Data Synthesis Methods Research Articles

Evaluating the Effect of Surrogate Data Generation on Healthcare Data Assessment

In healthcare applications, often it is not possible to record sufficient data as required for deep learning or data-driven classification and feature detection systems due to the patient condition, various clinical or experimental limitations, or time constraints. On the other hand, data imbalance invalidates many of the test results crucial for clinical approvals. Generating synthetic (artificial or dummy) data has become a potential solution to address this issue. Such data should possess adequate information, properties, and characteristics to mimic the real-world data recorded in natural circumstances. Several methods have been proposed for this purpose, and results often show that adding surrogates improves the decision-making accuracy. This article evaluates the most recent surrogate data generation and data synthesis methods to investigate the effects of the number of surrogates on improving the classification results. It is shown that the data analysis/classification results improve with an increasing number of surrogates, but this no longer continues after a certain number of surrogates. This achievement helps in deciding on the number of surrogates for each strategy, resulting in the alleviation of the computation cost.

Fixed-effect versus random-effect model in meta-analysis: How to decide?

Interpretation of meta-analysis requires an appropriate understanding of its statistical methods. Data synthesis methods are the basis of the meta-analytic process, and a thorough knowledge of the models used is essential. The two most widely used methods for data synthesis are the ‘fixed-effect method’ and the ‘random-effect method’. The latter is commonly used when a ‘significant heterogeneity’ exists. This narrative review explains fixed- and random-effect models, the two most commonly used data synthesis models.

Critical appraisal of systematic reviews and meta-analyses

Systematic reviews and meta-analyses (SRMAs) are cornerstone methodologies in evidence-based medicine, synthesising vast research to provide comprehensive insights. Despite their value, the increasing prevalence of SRMAs has highlighted the need for robust critical appraisal to ensure their reliability and applicability. This document outlines a structured framework for evaluating SRMAs, emphasising key elements such as clearly defined research questions, appropriate inclusion criteria, comprehensive search strategies and the assessment of bias in included studies. In addition, it discusses the importance of rigorous data extraction, synthesis methods and evaluation of publication bias. Tools like Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), Measurement Tools to Assess Systematic Reviews, version 2 and Grading of Recommendations Assessment, Development and Evaluation (GRADE) can facilitate appraisal and help translate findings into clinical practice. By identifying knowledge gaps and methodological strengths or weaknesses, critical appraisal enhances the quality of SRMAs and safeguards their role in guiding evidence-based decision-making.

Improved deep learning-based IVIM parameter estimation via the use of more "realistic" simulated brain data.

Due to the low signal-to-noise ratio (SNR) and the limited number of b-values, precise parameter estimation of intravoxel incoherent motion (IVIM) imaging remains an open issue to date, especially for brain imaging where the relatively small difference between D and D* easily leads to outliers and obvious graininess in estimated results. To propose a synthetic data driven supervised learning method (SDD-IVIM) for improving precision and noise robustness in IVIM parameter estimation without relying on real-world data for neural network training. On account of the absence of standard IVIM parametric maps from real-world data, a novel model-based method for generating synthetic human brain IVIM data was introduced. Initially, the parameter values of synthetic IVIM parametric maps were sampled from the complex distributions composed of a series of simple and uniform distributions. Subsequently, these parametric maps were modulated with human brain texture to imitate brain tissue structure. Finally, they were used to generate synthetic human brain multi-b-value diffusion-weighted (DW) images based on the IVIM bi-exponential model. With the proposed data synthesis method, an ordinary U-Net with spatial smoothness was employed for IVIM parameter mapping within a supervised learning framework. The performance of SDD-IVIM was evaluated on both numerical phantom and 20 glioma patients. The estimated IVIM parametric maps were compared to those derived from five state-of-the-art methods. In numerical phantom experiments, SDD-IVIM method produces IVIM parametric maps with lower mean absolute error, lower mean bias, and higher structural similarity compared to the other five methods, especially when the SNR of DW images is low. In glioma patient experiments, SDD-IVIM method offers lower coefficient of variation and more reasonable contrast-to-noise ratio between tumor and contralateral normal appearing white matter than the other five methods. Our method owns superior performance in parametric map quality, parameter estimation precision, and lesion characterization in IVIM parameter estimation, with strong resistance to noise.

