Materials characterization: Can artificial intelligence be used to address reproducibility challenges?

Aim: The majority of machine learning (ML) models in healthcare are built on retrospective data, much of which is collected without explicit patient consent for use in artificial intelligence (AI) and ML applications. The primary aim of this study was to evaluate whether clinicians and scientific researchers themselves consent to provide their own data for the training of ML models. Materials and Methods: This survey was conducted through an anonymous online survey, utilizing platforms such as Telegram, LinkedIn, and Viber. The target audience comprised specific international groups, primarily Russian, German, and English-speaking, of clinicians and scientific researchers. These participants ranged in their levels of expertise and experience, from beginners to veterans. The survey centered on a singular, pivotal question: “Do You Consent to the Use of Your Biological and Private Data for Training Machine Learning and AI Models?” Respondents had the option to choose from three responses: “Yes” and “No”. Results: The survey was conducted in January 2024. A total of 119 unique and verified individuals participated in the survey. The results revealed that only 50% of respondents (63 persons) expressed consent to provide their own data for the training of ML and AI models. Conclusion: In the development of ML and AI models, particularly open-source ones, it is crucial to ascertain whether participants are willing to provide their private data. While ML algorithms can transform the nature of data, it is important to remember that the primary owner of this data is the individual. Our findings show that in 50% of the cases, even participants from scientific research and clinical backgrounds – individuals typically accountable for ensuring data quality in AI and ML model development – do not consent to the use of their data in AI and ML settings. This highlights the need for more stringent consent processes and ethical considerations in the utilization of personal data in AI and ML research.

HADA: An automated tool for hardware dimensioning of AI applications

The integration of artificial intelligence (AI) is transforming medical research and the healthcare system with its role in the management of complex and global medical issues such as cancer. AI can improve operational efficiency to reduce the burden on the medical professional. Integration of AI into the flow from early diagnosis to treatment and remission can be beneficial to efficiently manage various diseases and help improve patient life expectancy. Integration of AI, Machine Learning (ML) and Deep Learning (DL) in healthcare to build accurate, efficient, and effective tools to solve complex healthcare challenges can open the path to making healthcare accessible, affordable, and equitable throughout the world. Gene expressions and regulations are the key to cell development, DNA damage repair, genomic integrity, and responses to drug therapies. The TP53 gene, which is also known as the 'Guardian of Genomes', regulates the production of the p53 protein. Various environmental and bodily stresses can trigger the synthesis of mutated p53, a common finding in many types of cancers. Many studies focus on molecular understanding of mutations, genomic networks, and mutation effects. We propose a machine learning (ML) framework to build generic predictive models for classification, sub-typing, and progression of cancers using mutation-based TP53 data. The primary goal of our study is to showcase the potential of using TP53 based data in building ML tools that can be useful for cancer research and diagnosis. One key challenge to achieve this goal is to find a comprehensive TP53 data set which includes important features to train the models and provide useful results. However, most publicly available datasets are curated from previous publications and lack uniformity or correlation. In this work, we use the curated TP53 mutation data set to build and train ML frameworks and models after applying data processing techniques to infer cancer information. Through data augmentation and feature selection, we achieved <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$&gt;80 \%$</tex> accuracy in both classification and subtyping. The progression model to predict tumor grade and cancer stage achieves an accuracy of around <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$&gt; 4 8 \%$</tex>. Our study identifies gaps in the available data that can be addressed to improve the model results in the future. The data augmentation can result in data leakage, thereby producing overly accurate results, the limitation that are discussed in the paper. Our approach underscores the potential of integrating multiple types of TP53 genomic data with machine learning models to build useful analytical tools for cancer research.

