Application Domains Research Articles

Bridging the Gap Between Human Toxicology and Ecotoxicology Under One Health Perspective by a Cross-Species Adverse Outcome Pathway Network for Reproductive Toxicity.

Although ecotoxicological and toxicological risk assessments are performed separately from each other, recent efforts have been made in both disciplines to reduce animal testing and develop predictive approaches instead, for example, via conserved molecular markers, and in vitro and in silico approaches. Among them, adverse outcome pathways (AOPs) have been proposed to facilitate the prediction of molecular toxic effects at larger biological scales. Thus, more toxicological data are used to inform on ecotoxicological risks and vice versa. An AOP has been previously developed to predict reproductive toxicity of silver nanoparticles via oxidative stress on the nematode Caenorhabditis elegans (AOPwiki ID 207). Following this previous study, our present study aims to extend the biologically plausible taxonomic domain of applicability (tDOA) of AOP 207. Various types of data, including in vitro human cells, in vivo, and molecular to individual, from previous studies have been collected and structured into a cross-species AOP network that can inform both human toxicology and ecotoxicology risk assessments. The first step was the collection and analysis of literature data to fit the AOP criteria and build a first AOP network. Then, key event relationships were assessed using a Bayesian network modeling approach, which gave more confidence in our overall AOP network. Finally, the biologically plausible tDOA was extended using in silico approaches (Genes-to-Pathways Species Conservation Analysis and Sequence Alignment to Predict Across Species Susceptibility), which led to the extrapolation of our AOP network across over 100 taxonomic groups. Our approach shows that various types of data can be integrated into an AOP framework, and thus facilitates access to knowledge and prediction of toxic mechanisms without the need for further animal testing. Environ Toxicol Chem 2024;00:1-14. © 2024 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Databases in Edge and Fog Environments: A Survey

While a significant number of databases are deployed in cloud environments, pushing part or all data storage and querying planes closer to their sources (i.e., to the edge) can provide advantages in latency, connectivity, privacy, energy, and scalability. This article dissects the advantages provided by databases in edge and fog environments by surveying application domains and discussing the key drivers for pushing database systems to the edge. At the same time, it also identifies the main challenges faced by developers in this new environment and analyzes the mechanisms employed to deal with them. By providing an overview of the current state of edge and fog databases, this survey provides valuable insights into future research directions.

Prediction of Human Liver Microsome Clearance with Chirality-Focused Graph Neural Networks.

In drug candidate design, clearance is one of the most crucial pharmacokinetic parameters to consider. Recent advancements in machine learning techniques coupled with the growing accumulation of drug data have paved the way for the construction of computational models to predict drug clearance. However, concerns persist regarding the reliability of data collected from public sources, and a majority of current in silico quantitative structure-property relationship models tend to neglect the influence of molecular chirality. In this study, we meticulously examined human liver microsome (HLM) data from public databases and constructed two distinct data sets with varying HLM data quantity and quality. Two baseline models (RF and DNN) and three chirality-focused GNNs (DMPNN, TetraDMPNN, and ChIRo) were proposed, and their performance on HLM data was evaluated and compared with each other. The TetraDMPNN model, which leverages chirality from 2D structure, exhibited the best performance with a test R2 of 0.639 and a test root-mean-squared error of 0.429. The applicability domain of the model was also defined by using a molecular similarity-based method. Our research indicates that graph neural networks capable of capturing molecular chirality have significant potential for practical application and can deliver superior performance.

