the EMD-based 5-year moving process can generate daily sea temperature sequences for the coming year in this region, offering valuable insights for assessing El Niño probabilities and facilitating annual updates.

RT-SNDETR: real-time supernova detection via end-to-end image transformers

ABSTRACT In large-scale astronomical surveys, traditional supernova detection pipelines rely on complex and relatively inefficient image differencing techniques. This paper proposes an end-to-end deep-learning supernova detection network, the Real-Time SuperNova DEtection TRansformer (RT-SNDETR). This network partially replaces traditional pipelines by integrating image differencing, source detection, and Real-bogus classification, achieving a speed 51.49 times that of the fastest image differencing method, SFFT. Additionally, it remains competitive with methods like YOLO v8, offering a well-balanced trade-off between speed and accuracy. Experimental results highlight RT-SNDETR’s superior performance, with an average precision(AP) of 96.30 per cent on synthetic samples and 76.60 per cent on real supernova samples. It significantly outperforms other detection networks, including RT-DETR (+5.6 per cent AP on synthetic/+5.1 per cent AP on real samples) and Cascade R-CNN (+8.9 per cent AP on synthetic/ +28.6 per cent AP on real samples). The incorporation of CycleGAN-based data generation methods plays a significant role in enhancing RT-SNDETR’s performance. These methods simulate realistic PSF variations, enabling the object detection network to learn more robust features and improving its generalization to real supernovae data. Additionally, by integrating unsupervised domain adaptation techniques, RT-SNDETR achieves an AP of 81.70 per cent on real SDSS supernova survey samples. This study demonstrates RT-SNDETR’s potential to significantly enhance both the speed and accuracy of supernova detection, making it a highly effective solution for large-scale astronomical surveys.

Does Differentially Private Synthetic Data Lead to Synthetic Discoveries?

Synthetic data have been proposed as a solution for sharing anonymized versions of sensitive biomedical datasets. Ideally, synthetic data should preserve the structure and statistical properties of the original data, while protecting the privacy of the individual subjects. Differential Privacy (DP) is currently considered the gold standard approach for balancing this trade-off. The aim of this study is to investigate how trustworthy are group differences discovered by independent sample tests from DP-synthetic data. The evaluation is carried out in terms of the tests' Type I and Type II errors. With the former, we can quantify the tests' validity, i.e., whether the probability of false discoveries is indeed below the significance level, and the latter indicates the tests' power in making real discoveries. We evaluate the Mann-Whitney U test, Student's t-test, chi-squared test, and median test on DP-synthetic data. The private synthetic datasets are generated from real-world data, including a prostate cancer dataset (n = 500) and a cardiovascular dataset (n = 70,000), as well as on bivariate and multivariate simulated data. Five different DP-synthetic data generation methods are evaluated, including two basic DP histogram release methods and MWEM, Private-PGM, and DP GAN algorithms. A large portion of the evaluation results expressed dramatically inflated Type I errors, especially at levels of ϵ ≤ 1. This result calls for caution when releasing and analyzing DP-synthetic data: low p-values may be obtained in statistical tests simply as a byproduct of the noise added to protect privacy. A DP Smoothed Histogram-based synthetic data generation method was shown to produce valid Type I error for all privacy levels tested but required a large original dataset size and a modest privacy budget (ϵ ≥ 5) in order to have reasonable Type II error levels.

Bone metastasis scintigram generation using generative adversarial learning with multi-receptive field learning and two-stage training.

Deep learning is the primary method for conducting automated analysis of SPECT bone scintigrams. The lack of available large-scale data significantly hinders the development of well-performing deep learning models, as the performance of a deep learning model is positively correlated with the size of the dataset used. Therefore, there is an urgent demand for an automated data generation method to enlarge the dataset of SPECT bone scintigrams. We introduce a deep learning-based generation model that can generate realistic but not identical samples from the original SPECT bone scintigrams. Following the generative adversarial learning architecture, a bone metastasis scintigram generation model christened BMS-Gen is proposed. First, BMS-Gen takes multiple input conditions and employs multi-receptive field learning to ensure that the generated samples are as realistic as possible. Second, BMS-Gen adopts generative adversarial learning to retain the diversity of the generated samples. Last, BMS-Gen uses a two-stage training strategy to improve the quality of the generated samples. Experimental evaluation conducted on a set of clinical data of SPECT BM scintigrams has shown the performance of the proposed BMS-Gen, achieving the best overall scores of 1678.0, 69.33, and 19.51 for FID (Fréchet Inception Distance), MSE (Mean Square Error), and PSNR (Peak Signal-to-Noise Ratio) metrics. The introduction of samples generated by BMS-Gen contributes a maximum (minimum) increase of 3.01% (0.15%) on the F-1 score and a maximum (minimum) increase of 6.83% (2.21%) on the DSC score for the image classification and segmentation tasks, respectively. The proposed BMS-Gen model can be used as a promising tool for augmenting the data of bone scintigrams, greatly facilitating the development of deep learning-based automated analysis of SPECT bone scintigrams.