Nailfold capillaroscopy (NFC) is a well-established, highly specialized, non-invasive method for assessing microcirculation and represents the “gold standard” in rheumatology for distinguishing primary from secondary Raynaud’s phenomenon (RP). Through analysis of capillary structures in the nailfold region, clinicians can identify diff erent patterns (scleroderma, scleroderma-like, and non-specifi c). Despite its clinical signifi cance, image interpretation remains challenging due to subjectivity and the requirement for substantial examiner expertise. In recent years, artifi cial intelligence (AI) has off ered novel solutions through the automated recognition and analysis of capillary structures. Machine learning and deep learning approaches have demonstrated high eff ectiveness in detecting abnormalities, classifying capillary patterns, and predicting the risk of systemic sclerosis. Projects such as CAPI-Detect, along with various other machine learning, deep learning, and neural network models (DenseNet-121, Effi cientNet-B0, ResNet-34, NFC-Net), have shown that AI can identify novel quantitative parameters not accessible through traditional visual assessment and can increase the objectivity and reproducibility of results up to 90%. Most studies in this fi eld focus on capillaroscopic patterns in systemic sclerosis (SSc); however, pilot studies in juvenile dermatomyositis, diabetes mellitus, and hypertension further highlight the applicability of these technologies. Some AI models are even capable of distinguishing onychomycosis, nail psoriasis, and subungual melanoma. Machine learning and deep learning models based on image analysis (Vision Transformer, ViT) appear to represent an additional valid system for the early and rapid interpretation of NFC images and morphological biomarkers in systemic sclerosis (SSc), integrating EULAR-validated algorithms for distinguishing scleroderma from non-scleroderma patterns with artifi cial intelligence (AI). Beyond its diagnostic applications, AI also holds potential in disease monitoring, assessment of therapeutic response, and the development of personalized treatment strategies. The processing of digital images through validated machine learning algorithms, neural networks, and related approaches is aligned with the priorities of the National Health Strategy 2030 for digitalization and the development of eHealth (Policy 2.5). The establishment of a National Digital Platform for Medical Diagnostics is envisaged, aimed at supporting all medical specialties. This platform will be integrated with the National Health Information System and the electronic patient record, with the primary objective of improving the quality of healthcare services. The integration of AI into medical practice opens new opportunities to support early diagnosis and improved management of patients with microangiopathy, facilitating timely detection and the prevention of complications, with the ultimate aim of ensuring better quality of life and preserved functional capacity.

Artificial intelligence (AI) and machine learning (ML) models have been increasingly used in the diagnosis of neurodegenerative disorders. These models have the potential to improve diagnostic accuracy, reduce the burden on healthcare systems, and improve patient outcomes. However, there are several challenges that need to be addressed for these models to be widely adopted in clinical practice. This article provides a summary of the current state of AI and ML models for neurodegenerative disorder diagnosis, including their strengths and limitations. It also discusses the challenges faced in the field, such as the need for large and diverse datasets, the difficulty of obtaining accurate and reliable medical imaging data, and the need for robust and interpretable models. Furthermore, it gives an overview of the recent developments in the field such as the use of deep learning, transfer learning, and multimodal medical image fusion techniques for the diagnosis of neurodegenerative disorders. The article highlights the need for more research and development in the field, specifically in areas such as the integration of multiple data modalities, the use of explainable AI for clinical decision making, and the development of personalized treatment plans. Finally, it suggests future research directions for the field, such as the need for more rigorous evaluation of AI models in clinical settings, the integration of AI with other diagnostic and therapeutic modalities, and the development of AI-based decision support systems for clinicians.