POST-WAR REINTEGRATION OF THE VETERANS: THE EXPERIENCE OF AMERICAN UNIVERSITIES

The article analyzes the experience of American universities in reintegrating veterans. It is predicted that by the end of the war in Ukraine, almost 15% of the population will be veterans and their families, so the reintegration of veterans is an extremely important, large-scale and urgent issue for Ukraine. The article uses general scientific methods of data analysis and synthesis, comparison and systematization. The information base of the article is based on American scientific articles and programs for the reintegration of veterans. The article outlines the main areas of veterans' reintegration based on the US experience: state support, partnerships between veterans and non-governmental organizations, veteran business incubators, integration of veterans in the labor market, and support for veterans' mental health. A separate important place in this list is occupied by American universities, which create relevant programs, centers, and hubs to support the reintegration of veterans, as well as conduct numerous studies on veterans' reintegration. The article provides several examples of American universities that actively cooperate with veterans and help create social entrepreneurship. By creating a favorable environment for the development of social entrepreneurship, American universities contribute to solving important social problems, which is relevant for Ukraine in terms of the post-war period. As a result of the study of the American experience, the author proposed promising directions for the activities of Ukrainian universities regarding the post-war reintegration of veterans: development of stratagems by veterans at university hubs, development of educational programs for veterans, development of retraining programs for veterans, in particular in cooperation with business; psychological support and adaptation at universities through counseling centers, support groups and volunteer programs; conducting joint research with veterans, launching innovative and social projects; support for an inclusive environment. Universities can thus play a key role in returning veterans to a peaceful life through education, vocational training, support for psychological adjustment and the creation of social entrepreneurship.

Fungal infection in patients treated with Bruton tyrosine kinase inhibitor-from epidemiology to clinical outcome: a systematic review.

The Bruton tyrosine kinase inhibitor (BTKi) has emerged as a key treatment for B-cell lymphomas. Despite its efficacy in the treatment of malignancies, numerous cases of invasive fungal infections (IFI) have been reported in patients receiving ibrutinib, a first-generation BTKi. Cases of invasive aspergillosis have also been reported with acalabrutinib and zanubrutinib. The objective of this study was to provide an overview of the pathogens involved, the time of onset of infections and factors influencing survival. Data sources: PubMed, Embase and Web of Science databases were used, and the results were reported according to Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Case reports, case series, clinical trials and cohort studies were included. All reported cases of IFI in patients treated with BTKi were analysed. For case reports/case series, demographic, microbiological and outcome data were retrieved. Assessment of risk of bias: Given the significant heterogeneity in clinical trials/cohort studies, only epidemiological analysis was performed, without formal incidence analysis. Methods of data synthesis: Epidemiologic data were presented as descriptive statistics. In total, 25215 patients from 92 retrospective and prospective clinical trials/cohort studies and 211 patients from 115 case reports/case series were included. Among clinical trials/cohorts, 736 IFI were reported, including 234 candidiasis (31.8%), 227 aspergillosis (30.8%) and 124 Pneumocystis jirovecii pneumonia (PJP) (16.8%). Among the case reports/case series, 155 (73.5%) had chronic lymphocytic leukaemia, and 56 (26.5%) had other malignancies. The main IFI were aspergillosis (n=107, 50.7%), cryptococcosis (n=33, 15.6%), PJP (n=26, 12.3%) and mucormycosis (n=23, 10.9%). The median delay between the initiation of BTKi and IFI was 2.3, 4.0, 3.0 and 3.0 for aspergillosis, cryptococcosis, PJP and mucormycosis, respectively. The survival rate improved when BTKi was discontinued during infection. Targeted therapies in lymphocytic malignancies raised new issues concerning infectious complications. Monitoring IFI in patients receiving second- and third-generation BTKi is crucial for improving the management of these manifestations.