Logistics 4.0 – digital transformation with smart connected tracking and tracing devices

Tracking and tracing devices can provide real-time information from the supply chain and enable digital transformation in the logistics and supply chain industry. In this connection, Logistics 4.0 refers to the potential for information technology and smart connected assets to be used in logistics in the same way as the Industry 4.0 concept which has been applied in operations and manufacturing. This paper analyses current tracking and tracing-focused applications that can provide value for logistics operations through a case study approach.This study employs an exploratory multiple-case study approach, which is based on interviews with development project stakeholders. The paper analyses three industrial case studies and how tracking/tracing applications are connected to value processes.The findings of this study show that the value of Logistics 4.0, through the deployment of tracking/tracing applications, is delivered in terms of operational efficiency, visibility, transparency, and safety/security. The payback depends on the volume of transactions, the possibility of reusing the tracking tags, the duration of trips, and the supply chain structure. The paper provides insight into how Logistics 4.0 technology can enhance logistics performance value. Based on the analysis, the study proposes the following potential application domains: (1) intermodal tracking of the shipments for operations control, (2) asset management of containers, and (3) certification of the process steps and authenticity.

Forming the conceptual foundations of a sustainable development accounting system and reflecting its results in reporting

The subject of this study is sustainable development accounting, particularly its role in shaping environmental and social indicators of enterprises. The paper pays special attention to the issue of insufficient integration of environmental and social aspects into conventional financial accounting. It identifies the need to devise methodologies that include social and ecological indicators in financial reporting, to provide a more comprehensive understanding of the impact of enterprise activities on society and the environment. The results demonstrate that the implementation of sustainable development accounting allows for a comprehensive assessment of the financial, social, and environmental impact of the enterprise. Analysis of sources proves that enterprises that use this approach have better opportunities for risk forecasting and resource management, which increases their competitiveness and responsibility to society. The interpretation of the results emphasizes that the success of sustainable development accounting is conditioned by the ability to integrate comprehensive data about the impact of enterprise activities on the environment and society into the general management system. Such integration allows enterprises not only to meet regulatory requirements but also to enhance their overall value and image. The features and distinctive characteristics of the results are that they enable companies to conduct more transparent and responsible business. The implementation of sustainable development accounting systems gives enterprises the ability not only to reduce the negative impact on the environment but also to effectively engage resources to address social issues. The domain of practical application of the results extends to corporate governance, socially responsible business, investment analysis, and environmental assessment. It includes internal and external regulation, stakeholder requirements, and strategic planning at enterprises

Digitalization, AI, and robotics for good care and work? German policy imaginaries of healthcare technologies

Abstract Healthcare is being discussed as an increasingly important application domain for new digital, artificial intelligence–enabled, and robotic technologies. This area has garnered interest from policymakers who seek to harness technological advancements in their quest for providing adequate healthcare to aging populations and modernizing healthcare jobs. In this article, we analyze the sociotechnical imaginary of healthcare technology constructed in German public policy. We analyze 22 pertinent German policy documents from the years 2018–22, drawing on a conceptual frame of sociotechnical imaginaries and dramaturgical studies of futuring. Our analysis shows how healthcare is imagined as a sector in crisis, how technology is promoted as a solution, and how anticipated resistances to technological solutions are being met with the narratives of “technological assistance,” the provision of “good care,” and the facilitation of “good work” within the healthcare sector.

Technical debt in AI-enabled systems: On the prevalence, severity, impact, and management strategies for code and architecture

Context:Artificial Intelligence (AI) is pervasive in several application domains and promises to be even more diffused in the next decades. Developing high-quality AI-enabled systems — software systems embedding one or multiple AI components, algorithms, and models — could introduce critical challenges for mitigating specific risks related to the systems’ quality. Such development alone is insufficient to fully address socio-technical consequences and the need for rapid adaptation to evolutionary changes. Recent work proposed the concept of AI technical debt, a potential liability concerned with developing AI-enabled systems whose impact can affect the overall systems’ quality. While the problem of AI technical debt is rapidly gaining the attention of the software engineering research community, scientific knowledge that contributes to understanding and managing the matter is still limited. Objective:In this paper, we leverage the expertise of practitioners to offer useful insights to the research community, aiming to enhance researchers’ awareness about the detection and mitigation of AI technical debt. Our ultimate goal is to empower practitioners by providing them with tools and methods. Additionally, our study sheds light on novel aspects that practitioners might not be fully acquainted with, contributing to a deeper understanding of the subject. Method:We develop a survey study featuring 53 AI practitioners, in which we collect information on the practical prevalence, severity, and impact of AI technical debt issues affecting the code and the architecture other than the strategies applied by practitioners to identify and mitigate them. Results:The key findings of the study reveal the multiple impacts that AI technical debt issues may have on the quality of AI-enabled systems (e.g., the high negative impact that Undeclared consumers has on security, whereas Jumbled Model Architecture can induce the code to be hard to maintain) and the little support practitioners have to deal with them, limited to apply manual effort for identification and refactoring. Conclusion:We conclude the article by distilling lessons learned and actionable insights for researchers.