Curriculum vision as a tool to the rescue of Prof M: a life history study

ABSTRACT Life-curriculum vision is a reasoning plan for or of educating that guides human actions. A vision is a reasoning plan for educating, should a plan be prescribed as strictly guiding human actions; or a plan of educating, should a plan be generated after actions have already taken place. As such, this life-history study explored and interrogated lessons learnt by means of and through Prof M’s life-curriculum vision that helped him to understand his unique identities and to manage his life-objective reality. Prof M’s vision and pragmatic paradigm produced a natural identity framework for this purposive and convenient individual life-history study based on photo voices, semi-structured interviews, and reflective activities as the data-generation methods. Prof M’s narratives were evaluated by his family, relatives, colleagues, and friends. The findings revealed that, through his life-curriculum vision, he had become a pragmatist, naturalist, and agnostic, deploying his experiences of societal, professional, personal, and natural identities. Consequently, this study recommends that individuals reflect on and critique their experiences. Such reflection and critique may allow them to create their own relevant unique tools, enabling them to address both their personal needs (personal identity) and their societal and professional needs.

TabSAL: Synthesizing Tabular data with Small agent Assisted Language models

Tabular data are widely used in machine-learning tasks because of their prevalence in various fields; however, the potential risks of data breaches in tabular data and privacy protection regulations render such data almost unavailable. Tabular data generation methods alleviate data unavailability by synthesizing privacy-free data, and generating data using language models is a novel innovation. Language models can synthesize high-quality datasets by learning knowledge from nondestructive information and recognizing the semantics of table columns. However, when current language models function as generators, their encoding methods are hindered by complicated decoding processes, and the limited predictive ability of language models restricts their generative capability. To this end, we propose an encoding method based on interactive data structures such as JavaScript Object Notation for converting tabular data. We design TabSAL, which is a pluggable tabular data generation framework with small agent assisted language models, to boost the predictive capability, resulting in high-quality synthetic datasets with a much lower computational resource cost. In addition, a benchmark that integrates eight datasets, three methods, and three assessment directions has been issued, which indicates that TabSAL surpasses the state of the art by up to 60%.

Local fusion generative adversarial network with dual-discriminator and parallel multipath and its application in machinery fault diagnosis with imbalanced data

Abstract Diagnosing faults in critical machinery components is imperative for effective condition monitoring and real-world datasets often suffer from data imbalance. To address this issue, numerous data generation methods have been developed, such as improved local fusion generative adversarial network (ILoFGAN), variational autoencoding GAN (VAEGAN), etc. However, the existing data generation methods primarily concentrate on global and single-scale features and often ignore local or multi-scale features, which leads to the omission of key features or nuances in the generated data. Therefore, a novel approach called the Local Fusion Generative Adversarial Network with Dual-discriminator and Parallel multipath (LoFGAN-DP) is designed to enhance the fault diagnosis performance in the context of imbalanced data. The LoFGAN-DP features a Parallel Multi-Path (PMP) module along with a dual-discriminator scheme, in which the multipath module facilitates feature extraction at various scales through convolution across paths of diverse sizes, and the dual-discriminator scheme can better improve the quality and diversity of the samples generated by the generator. The PMP module and dual-discriminator scheme enhance the proposed method's robustness against variations in input data. After generating data by LoFGAN-DP, a two-dimensional capsule network is further used to achieve the efficient recognition of fault features. To validate the proposed LoFGAN-DP in the machinery fault diagnosis with imbalanced data, the gear dataset and the self-constructed bearing dataset were utilized. Experimental results show that LoFGAN-DP significantly improves Structural Similarity Index (SSIM), Fréchet Inception Distance (FID), and fault classification accuracy compared to several advanced methods.

Evaluating Synthetic Data Augmentation to Correct for Data Imbalance in Realistic Clinical Prediction Settings.

