Potential Of Neural Networks Research Articles

Assessing the quality of random number generators through neural networks

In this paper we address the use of Neural Networks (NNs) for the assessment of the quality and hence safety of several Random Number Generators (RNGs), focusing both on the vulnerability of classical Pseudo Random Number Generators (PRNGs), such as Linear Congruential Generators (LCGs) and the RC4 algorithm, and extending our analysis to non-conventional data sources, such as Quantum Random Number Generators (QRNGs) based on Vertical-Cavity Surface-Emitting Laser (VCSEL). Among the results found, we have classified the generators based on the capability of the NN to distinguish between the RNG and a Golden Standard RNG (GSRNG). We show that sequences from simple PRNGs like LCGs and RC4 can be distinguished from the GSRNG. We also show that sequences from LCG on elliptic curves and VCSEL-based QRNG can not be distinguished from the GSRNG even with the biggest long-short term memory or convolutional neural networks (CNNs) that we have considered. We underline the fundamental role of design decisions in enhancing the safety of RNGs. The influence of network architecture design and associated hyper-parameters variations was also explored. We show that longer sequence lengths and CNNs are more effective for discriminating RNGs against the GSRNG. Moreover, in the prediction domain, the proposed model is able to deftly distinguish between the raw data of our QRNG and data from the GSRNG exhibiting a cross-entropy error of 0.52 on the test data-set used. All these findings reveal the potential of NNs to enhance the security of RNGs, while highlighting the robustness of certain QRNGs, in particular the VCSEL-based variants, for high-quality random number generation applications.

Superdiffusive Rotation of Interfacial Water on Noble Metal Surface.

Interfacial water on a metal surface acts as an active layer through the reorientation of water, thereby facilitating the energy transfer and chemical reaction across the metal surface in various physicochemical and industrial processes. However, how this active interfacial water collectively behaves on flat noble metal substrates remains largely unknown due to the experimental limitation in capturing librational vibrational motion of interfacial water and prohibitive computational costs at the first-principles level. Herein, by implementing a machine-learning approach to train neural network potentials, we enable performing advanced molecular dynamics simulations with ab initio accuracy at a nanosecond scale to map the distinct rotational motion of water molecules on a metal surface at room temperature. The vibrational density of states of the interfacial water with two-layer profiles reveals that the rotation and vibration of water within the strong adsorption layer on the metal surface behave as if the water molecules in the bulk ice, wherein the O-H stretching frequency is well consistent with the experimental results. Unexpectedly, the water molecules within the adjacent weak adsorption layer exhibit superdiffusive rotation, contrary to the conventional diffusive rotation of bulk water, while the vibrational motion maintains the characteristic of bulk water. The mechanism underlying this abnormal superdiffusive rotation is attributed to the translation-rotation decoupling of water, in which the translation is restrained by the strong hydrogen bonding within the bilayer interfacial water, whereas the rotation is accelerated freely by the asymmetric water environment. This superdiffusive rotation dynamics may elucidate the experimentally observed large fluctuation of the potential of zero charge on Pt and thereby the conventional Helmholtz layer model revised by including the contribution of interfacial water orientation. The surprising superdiffusive rotation of vicinal water next to noble metals will shed new light on the physicochemical processes and the activity of water molecules near metal electrodes or catalysts.

Prebiotic chemical reactivity in solution with quantum accuracy and microsecond sampling using neural network potentials

While RNA appears as a good candidate for the first autocatalytic systems preceding the emergence of modern life, the synthesis of RNA oligonucleotides without enzymes remains challenging. Because the uncatalyzed reaction is extremely slow, experimental studies bring limited and indirect information on the reaction mechanism, the nature of which remains debated. Here, we develop neural network potentials (NNPs) to study the phosphoester bond formation in water. While NNPs are becoming routinely applied to nonreactive systems or simple reactions, we demonstrate how they can systematically be trained to explore the reaction phase space for complex reactions involving several proton transfers and exchanges of heavy atoms. We then propagate at moderate computational cost hundreds of nanoseconds of a variety of enhanced sampling simulations with quantum accuracy in explicit solvent conditions. The thermodynamically preferred reaction pathway is a concerted, dissociative mechanism, with the transient formation of a metaphosphate transition state and direct participation of water solvent molecules that facilitate the exchange of protons through the nonbridging phosphate oxygens. Associative-dissociative pathways, characterized by a much tighter pentacoordinated phosphate, are higher in free energy. Our simulations also suggest that diprotonated phosphate, whose reactivity is never directly assessed in the experiments, is significantly less reactive than the monoprotonated species, suggesting that it is probably never the reactive species in normal pH conditions. These observations rationalize unexplained experimental results and the temperature dependence of the reaction rate, and they pave the way for the design of more efficient abiotic catalysts and activating groups.