Step Forward Cross Validation for Bioactivity Prediction: Out of Distribution Validation in Drug Discovery.

Recent advances in machine learning methods for materials science have significantly enhanced accurate predictions of the properties of novel materials. Here, we explore whether these advances can be adapted to drug discovery by addressing the problem of prospective validation - the assessment of the performance of a method on out-of-distribution data. First, we tested whether k-fold n-step forward cross-validation could improve the accuracy of out-of-distribution small molecule bioactivity predictions. We found that it is more helpful than conventional random split cross-validation in describing the accuracy of a model in real-world drug discovery settings. We also analyzed discovery yield and novelty error, finding that these two metrics provide an understanding of the applicability domain of models and an assessment of their ability to predict molecules with desirable bioactivity compared to other small molecules. Based on these results, we recommend incorporating a k-fold n-step forward cross-validation and these metrics when building state-of-the-art models for bioactivity prediction in drug discovery.

Low Resource Summarization using Pre-trained Language Models

With the advent of Deep Learning based Artificial Neural Networks models, Natural Language Processing (NLP) has witnessed significant improvements in textual data processing in terms of its efficiency and accuracy. However, the research is mostly restricted to high-resource languages such as English and low-resource languages still suffer from a lack of available resources in terms of training datasets as well as models with even baseline evaluation results. Considering the limited availability of resources for low-resource languages, we propose a methodology for adapting self-attentive transformer-based architecture models (mBERT, mT5) for low-resource summarization, supplemented by the construction of a new baseline dataset (76.5k article, summary pairs) in a low-resource language Urdu. Choosing news (a publicly available source) as the application domain has the potential to make the proposed methodology useful for reproducing in other languages with limited resources. Our adapted summarization model urT5 with up to 44.78% reduction in size as compared to mT5 can capture contextual information of low resource language effectively with evaluation score (up to 46.35 ROUGE-1, 77 BERTScore) at par with state-of-the-art models in high resource language English (PEGASUS: 47.21, BART: 45.14 on XSUM Dataset) . The proposed method provided a baseline approach towards extractive as well as abstractive summarization with competitive evaluation results in a limited resource setup.

Prompting GPT –4 to support automatic safety case generation

In the ever-evolving field of software engineering, the advent of large language models and conversational interfaces, exemplified by ChatGPT, represents a significant revolution. While their potential is evident in various domains, this paper expands upon our previous research, where we experimented with GPT–4, on its ability to create safety cases. A safety case is a structured argument supported by a body of evidence to demonstrate that a given system is safe to operate in a given environment. In this paper, we first determine GPT–4’s comprehension of the Goal Structuring Notation (GSN), a well-established notation for visually representing safety cases. Additionally, we conduct four distinct experiments using GPT–4 to evaluate its ability to generate safety cases within a specified system and application domain. To assess GPT–4’s performance in this context, we compare the results it produces with the ground-truth safety cases developed for an X-ray system, a machine learning-enabled component for tire noise recognition in a vehicle, and a lane management system from the automotive domain. This comparison enables us to gain valuable insights into the model’s generative capabilities. Our findings indicate that GPT–4 is able to generate moderately accurate and reasonable safety cases.