Synthesizing Feature-Aligned and Category-aware Electronic Medical Records for Intracranial Aneurysm Rupture Prediction.

Rupture prediction is crucial for precise treatment and follow-up management of patients with intracranial aneurysms (IAs). Considerable machine learning (ML) methods have been proposed to improve rupture prediction by leveraging electronic medical records (EMRs), however, data scarcity and category imbalance strongly influence performance. Thus, we propose a novel data synthesis method i.e., Transformer-based conditional GAN (TransCGAN), to synthesize highly authentic and category-aware EMRs to address above challenges. Specifically, we first align feature-wise context relationship and distribution between synthetic and original data to enhance synthetic data quality. To achieve this, we first integrate the Transformer structure into GAN to match the contextual relationship by processing the long-range dependencies among clinical factors and introduce a statistical loss to maintain distributional consistency by constraining the mean and variance of the synthesis features. Additionally, a conditional module is designed to assign the category of the synthesis data, thereby addressing the challenge of category imbalance. Subsequently, the synthetic data are merged with the original data to form a large-scale and category-balanced training dataset for IAs rupture prediction. Experimental results show that using TransCGAN's synthetic data enhances classifier performance, achieving AUC of 0.89 and outperforming state-of-the-art resampling methods by 5 %-33 % in F1 score.

Clinical Readiness for Practice of Nursing Students: A Concept Analysis.

The concept of clinical readiness for practice among nursing students is yet to be analyzed, and there is a lack of empirical evidence on its usage among academics and clinicians. This concept analysis is anchored on a systematic literature review that adhered to the PRISMA guidelines and incorporated the eight iterative steps of Walker and Avant's concept analysis method. This concept analysis method involved: (1) choosing a concept; (2) determining the objectives of the analysis; (3) identifying usages of the concept; (4) determining the defining attributes; (5) identifying a model case; (6) identifying other cases, including borderline, contrary, and related cases; (7) identifying antecedents and consequences; and (8) defining empirical references. The integrative thematic data synthesis method was adopted. The concept of nursing students' clinical readiness for practice is said to have four interrelated attributes. These attributes included (1) professional skills, (2) communication skills, (3) self-management skills, and (4) self-confidence. The two antecedents for nursing students' clinical readiness to practice are (1) personal factors, including demographic characteristics, prior healthcare experience, income, and emotional intelligence; and (2) educational factors, including the clinical learning environment, clinical internship program, learning resource, and learning strategy. The consequence of clinical readiness for the practice of nursing students includes obtaining practice skills that can lead to more personal and job-related satisfactory outcomes. clinical readiness for practice in nursing encompasses the acquisition and integration of professional knowledge, skills, effective communication abilities, and self-management capabilities and the application of these competencies with confidence toward the provision of high-quality care to patients. Understanding the components of clinical readiness is crucial for nursing educators, preceptors, and healthcare institutions to ensure that nursing students are adequately prepared for the challenges they will face in clinical practice. By recognizing the importance of professional knowledge, skills, communication, and self-management in clinical readiness, educators and training institutions can tailor their curricula, programs, and support systems to better prepare nursing students for the demands of real-world healthcare settings. This focus on clinical readiness ultimately delivers safe, effective, and compassionate patient care.

Endodontic treatment modifies circulatory inflammatory mediator levels: A systematic review with meta-analysis.