Achieving operational excellence and efficiency is a constant goal in manufacturing and production, driven by the need to reduce waste and optimize processes in today’s competitive environment. The era of Industry 4.0 offers advanced technologies such as Artificial Intelligence (AI), robotics, IoT and cloud computing to support this pursuit, aligning with philosophies like lean manufacturing. However, modeling and optimizing complex production systems, which involve nonlinear behavior, interdependencies and time delays, remains a significant challenge. This research proposes a hybrid AI solution for optimizing production efficiency by combining Data Envelopment Analysis (DEA), Machine Learning (ML)-based simulation and Genetic Algorithms (GA). DEA is used to identify the efficient frontier of the production system based on historical or synthetic data. This data is used to train an ML model capable of simulating system behavior. GA is then applied to search for the optimal input configurations and control parameters to maximize efficiency, considering operational constraints. The methodology was applied and validated using real sensor data from a thermoelectric power plant. The process included data preparation with outlier removal using the Z-score method, calculation of relative efficiency using DEA-CCR, training ML models to predict outputs and efficiencies, and applying GA to identify ideal settings. The implementation resulted in significant improvements, with efficiency gains of 25% and 58% in two tested models, outperforming conventional closed-loop control systems. The approach also demonstrated that separating the optimization phase using GA from the DEA and ML training phases can significantly reduce computational time. As a result, the solution not only optimizes resource use and reduces production costs but also provides greater flexibility to production systems by reducing the direct dependency between high output levels and overall efficiency. The developed framework presents a practical, accessible and effective solution for production systems. It is adaptable to various industrial sectors and contributes to the data-driven digital transformation within the scope of Industry 4.0.

Batteries, recognized as effective energy storage solutions, are considered the main facilitators of the world-wide transition towards clean and renewable energy sources. Among different types of batteries, lithium-ion (Li-ion) variants offer higher energy densities and relatively longer life spans when compared to other types. Nonetheless, a primary concern with these batteries is their lifetime. Batteries undergo various degradation mechanisms under storage and use, significantly impacting their lifespan. To this end, it is crucial to predict the degradation and lifetime of Li-ion batteries under given conditions.Researchers use three main methodologies to perform battery health diagnostics and to predict the lifetime of batteries. One approach revolves around using mechanistic, first-principle electrochemical models, also known as physics-based models. If equipped with proper thermodynamic theories, such models can show promising capabilities; however, holistic degradation prediction with these models is still challenging due to the computational complexities and the multitude of parameters that need to be fine-tuned in this approach. The other common strategy is to use empirical models to predict battery degradation. These models generally entail fewer parameters to be identified and are computationally less intensive to solve. Nonetheless, empirical models can suffer from the accuracy point-of-view as they are constrained to predict degradation trends introduced by certain degradation modes. Another common technique in battery life prediction is to utilize purely data-driven methods, such as machine learning (ML) algorithms, which also have shown promising results in the literature on rapid health predictions. However, these methods require large volumes of experimental data for training and testing ML models to ensure accuracy. In addition, data-driven methods are likely to extrapolate poorly to conditions beyond their training data and are indifferent towards the underlying degradation mechanisms. Recently, physics-informed machine learning (PI-ML) methods have garnered significant attention. They integrate physics-based or empirical models (developed based on physics) with a data-driven approach and allow one to train ML models on a smaller set of experimental data.To the best of the authors’ knowledge, the performance comparison between first-principle and empirical models when integrated within PI-ML remains unclear. Therefore, in this work, we aim to compare these two models when applied to prognostics (capacity forecasting and remaining useful life prediction) of a set of Li-ion batteries. To perform this study, we generate aging data for 40 Li-ion coin cells cycled under randomized conditions. Each cell undergoes a three-step charging stage followed by a two-step discharge stage. After obtaining the aging data, we will develop two PI-ML models, one equipped with a physics-based model and another with a set of empirical models. Both PI-ML models in this work will follow the sequential integration approach, where the training data for the final PI-ML model come from both experimental and computational data, the latter of which are obtained from the physics-based or empirical models. The parameters for the physics-based and empirical models are identified from another set of experimental data. Finally, the PI-ML models will be tested with experimental data obtained at different cycling conditions. The data flow for the sequential architecture of PI-ML is shown in the attached figure. This comparative study will help identify the performance of physics-based and empirical models when integrated into PI-ML. The main performance metric considered in this work is each model’s ability to extrapolate beyond the experimental training data set, hence aiding the final PI-ML model in generalizing to conditions not covered by its experimental training data set. Figure 1