Attenuating majority attack class bias using hybrid deep learning based IDS framework

In real-time application domains, like finance, healthcare and defence, delay in service or stealing information may lead to unrecoverable consequences. So, early detection of intrusion is important to prevent security breaches. In recent days, anomaly-based intrusion detection using Hybrid Deep Learning approaches are becoming more popular. The most used benchmark datasets in the literature are NSL-KDD and UNSW-NB15, and these datasets are imbalanced. The models built on imbalanced datasets may lead to biased results towards majority classes by neglecting the minority class, even though they are equally important. In many cases, high accuracy is achieved for majority classes in the imbalanced datasets. But, the class-level performances are poor with respect to the minority class. The class balancing will also play an important role in attenuating the bias in prediction for imbalanced datasets. In this paper, a Hybrid Deep Learning Based Intrusion Detection (HDLBID) framework is proposed with CNN-BiLSTM combination. The four techniques, namely, Random Oversampling (ROS), ADASYN, SMOTE, and SMOTE-Tomek, are used for class balancing in the proposed HDLBID framework. The proposed HDLBID with SMOTE-Tomek achieves an overall accuracy of 99.6% with NSL-KDD and 89.02% for UNSW-NB15. It results in an improvement of 13.67% for NSL-KDD and 10.62% for UNSW-NB15 over the existing recent related models. In the proposed HDLBID, in addition to overall accuracy, the class-level F1 score is also calculated. A comparative study is presented to show the effectiveness of balancing dataset compared to imbalanced dataset, and observed that the SMOTE-Tomek class balancing comparatively performed well. An improvement of 37.43% is observed in the U2R class of the NSL-KDD dataset and 61.65% improvement is seen in the Worms class of the UNSW-NB15 dataset, both with SMOTE-Tomek class balancing. Therefore, the proposed HDLBID with SMOTE-Tomek class balancing reports the best results in terms of overall accuracy compared to existing recent related approaches. Also, in terms of class-level analysis, HDLBID reports best results with SMOTE-Tomek over imbalanced version of datasets.

A SAR and QSAR study on 3CLpro inhibitors of SARS-CoV-2 using machine learning methods

ABSTRACT The 3C-like Proteinase (3CLpro) of novel coronaviruses is intricately linked to viral replication, making it a crucial target for antiviral agents. In this study, we employed two fingerprint descriptors (ECFP_4 and MACCS) to comprehensively characterize 889 compounds in our dataset. We constructed 24 classification models using machine learning algorithms, including Support Vector Machine (SVM), Random Forest (RF), extreme Gradient Boosting (XGBoost), and Deep Neural Networks (DNN). Among these models, the DNN- and ECFP_4-based Model 1D_2 achieved the most promising results, with a remarkable Matthews correlation coefficient (MCC) value of 0.796 in the 5-fold cross-validation and 0.722 on the test set. The application domains of the models were analysed using dSTD-PRO calculations. The collected 889 compounds were clustered by K-means algorithm, and the relationships between structural fragments and inhibitory activities of the highly active compounds were analysed for the 10 obtained subsets. In addition, based on 464 3CLpro inhibitors, 27 QSAR models were constructed using three machine learning algorithms with a minimum root mean square error (RMSE) of 0.509 on the test set. The applicability domains of the models and the structure-activity relationships responded from the descriptors were also analysed.

A Review of Impedance Spectroscopy Technique: Applications, Modelling, and Case Study of Relative Humidity Sensors Development

Impedance Spectroscopy (IS) is a general term for the technique referring to small-signal measurements of the linear electrical response of a domain of interest. This method is based on the analysis of the system’s electrical response to yield helpful information about its domain-dependent physicochemical properties (generally, the analysis is carried out in the frequency domain). Nowadays, there are many areas of application where IS can be used to evaluate or enhance the development of emerging products and processes. As a contribution to this field of research, this paper presents relevant theoretical–practical aspects of the interpretation and analysis of the electrical behavior of materials based on IS and IS modelling. The work starts by historically introducing IS and then goes through different domains of application of the technique, such as Materials Science and correlated areas. Afterwards, an introduction to IS usage for constructing equivalent electrical circuits is presented, aiming at modelling the materials’ electrical behavior, followed by examples from the literature that use the two possible circuit development approaches, the series and the parallel association of circuit elements. Lastly, the authors present a case study of their most recent efforts of a circuit model development of relative humidity (RH) sensors based on heterogeneous mixed metal oxide (MMO) nanostructures, used to understand and identify existing contributions to the overall electrical response of the sensors to moisture; in their case, the electrical response of the MMO sensors was modelled with a high level of superposition between the experimental and fitted data, using a parallel combination of circuit elements, which is an unconventional one.