There is limited and conflicting data on the reduction of circulatory inflammatory mediators in patients with apical periodontitis (AP) following endodontic treatment. To answer the following research question: in adult healthy patients with AP [Population (P)], is there a difference before [Comparator (C)] and after various endodontic treatments (nonsurgical, surgical or retreatment) [Intervention (I)] on systemic levels of inflammatory biomarkers [Outcome (O)] in the follow-up period[Time (T)]? An electronic literature search was conducted in the databases Scopus, PubMed, Clarivate Analytics' Web of Science, Cochrane Database of Systematic Reviews and Grey literature from inception to July 2024 with no language restrictions. Observational studies examining changes in serum levels of inflammatory mediators were included. Two independent reviewers selected studies, extracted data and critically appraised the included studies. Qualitative and quantitative (meta-analysis) data synthesis methods were employed. The Newcastle-Ottawa Scale was used to assess the quality of the included studies. Sixteen studies met the inclusion criteria, of which six were included in the meta-analysis. These studies were published between 1992 and 2024, involving a total of 596 patients (54% females) aged between 16 and 75 years. The meta-analysis of pooled data showed a significant decrease in high-sensitive C-reactive protein (hs-CRP) levels in the serum of patients with AP 6 months after treatment [2.26 ± 1.76 versus 1.28 ± 1.06 mg/L, (Z = 2.03, p = .04)] and a decrease in interleukin-1β (IL-1β) levels 12 months after treatment [13.01 ± 5.95 versus 10.86 ± 3.52 pg/mL, (Z = 3.72, p < .01)]. One study was assessed as poor quality, while all others were considered high quality. Despite the differences in methodologies across the included studies, it has been established that effective endodontic treatment leads to a reduction in systemic inflammatory biomarkers in the body. Following effective endodontic treatment in patients with AP, the systemic levels of hs-CRP and IL-1β exhibit a significant reduction at 6 and 12 months, respectively. Further clinical studies should investigate whether effective endodontic treatment and reduced levels of investigated biomarkers may change the clinical presentation of systemic diseases. PROSPERO database (CRD42024559271).

POSTWAR SOCIALISATION OF THE UKRAINIAN ECONOMY IN THE CONDITIONS OF EUROPEAN INTEGRATION

The full-scale war in Ukraine has exacerbated a number of social problems that will need to be addressed during the post-war reconstruction period through the socialization of the economy. Ukraine's European integration will have a positive impact on this process. The relevance of post-war socialization of Ukraine's economy in the context of European integration is driven by many factors related to the need to restore the country after the conflict and adapt to the standards of the European Union. European integration is a key goal of Ukraine, and it requires serious transformations in the country's economic, social and political life. The purpose of the article is to study the development of post-war socialization of Ukraine's economy in the context of European integration. To achieve the purpose of the article, the following tasks were set and solved: to identify the main aspects of the post-war socialization of Ukraine's economy, to analyze the impact of European integration on the post-war development of the social sphere. The article uses general scientific methods of data analysis and synthesis, comparison and systematization. The information base of the article is made up of scientific articles and online statistical databases. The article analyzes statistical data on the damage to the destroyed social infrastructure, the dynamics of the unemployment rate in 2020-2024, and the number of registered Ukrainian refugees by country. The author outlines the directions of post-war socialization of Ukraine's economy using the prospects of European integration processes. The article offers recommendations on how to use the potential of Ukraine's integration into the EU to build the social sphere in the postwar period. The postwar socialization of Ukraine's economy is strategically important for achieving social stability, strengthening the country's competitive position and further integration into the European Union. This approach will ensure Ukraine's sustainable development and improve the quality of life of the population within the European community.The practical value of the article lies in the fact that its material can be used by state authorities in the post-war reconstruction of Ukraine with the use of the potential of European integration.

Montesquieu’s Thought on Rule of Law-Set Issues for Political Practice in Vietnam

Building the State with the objective: prosperous people, strong nation, democracy, justice and civilization has set an indispensable requirement of the comprehensive renovation of the apparatus of Vietnam in the direction of a modern state, which is the actual reason why we choose the political thought of Montesquieu as a theoretical basis for this article. The purpose of the article is to assess the historical significance, epochal value, positive contributions of the doctrine to theory and practice of mankind through a generalization of Montesquieu's political thought, thereby we draw lessons and the necessity of creatively applying the doctrine to the construction of the rule-of-law state in Vietnam. To achieve this purpose, we structure the content of the article into two sections: In the first section, the article briefly outlines basic contents of the thought that is directly related to our recommendation; In the second section, from the theoretical and practical values of the doctrine, we recommend directional solutions that we believe that they will bring practical effects to the renewal of the political system in Vietnam. With the above content, in the article, we use methods such as: qualitative analysis method, generalization and systemization method, historical - logical method, data synthesis method.