Correlation-Pattern-Based Continuous Variable Entanglement Detection through Neural Networks.

Entanglement in continuous-variable non-Gaussian states provides irreplaceable advantages in many quantum information tasks. However, the sheer amount of information in such states grows exponentially and makes a full characterization impossible. Here, we develop a neural network that allows us to use correlation patterns to effectively detect continuous-variable entanglement through homodyne detection. Using a recently defined stellar hierarchy to rank the states used for training, our algorithm works not only on any kind of Gaussian state but also on a whole class of experimentally achievable non-Gaussian states, including photon-subtracted states. With the same limited amount of data, our method provides higher accuracy than usual methods to detect entanglement based on maximum-likelihood tomography. Moreover, in order to visualize the effect of the neural network, we employ a dimension reduction algorithm on the patterns. This shows that a clear boundary appears between the entangled states and others after the neural network processing. In addition, these techniques allow us to compare different entanglement witnesses and understand their working. Our findings provide a new approach for experimental detection of continuous-variable quantum correlations without resorting to a full tomography of the state and confirm the exciting potential of neural networks in quantum information processing.

Physics-aware recurrent convolutional neural networks for modeling multiphase compressible flows

Multiphase compressible flow systems can exhibit unsteady and fast-transient dynamics, marked by sharp gradients and discontinuities, and material boundaries that interact with the evolving flow. The transient nature of the dynamics presents challenges to employing artificial intelligence (AI) and data-driven models for predicting flow behaviors. In this study, we explore the potential of physics-aware recurrent convolutional neural networks (PARC) to model the spatiotemporal dynamics of multiphase flows in the presence of shocks and reaction fronts. PARC is a neural network model that incorporates the generic form of the diffusion–advection–reaction equation in its network architecture, which mimics the process of solving the governing equations of fluid flows. In contrast to physics-informed machine learning approaches such as physics-informed neural networks (PINNs) where models are trained to directly minimize the residual of governing equations, PARC takes a dynamical systems viewpoint and does not seek to minimize potentially nonconvex and nonlinear loss terms. To assess the ability of PARC to accurately learn and simulate the physics of multiphase flows, we train and test PARC on various flow simulation problems, including the Burgers’ equation, fluid flow behind a cylindrical cross-section, and unsteady shock interactions with a particle at varying Mach numbers. We analyze PARC’s performance and examine sources of error in its prediction, in terms of differentiation and integration schemes and different weighting strategies for the model update. Based on our observations, we discuss PARC’s capabilities and limitations in multiphase flow applications and propose future research directions.

Assessment of Embedding Schemes in a Hybrid Machine Learning/Classical Potentials (ML/MM) Approach.

Machine learning (ML) methods have reached high accuracy levels for the prediction of in vacuo molecular properties. However, the simulation of large systems solely through ML methods (such as those based on neural network potentials) is still a challenge. In this context, one of the most promising frameworks for integrating ML schemes in the simulation of complex molecular systems are the so-called ML/MM methods. These multiscale approaches combine ML methods with classical force fields (MM), in the same spirit as the successful hybrid quantum mechanics-molecular mechanics methods (QM/MM). The key issue for such ML/MM methods is an adequate description of the coupling between the region of the system described by ML and the region described at the MM level. In the context of QM/MM schemes, the main ingredient of the interaction is electrostatic, and the state of the art is the so-called electrostatic-embedding. In this study, we analyze the quality of simpler mechanical embedding-based approaches, specifically focusing on their application within a ML/MM framework utilizing atomic partial charges derived in vacuo. Taking as reference electrostatic embedding calculations performed at a QM(DFT)/MM level, we explore different atomic charges schemes, as well as a polarization correction computed using atomic polarizabilites. Our benchmark data set comprises a set of about 80k small organic structures from the ANI-1x and ANI-2x databases, solvated in water. The results suggest that the minimal basis iterative stockholder (MBIS) atomic charges yield the best agreement with the reference coupling energy. Remarkable enhancements are achieved by including a simple polarization correction.