The future 6G network will be able to connect massive and various unmanned vehicles (UxVs) into the network, bringing novel requirements toward UxV network security and data privacy. Deploying machine learning (ML)-based intrusion detection on UxVs can be one promising approach. The conventional approach cannot fulfill security and privacy requirements, as the structure can be more flexible, and there may be a lack of a central node in the 6G-based UxV network. Moreover, UxV clients can exhibit high heterogeneity with extremely class-imbalanced training samples. With the presence of federated learning (FL), UxVs can collaborate to train and update ML models while preserving data privacy in the UxV security critical scenario. However, most past works focused on centralized approaches, using cloud servers to assist intrusion detection and service deployments. This article introduces an FL framework with a decentralized design for training ML models on UxVs using intrusion detection as a study case. This scheme can be more flexible, as UxV clients can collaborate to train and synchronize models without the central server. Simulation experiments were conducted to evaluate the efficiency of the proposed approach. Experimental results and theoretical analysis confirm that the proposed method outperforms baseline models with FedAvg and local-trained client models.

Accurate estimation of reference evapotranspiration (ETo) plays a vital role in irrigation and water resource planning. The Penman–Monteith method recommended by the Food and Agriculture Organization (FAO PM56) is widely used and considered a standard to calculate ETo. However, FAO PM56 cannot be used with limited meteorological variables, so it is compulsory to choose an alternative model for ETo estimation, which requires fewer variables. This study built ten machine learning (ML) models based on multi-function, neural network, and tree-based structure against the FAO PM56 method. For this purpose, time series temperature data on a monthly scale are only used to train ML models. The developed ML models were applied to estimate ETo at different test stations and the obtained results were compared with the FAO PM56 method to verify and validate their performance in ETo estimation for the selected stations. In addition, multiple statistical indicators, including root-mean-square error (RMSE), coefficient of determination (R2), mean absolute error (MAE), Nash–Sutcliffe efficiency (NSE), and correlation coefficient (r) were calculated to compare the performance of each ML model on ETo estimation. Among the applied ML models, the ETo tree boost (TB) ML model outperformed the other ML models in estimating ETo in diverse climatic conditions based on statistical indicators (R2, NSE, r, RMSE, and MAE). Moreover, the observed R2, NSE, and r were the highest for the TB ML model, while RMSE and MAE were found to be the lowest at the study sites compared to other applied ML models. Lastly, ETo point data yielded from the TB ML model was used in an interpolation process to create monthly and annual ETo maps. Based on the ETo maps, this study suggests mainly a focus on areas with high ETo values and proper irrigation scheduling of crops to ensure water sustainability.

The rapid adoption of DevOps practices has transformed the software development landscape by emphasizing continuous integration, continuous delivery (CI/CD), and agile methodologies. However, this rapid pace of development often introduces significant security challenges, as traditional security testing methods struggle to keep up with the accelerated release cycles. To address these challenges, the integration of Artificial Intelligence (AI) and Machine Learning (ML) into automated security testing has emerged as a promising solution. This paper explores the use of AI and ML to enhance automated security testing within DevOps environments, offering a comprehensive approach to identifying, predicting, and mitigating security vulnerabilities in real time. Automated security testing leverages AI and ML algorithms to analyze code, detect anomalies, and predict potential security threats. These technologies enable the continuous monitoring of codebases, allowing for the early identification of vulnerabilities before they are exploited. By incorporating AI-driven security testing into the CI/CD pipeline, organizations can ensure that security is not an afterthought but a continuous process integrated into every stage of the software development lifecycle. AI and ML models can be trained to recognize patterns associated with security risks, such as code injection, unauthorized access, and data leakage. These models continuously learn from new data, improving their accuracy over time and adapting to evolving threats. The dynamic nature of AI-driven security testing makes it particularly suited for DevOps environments, where frequent code changes and updates can introduce new vulnerabilities. Moreover, AI and ML can assist in automating complex tasks, such as threat modeling, risk assessment, and the prioritization of security issues, enabling security teams to focus on higher-order tasks that require human expertise.