ТЕОРЕТИЧНИЙ АНАЛІЗ ПОНЯТТЯ РЕГУЛЯТОРНА ПОЛІТИКА У ПЛАНУВАННІ ТА ЗАБУДОВІ МІСТА: ОСНОВНІ ПІДХОДИ ТА СУЧАСНІ ТЕНДЕНЦІЇ

Regulatory policy in the field of urban planning and development is a key element of public administration aimed at ensuring economic stability, social welfare and sustainable development. In the conditions of globalization and intensive technological progress, effective regulation of urban planning processes is becoming more and more important for solving modern challenges. The purpose of this study is to carry out a theoretical analysis of the concept of regulatory policy in urban planning and development, to study the main approaches to its implementation, as well as to determine modern trends and opportunities for adapting international experience for Ukraine. In particular, the research aims to identify effective models of regulatory policy that can be integrated into national legislation and practice to improve the management of urban development. The research uses a comprehensive approach, which includes the analysis of scientific publications, comparative analysis of regulatory policies of different countries, methods of critical literature review and data synthesis. Special attention is paid to the experience of countries such as the USA, Germany and Singapore, where various approaches to the regulation of urban planning processes are used. The analysis showed that regulatory policy in urban planning and development is a multifaceted phenomenon, which includes such basic approaches as zoning, strategic spatial planning, energy efficiency, integration of transport infrastructure and active involvement of the public in the decision‒ making process. Modern trends in regulatory policy have been identified, including innovative approaches to planning, the use of digital technologies, an emphasis on sustainable development and environmental safety. A comparative analysis of the regulatory policies of the USA, Germany, and Singapore made it possible to identify effective models that can be adapted to the conditions of Ukraine. The study confirmed that an effective regulatory policy is an important prerequisite for sustainable economic development, social stability and protection of citizens' rights. Further research should focus on the integration of digital technologies into regulatory processes, improvement of control and monitoring mechanisms, which will contribute to the sustainable development of cities in Ukraine.

METHODOLOGICAL APPROACHES TO THE ECONOMIC ASSESSMENT OF THE INNOVATIVE ACTIVITIES OF RAILWAY TRANSPORT

Objective. Justification of the need to improve the economic assessment of the innovative activity of railway transport, the development of methodological approaches and practical recommendations for the assessment of the innovative activity of railway transport, the study of requirements and assessment indicators characterizing the results of the implementation of innovations. Methods. Various methods and approaches were used in the process of performing the tasks in this work. The abstract-logical method was used to determine the research objectives. Analytical and monographic approach was used in the analysis of literary sources. The method of data analysis and synthesis for the analysis of existing methodological approaches and indicators. A comprehensive review and strategic analysis to describe promising directions for the innovative development of railway transport. Systemic, analogical, and structural-functional research methods for determining indicators of economic evaluation of innovative activity. Results. The article examines the main methodological approaches and proposes a new approach to the economic assessment of innovative activity. Scientific novelty. Recommendations on the use of an additional system of indicators for evaluating the innovative activity of railway transport are given. Practical significance. The recommendations given in the article regarding the use of the indicator system can be applied in railway transport units to make a decision on the introduction of innovations.