On-the-fly kinetic Monte Carlo simulations with neural network potentials for surface diffusion and reaction.

We develop an adaptive scheme in the kinetic Monte Carlo simulations, where the adsorption and activation energies of all elementary steps, including the effects of other adsorbates, are evaluated "on-the-fly" by employing the neural network potentials. The configurations and energies evaluated during the simulations are stored for reuse when the same configurations are sampled in a later step. The present scheme is applied to hydrogen adsorption and diffusion on the Pd(111) and Pt(111) surfaces and the CO oxidation reaction on the Pt(111) surface. The effects of interactions between adsorbates, i.e., adsorbate-adsorbate lateral interactions, are examined in detail by comparing the simulations without considering lateral interactions. This study demonstrates the importance of lateral interactions in surface diffusion and reactions and the potential of our scheme for applications in a wide variety of heterogeneous catalytic reactions.

Constructing and Evaluating Machine-Learned Interatomic Potentials for Li-Based Disordered Rocksalts.

Lithium-based disordered rocksalts (LDRs), which are an important class of positive electrode materials that can increase the energy density of current Li-ion batteries, represent a significantly complex chemical and configurational space for conventional density functional theory (DFT)-based high-throughput screening approaches. Notably, atom-centered machine-learned interatomic potentials (MLIPs) are a promising pathway to accurately model the potential energy surface of highly disordered chemical spaces, such as LDRs, where the performance of such MLIPs has not been rigorously explored yet. Here, we represent a comprehensive evaluation of the accuracy, transferability, and ease of training of five atom-centered MLIPs, including the artificial neural network potentials developed by the atomic energy network (AENET), the Gaussian approximation potential (GAP), the spectral neighbor analysis potential (SNAP) and its quadratic extension (qSNAP), and the moment tensor potential (MTP), in modeling a 11-component LDR chemical space. Specifically, we generate a DFT-calculated data set of 10,842 configurations of disordered LiTMO2 and TMO2 compositions, where TM = Sc, Ti, V, Cr, Mn, Fe, Co, Ni, and/or Cu. To provide a point-of-comparison on the performance of atom-centered MLIPs, we also trained the neural equivariant interatomic potential (NequIP) on a subset of our data. Importantly, we find AENET to be the best potential in terms of accuracy and transferability for energy predictions, while MTP is the best for atomic forces. While AENET is the fastest to train among the MLIPs considered at low number of epochs (300), the training time increases significantly as epochs increase (3300), with a corresponding reduction in training errors (∼60%). Note that AENET and GAP tend to overfit in small data sets, with the extent of overfitting reducing with larger data sets. Finally, we observe AENET to provide reasonable predictions of average Li-intercalation voltages in layered, single-TM LiTMO2 frameworks, compared to DFT (∼10% error on average). Our study should pave the way both for discovering novel disordered rocksalt electrodes and for modeling other configurationally complex systems, such as high-entropy ceramics and alloys.

Songs Classification Problem Research by Genre Based on Neural Network

This article explores the use of neural networks for classifying musical compositions by genre based on their audio characteristics. With the increasing availability of large musical datasets and advancements in computational technologies, this topic has gained significant relevance. The article emphasizes the importance of selecting and processing audio features, particularly focusing on Mel-Frequency Cepstral Coefficients (MFCC), which reflect the key sound characteristics essential for genre identification. The approach to creating a neural network capable of effectively processing these characteristics and classifying songs is examined within the research. It outlines the main challenges associated with classification accuracy, error calculation, and the need to adapt to rapidly changing musical trends. This article also reviews and constructs a test neural network for classifying songs into three genres, creates diagrams to depict classification accuracy, and shows the distribution of each class by frequency characteristics. Additionally, it analyzes recent research in this field and highlights the importance of using MFCC. The article illuminates not only the technical aspects of creating and training neural networks for song classification but also the significance of such systems in the context of the widespread availability of musical content. It underscores the role of these systems in enhancing the quality of recommendation services, providing a more comprehensive musical experience to users, and supporting artists and musical styles. In conclusion, the article contributes valuable insights into the potential of neural networks in song genre classification, pointing to future research and development directions in this area.