The integration of Artificial Intelligence (AI) into financial forecasting has transformed traditional stock market prediction methods. This research paper explores the effectiveness of AI techniques in forecasting stock trends within the Indian stock market over the last few years. We examine AI methodologies, including machine learning (ML), deep learning (DL), and hybrid models applied to the Bombay Stock Exchange (BSE) and the National Stock Exchange (NSE). By comparing historical data with AI-predicted trends, this study evaluates prediction accuracy and market relevance (Patel et al., 2015). Furthermore, the research outlines existing study gaps and proposes a future scope of integrating AI with behavioral finance and real-time analytics (Chen et al., 2022). The prediction of stock market trends remains a significant challenge due to the stochastic and non-linear nature of financial time series data (Zhang &amp; Zhou, 2020). With the proliferation of Artificial Intelligence (AI), particularly Machine Learning (ML) and Deep Learning (DL) techniques, there has been a paradigm shift in the modeling and forecasting of stock price movements (Fischer &amp; Krauss, 2018). This paper presents a comprehensive study on the effectiveness of AI in predicting stock market trends within the Indian financial ecosystem, focusing on a comparative analysis of AI models implemented over the last five years (2018–2023) on major indices and stocks listed on the NSE and BSE. This research evaluates the performance of multiple AI algorithms—including Support Vector Machines (SVM), Random Forests (RF), Long Short-Term Memory (LSTM) networks, Gated Recurrent Units (GRU), and hybrid ensemble models—in forecasting short- and medium-term price trends using historical stock data (Krauss et al., 2017; Chen &amp; He, 2021). Standard evaluation metrics such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and classification accuracy are employed to assess model efficacy. The results indicate that while traditional machine learning algorithms offer moderate predictive power, advanced deep learning and hybrid models significantly outperform them, particularly during periods of high market volatility, such as the COVID-19 pandemic and post-lockdown recovery phases (Weng et al., 2018; Jain &amp; Jain, 2021). The study identifies key research gaps, including limited sectoral diversification in datasets, inadequate integration of sentiment and behavioral data, overfitting issues in complex models, and the lack of real-time prediction systems in the Indian context (Gupta &amp; Pathak, 2022). Furthermore, current models often ignore nonquantitative factors such as investor sentiment, macroeconomic indicators, and global events, which can critically impact prediction accuracy (Nassirtoussi et al., 2014). These limitations suggest the need for a more holistic and interdisciplinary approach to AI-driven financial forecasting. The scope for future research includes the development of real-time, adaptive AI systems using high-frequency trading data, the incorporation of behavioral finance through social media and news analytics, and the exploration of quantum computing-based AI models (Liu et al., 2023). From a practical standpoint, the findings of this study offer valuable insights for institutional investors, financial analysts, regulatory bodies, and developers of AIpowered trading platforms. The study concludes that while AI is not a definitive solution for market prediction, it provides a powerful augmentative tool that, when designed with robustness, transparency, and adaptability, can significantly enhance decision-making in India’s fastevolving financial markets.

Digitalisation of health and the use of health data in artificial intelligence, and machine learning (ML), including for applications that will then in turn be used in healthcare are major...

Reconstructing the emergence of the modern continental crust is crucial to understand the evolution of the crust, the onset of plate tectonics, elemental cycling, and long-term climate. However, it remains highly contentious about when and how the major subaerial continental crust emerged over time. Here, we used a machine learning (ML) model (XGBoost) to reveal the exposed history of continental crust over the last 3.8 billion years ago (Ga). First, we compiled ~10,000 modern subaerial or submarine basalts with major and trace elements to train the ML model. Then, the trained ML model (with resampling) was utilized to predict and calculate the mean proportions of subaerially erupted continental basaltic rocks since 3.8 Ga. The result suggested that the subaerial proportion only reached about 50% at ~2.5 Ga, indicating the exposure of the continental crust was far from the present-day level at the end of the Archean era. On the other hand, since ~1.8 Ga, the subaerial proportion of the continental crust exhibited a dynamic balance between ~60% and 80%, reaching the present-day level.

Federated learning enables privacy-preserving and data-efficient dimension prediction and part qualification across additive manufacturing factories

Real-World Surveillance of FDA-Cleared Artificial Intelligence Models: Rationale and Logistics.