Challenges and opportunities of generative models on tabular data

Tabular data, organized like tables with rows and columns, is widely used. Existing models for tabular data synthesis often face limitations related to data size or complexity. In contrast, deep generative models, a part of deep learning, demonstrate proficiency in handling large and complex data sets. While these models have shown remarkable success in generating image and audio data, their application in tabular data synthesis is relatively new, lacking a comprehensive comparison with existing methods. To fill this gap, this study aims to systematically evaluate and compare the performance of deep generative models with these existing methods for tabular data synthesis, while also investigating the efficacy of post-processing techniques. We aim to identify strengths and limitations and provide insights for future research and practical applications. Our study showed that the Synthetic Minority Oversampling Technique (SMOTE) and its variants outperform deep generative models, especially for small datasets. However, we observed that an ensemble of deep generative models and post-generation processing performs better on large datasets than SMOTE alone. The results of our study indicate that deep generative models hold promise as a valuable tool for generating tabular data. Nonetheless, further research is warranted to enhance the performance of deep generative models and gain a comprehensive understanding of their limitations.

Self-Supervised Tracking via Target-Aware Data Synthesis.

While deep-learning-based tracking methods have achieved substantial progress, they entail large-scale and high-quality annotated data for sufficient training. To eliminate expensive and exhaustive annotation, we study self-supervised (SS) learning for visual tracking. In this work, we develop the crop-transform-paste operation, which is able to synthesize sufficient training data by simulating various appearance variations during tracking, including appearance variations of objects and background interference. Since the target state is known in all synthesized data, existing deep trackers can be trained in routine ways using the synthesized data without human annotation. The proposed target-aware data-synthesis method adapts existing tracking approaches within a SS learning framework without algorithmic changes. Thus, the proposed SS learning mechanism can be seamlessly integrated into existing tracking frameworks to perform training. Extensive experiments show that our method: 1) achieves favorable performance against supervised (Su) learning schemes under the cases with limited annotations; 2) helps deal with various tracking challenges such as object deformation, occlusion (OCC), or background clutter (BC) due to its manipulability; 3) performs favorably against the state-of-the-art unsupervised tracking methods; and 4) boosts the performance of various state-of-the-art Su learning frameworks, including SiamRPN++, DiMP, and TransT.

A review on learning based image reflection removal algorithms

Capturing images through semi-reflective surfaces, such as glass windows and transparent enclosures, often leads to a reduction in visual quality and can adversely affect the performance of computer vision algorithms. As a result, image reflection removal has garnered significant attention among computer vision researchers. With the growing application of deep learning methods in various computer vision tasks, such as super-resolution, inpainting, and denoising, convolutional neural networks (CNNs) have become an increasingly popular choice for image reflection removal. The purpose of this paper is to provide a comprehensive review of learning-based algorithms designed for image reflection removal. Firstly, we provide an overview of the key terminology and essential background concepts in this field. Next, we examine various datasets and data synthesis methods to assist researchers in selecting the most suitable options for their specific needs and targets. We then review existing methods with qualitative and quantitative results, highlighting their contributions and significance in this field. Finally, some considerations about challenges and future scope in image reflection removal techniques are discussed.

Synthetic Data-Driven Real-Time Detection Transformer Object Detection in Raining Weather Conditions

Images captured in rainy weather conditions often suffer from contamination, resulting in blurred or obscured objects, which can significantly impact detection performance due to the loss of identifiable texture and color information. Moreover, the quality of the detection model plays a pivotal role in determining detection outcomes. This study adopts a dual perspective, considering both pre-trained models and training data. It employs 15 image augmentation techniques, combined with neural style transfer (NST), CycleGAN, and an analytical method, to synthesize images under rainy conditions. The Real-Time Detection Transformer (RTDETR) and YOLOv8 pre-trained models are utilized to establish object detection frameworks tailored for rainy weather conditions. Testing is carried out using the DAWN (Detection in Adverse Weather Nature) dataset. The findings suggest compatibility between the pre-trained detection models and various data synthesis methods. Notably, YOLOv8 exhibits better compatibility with CycleGAN data synthesis, while RTDETR demonstrates a stronger alignment with the NST and analytical approaches. Upon the integration of synthesized rainy images into model training, RTDETR demonstrates significantly enhanced detection accuracy compared to YOLOv8, indicating a more pronounced improvement in performance. The proposed approach of combining RTDETR with NST in this study shows a significant improvement in Recall (R) and mAP50-95 by 16.35% and 15.50%, respectively, demonstrating the robust rainy weather resilience of this method. Additionally, RTDETR outperforms YOLOv8 in terms of inference speed and hardware requirements, making it easier to use and deploy in real-time applications.