Deep learning accelerated numerical simulation for three-dimensional compressible fluids

Numerical simulation of fluid flow is a long-standing challenge across many physical application domains, including engineering, climate, and the physical science. There has been a surge of interest in high order schemes aimed at improving simulation accuracy on coarse grids. However, for high-dimensional fluids, the computational cost escalates with the number of dimensions involved. In this paper, we propose a deep learning-based approach to accelerate the numerical computation and further improve the accuracy in simulating three-dimensional (3D) compressible fluids which can be described by Eulerian equations. The proposed work utilizes 3D Euler transformer networks to learn the interpolation coefficients for cell boundaries, which are applied to approximate the boundary fluxes of fluid on coarser grids. Benefiting from learning features of high-resolution fluid flow, our learned interpolation method yields finer performance on coarse grids, thereby accelerating the fluid simulations and improving the numerical accuracy. The numerical experiments confirm that the proposed method improves performance in inference of coarse-grained dynamics.

Unsupervised Feature Extraction From Raw Data for Gesture Recognition With Wearable Ultralow-Power Ultrasound.

Wearable ultrasound (US) is a novel sensing approach that shows promise in multiple application domains, and specifically in hand gesture recognition (HGR). In fact, US enables to collect information from deep musculoskeletal structures at high spatiotemporal resolution and high signal-to-noise ratio, making it a perfect candidate to complement surface electromyography for improved accuracy performance and on-the-edge classification. However, existing wearable solutions for US-based gesture recognition are not sufficiently low power for continuous, long-term operation. On top of that, practical hardware limitations of wearable US devices (limited power budget, reduced wireless throughput, and restricted computational power) set the need for the compressed size of models for feature extraction and classification. To overcome these limitations, this article presents a novel end-to-end approach for feature extraction from raw musculoskeletal US data suited for edge computing, coupled with an armband for HGR based on a truly wearable (12 cm2, 9 g), ultralow-power (ULP) (16 mW) US probe. The proposed approach uses a 1-D convolutional autoencoder (CAE) to compress raw US data by 20× while preserving the main amplitude features of the envelope signal. The latent features of the autoencoder are used to train an XGBoost classifier for HGR on datasets collected with a custom US armband, considering armband removal/repositioning in between sessions. Our approach achieves a classification accuracy of 96%. Furthermore, the proposed unsupervised feature extraction approach offers generalization capabilities for intersubject use, as demonstrated by testing the pretrained encoder on a different subject and conducting posttraining analysis, revealing that the operations performed by the encoder are subject-independent. The autoencoder is also quantized to 8-bit integers and deployed on a ULP wearable US probe along with the XGBoost classifier, allowing for a gesture recognition rate ≥ 25 Hz and leading to 21% lower power consumption [at 30 frames/s (FPS)] compared to the conventional approach (raw data transmission and remote processing).

FedComm: Federated Learning as a Medium for Covert Communication

Proposed as a solution to mitigate the privacy implications related to the adoption of deep learning, Federated Learning (FL) enables large numbers of participants to successfully train deep neural networks without revealing the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">actual</i> private training data. To date, a substantial amount of research has investigated the security and privacy properties of FL, resulting in a plethora of innovative attack and defense strategies. This paper thoroughly investigates the communication capabilities of an FL scheme. In particular, we show that a party involved in the FL learning process can use FL as a covert communication medium to send an arbitrary message. We introduce FedComm, a novel covert-communication technique that enables robust sharing and transfer of targeted payloads within the FL framework. Our extensive theoretical and empirical evaluations show that FedComm provides a stealthy communication channel, with minimal disruptions to the training process. Our experiments show that FedComm successfully delivers 100% of a payload in the order of kilobits before the FL procedure converges. Our evaluation also shows that FedComm is independent of the application domain and the neural network architecture used by the underlying FL scheme.