Predictive modeling holds a large potential in clinical decision-making, yet its effectiveness can be hindered by inherent data imbalances in clinical datasets. This study investigates the utility of synthetic data for improving the performance of predictive modeling on realistic small imbalanced clinical datasets. We compared various synthetic data generation methods including Generative Adversarial Networks, Normalizing Flows, and Variational Autoencoders to the standard baselines for correcting for class underrepresentation on four clinical datasets. Although results show improvement in F1 scores in some cases, even over multiple repetitions, we do not obtain statistically significant evidence that synthetic data generation outperforms standard baselines for correcting for class imbalance. This study challenges common beliefs about the efficacy of synthetic data for data augmentation and highlights the importance of evaluating new complex methods against simple baselines.

Synthetic Generation of Patient Service Utilization Data: A Scalability Study.

To address privacy and ethical issues in using health data for machine learning, we evaluate the scalability of advanced synthetic data generation methods like GANs, VAEs, copulaGAN, and transformer models specifically for patient service utilization data. Our study examines five models on data from a Canadian health authority, focusing on training and generation efficiency, data resemblance, and practical utility. Our findings indicate that statistical models excel in efficiency, while most models produce synthetic data that closely mirrors real data, and is also useful for real-world applications.

An interlaboratory proficiency test using metagenomic sequencing as a diagnostic tool for the detection of RNA viruses in swine fecal material.

Metagenomic shotgun sequencing (mNGS) is a generic molecular diagnostic method, involving laboratory preparation of samples, sequencing, bioinformatic analysis of millions of short sequences, and interpretation of the results. In this paper, we investigated the performance of mNGS on the detection of porcine astroviruses, a model for RNA viruses in a pig fecal material, among six European veterinary and public health laboratories. We showed that different methods for data generation affect mNGS performance among participants and that the selection of reference genomes is crucial for read classification. Follow-up investigation remains a critical factor for the diagnostic interpretation of mNGS results. The paper contributes to potential improvements of mNGS as a diagnostic tool in clinical settings.

Rapid prediction of wall shear stress in stenosed coronary arteries based on deep learning.

There is increasing evidence that coronary artery wall shear stress (WSS) measurement provides useful prognostic information that allows prediction of adverse cardiovascular events. Computational Fluid Dynamics (CFD) has been extensively used in research to measure vessel physiology and examine the role of the local haemodynamic forces on the evolution of atherosclerosis. Nonetheless, CFD modelling remains computationally expensive and time-consuming, making its direct use in clinical practice inconvenient. A number of studies have investigated the use of deep learning (DL) approaches for fast WSS prediction. However, in these reports, patient data were limited and most of them used synthetic data generation methods for developing the training set. In this paper, we implement 2 approaches for synthetic data generation and combine their output with real patient data in order to train a DL model with a U-net architecture for prediction of WSS in the coronary arteries. The model achieved 6.03% Normalised Mean Absolute Error (NMAE) with inference taking only 0.35s; making this solution time-efficient and clinically relevant.

Overcoming Data Deficiency for Multi-Person Pose Estimation.

Building multi-person pose estimation (MPPE) models that can handle complex foreground and uncommon scenes is an important challenge in computer vision. Aside from designing novel models, strengthening training data is a promising direction but remains largely unexploited for the MPPE task. In this article, we systematically identify the key deficiencies of existing pose datasets that prevent the power of well-designed models from being fully exploited and propose the corresponding solutions. Specifically, we find that the traditional data augmentation techniques are inadequate in addressing the two key deficiencies, imbalanced instance complexity (IC) (evaluated by our new metric IC) and insufficient realistic scenes. To overcome these deficiencies, we propose a model-agnostic full-view data generation (Full-DG) method to enrich the training data from the perspectives of both poses and scenes. By hallucinating images with more balanced pose complexity and richer real-world scenes, Full-DG can help improve pose estimators' robustness and generalizability. In addition, we introduce a plug-and-play adaptive category-aware loss (AC-loss) to alleviate the severe pixel-level imbalance between keypoints and backgrounds (i.e., around 1:600). Full-DG together with AC-loss can be readily applied to both the bottom-up and top-down models to improve their accuracy. Notably, plugging into the representative estimators HigherHRNet and HRNet, our method achieves substantial performance gains of 1.0%-2.9% AP on the COCO benchmark, and 1.0%-5.1% AP on the CrowdPose benchmark.