Psychosocial barriers and facilitators for cascade genetic testing in hereditary breast and ovarian cancer: a scoping review.

Despite increased awareness and availability of genetic testing for hereditary breast and ovarian cancer (HBOC) syndrome for over 20years, there is still significant underuse of cascade genetic testing among at-risk relatives. This scoping review synthesized evidence regarding psychosocial barriers and facilitators of family communication and/or uptake of cascade genetic testing in relatives from HBOC families. Search terms included 'hereditary breast and ovarian cancer' and 'cascade genetic testing' for studies published from 2012-2022. Through searching common databases, and manual search of references, 480 studies were identified after excluding duplications. Each article was reviewed by two researchers independently and 20 studies were included in the final analysis. CASP, RoBANS 2.0, RoB 2.0, and MMAT were used to assess the quality of included studies. A convergent data synthesis method was used to integrate evidence from quantitative and narrative data into categories and subcategories.Evidence points to 3 categories and 12 subcategories of psychosocial barriers and facilitators for cascade testing: (1) facilitators (belief in health protection and prevention; family closeness; decisional empowerment; family support, sense of responsibility; self-efficacy; supportive health professionals); (2) bidirectional concepts (information; perception of genetic/cancer consequences; negative emotions and attitude); and (3) barriers (negative reactions from family and negative family dynamics). Healthcare providers need to systematically evaluate these psychosocial factors, strengthen facilitators and alleviate barriers to promote informed decision-making for communication of genetic test results and uptake of genetic testing. Bidirectional factors merit special consideration and tailored approaches, as they can potentially have a positive or negative influence on family communication and uptake of genetic testing.

Complex smoke removal in image: Integrating frequency learning network with data synthesis method

This paper introduces a novel neural network architecture and an enhanced data synthesis method that significantly boost the performance in removing complex smoke from images. The architecture features a multi-branch and multi-scale feature fusion design, which effectively integrates multiple feature streams and adaptively restores the background by identifying specific smoke characteristics within the image. A newly designed Fourier residual block is incorporated to capture frequency domain information, enabling the network to process and transform information across both spatial and frequency domains. To improve the network’s generalization ability and robustness, an in-depth analysis of the imaging process in smoky environments was conducted, leading to an improved method for synthesizing smoke images. This methodology facilitates the creation of a more varied and realistic training dataset, substantially enhancing the neural network’s capabilities in image restoration. Experimental results show that this approach is highly effective on both synthetic and real-world smoke datasets, outperforming existing image de-smoking methods in terms of quantitative metrics and visual perception. The code for this method is available at https://github.com/Exiagit/MFSR.

Automated detection of railway defective fasteners based on YOLOv8-FAM and synthetic data using style transfer

Fastener damage detection is an integral component of track safety inspections. The lack of balanced dataset caused by insufficient data on defective fasteners poses a significant challenge to the current development of robust fastener detection models. This study proposes the YOLOv8-FAM detection model algorithm, which combines the enhanced capabilities of YOLOv8, and generates realistic images of defective fasteners by performing style transfer on masked images of non-defective fasteners, thereby creating a balanced dataset for training the detection model. Experimental results show that the defect detection accuracy of YOLOv8-FAM improves by 8% compared to the original model, while also reducing the data acquisition cost by 40%. The YOLOv8-FAM model surpasses existing models in detecting defective fasteners while minimizing inference costs to the maximum extent. The proposed style transfer data synthesis method drives the practical deployment of deep learning, offering an efficient and cost-effective solution for the transportation infrastructure industry.