Innovative Machine Learning Solutions: Investigating Algorithms and Applications

This comprehensive review paper provides a thorough examination of the ever-evolving landscape of machine learning (ML), spanning from its historical origins to contemporary applications, challenges, and future prospects. It begins by elucidating the foundational concepts of machine learning, encompassing the diverse types, key terminology, and the intricate pipeline involved in the machine learning process. Delving into its historical perspective, the paper chronicles the development of machine learning, tracing its roots in artificial intelligence and highlighting key milestones and influential researchers who have shaped its trajectory. The core of this review explores an array of machine learning algorithms and techniques, spanning regression, classification, clustering, dimensionality reduction, deep learning, and ensemble methods. These algorithms are contextualized with real-world applications, ranging from healthcare to finance, natural language processing (NLP), computer vision, recommender systems, and robotics. Each application domain is buttressed with illustrative examples and case studies. In recognizing the challenges and open problems that confront machine learning, the review delves into issues pertaining to data quality, model interpretability, bias, ethics, generalization, and overfitting. Moreover, it identifies pressing research questions and areas where advancements are needed. Recent trends and developments in machine learning, including transfer learning, explainable AI, federated learning, reinforcement learning, and ethical considerations, are also highlighted to provide a glimpse into the evolving landscape. The paper culminates in a thought-provoking discussion on the future of machine learning, its potential societal impact, and its transformative role across industries. In summary, this review amalgamates key findings and insights, offering a comprehensive view of machine learning's multifaceted journey, from its inception to its promising future.

Image Segmentation Techniques: An In-Depth Review and Analysis

In the fields of computer vision and image processing, image segmentation is a very important task. The image is partitioned into segments or regions then the visual data can be understood, analyzed and used easily. In this article, a comprehensive review of image segmentation methods is presented. It covers both the strengths and the advantages of some techniques as well as the weaknesses and limitations of other algorithms. The techniques used in image segmentation can be divided into four groups based on thresholding, region, edge and deep learning techniques. The fundamental principles, popularity of algorithms and evaluation metrics are mentioned. Also, the applications domains and future directions with current challenges are presented.

Comparison of Fuzzy Level Set Search and Rescue Optimization and Clustering-Based Segmentation Techniques for Brain Tumor Detection

Abstract: Brain tumor detection is a critical task in medical imaging that requires accurate and efficient image segmentation techniques. This paper presents a comparative study of two advanced segmentation methods: Fuzzy Level Set Search and Rescue Optimization (FLSSRO) and Clustering-Based Segmentation (CBS). By examining the theoretical foundations, algorithmic structures, and performance metrics of these techniques, we aim to highlight their strengths and limitations in the context of brain tumor detection. Through experimental evaluation on benchmark MRI datasets, we analyze the efficacy, accuracy, and computational efficiency of both methods. The study concludes with insights into the appropriate application domains for each technique and potential future research directions.

Revisited UV- spectra of chlorohydroxoaurate anions

The most important ionic precursor of gold, [AuCl4]-, is used in aqueous solution leading to chlorohydroxoaurates species, [AuCl4-x(OH)x]- (x = 1–4) due to partial hydrolysis. Their UV spectral signatures are still relatively unknown though very useful in many domains of application. Individual spectra of each of them are determined for the first time thanks to a thorough experimental investigation comprising the range 200–250 nm, surpringly ignored up to now. New isosbestic points useful for species partition analysis are evidenced. Electronic transition attribution is obtained from quantum chemical calculations based on TD-DFT. The prediction of the experimental blueshifted bands of the [AuCl4-x(OH)x]-1 anions was possible only after applying energy corrections calibrated on the full UV range two-band spectrum of the [AuCl4]- complex.