High-throughput investigation of stability and Li diffusion of doped solid electrolytes via neural network potential without configurational knowledge

Solid electrolytes hold substantial promise as vital components of all-solid-state batteries. Enhancing their performance necessitates simultaneous improvements in their stability and lithium conductivity. These properties can be calculated using first-principles simulations, provided that the crystal structure of the material and the diffusion pathway through the material are known. However, solid electrolytes typically incorporate dopants to enhance their properties, necessitating the optimization of the dopant configuration for the simulations. Yet, performing such calculations via the first-principles approach is so costly that existing approaches usually rely on predetermined dopant configurations informed by existing knowledge or are limited to systems doped with only a few atoms. The proposed method enables the optimization of the dopant configuration with the support of neural network potential (NNP). Our approach entails the use of molecular dynamics to analyze the diffusion after the optimization of the dopant configuration. The application of our approach to Li10\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${_{10}}$$\\end{document}MP2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${_{2}}$$\\end{document}S12-x\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${_{12-x}}$$\\end{document}Ox\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${_{x}}$$\\end{document} (M = Ge, Si, or Sn) reproduce the experimental results well. Furthermore, analysis of the lithium diffusion pathways suggests that the activation energy of diffusion undergoes a percolation transition. This study demonstrates the effectiveness of NNPs in the systematic exploration of solid electrolytes.

An overview about neural networks potentials in molecular dynamics simulation

AbstractAb‐initio molecular dynamics (AIMD) is a key method for realistic simulation of complex atomistic systems and processes in nanoscale. In AIMD, finite‐temperature dynamical trajectories are generated by using forces computed from electronic structure calculations. In systems with high numbers of components a typical AIMD run is computationally demanding. On the other hand, machine learning (ML) is a subfield of the artificial intelligence that consist in a set of algorithms that show learning by experience with the use of input and output data where algorithms are capable of analysing and predicting the future. At present, the main application of ML techniques in atomic simulations is the development of new interatomic potentials to correctly describe the potential energy surfaces (PES). This technique is in constant progress since its inception around 30 years ago. The ML potentials combine the advantages of classical and Ab‐initio methods, that is, the efficiency of a simple functional form and the accuracy of first principles calculations. In this article we review the evolution of four generations of machine learning potentials and some of their most notable applications. This review focuses on MLPs based on neural networks. Also, we present a state of art of this topic and future trends. Finally, we report the results of a scientometric study (covering the period 1995–2023) about the impact of ML techniques applied to atomistic simulations, distribution of publications by geographical regions and hot topics investigated in the literature.

A reactive neural network framework for water-loaded acidic zeolites

Under operating conditions, the dynamics of water and ions confined within protonic aluminosilicate zeolite micropores are responsible for many of their properties, including hydrothermal stability, acidity and catalytic activity. However, due to high computational cost, operando studies of acidic zeolites are currently rare and limited to specific cases and simplified models. In this work, we have developed a reactive neural network potential (NNP) attempting to cover the entire class of acidic zeolites, including the full range of experimentally relevant water concentrations and Si/Al ratios. This NNP has the potential to dramatically improve sampling, retaining the (meta)GGA DFT level accuracy, with the capacity for discovery of new chemistry, such as collective defect formation mechanisms at the zeolite surface. Furthermore, we exemplify how the NNP can be used as a basis for further extensions/improvements which include data-efficient adoption of higher-level (hybrid) references via Δ-learning and the acceleration of rare event sampling via automatic construction of collective variables. These developments represent a significant step towards accurate simulations of realistic catalysts under operando conditions.

Feature selection for high-dimensional neural network potentials with the adaptive group lasso

Neural network potentials are a powerful tool for atomistic simulations, allowing to accurately reproduce ab initio potential energy surfaces with computational performance approaching classical force fields. A central component of such potentials is the transformation of atomic positions into a set of atomic features in a most efficient and informative way. In this work, a feature selection method is introduced for high dimensional neural network potentials, based on the adaptive group lasso (AGL) approach. It is shown that the use of an embedded method, taking into account the interplay between features and their action in the estimator, is necessary to optimize the number of features. The method’s efficiency is tested on three different monoatomic systems, including Lennard–Jones as a simple test case, Aluminium as a system characterized by predominantly radial interactions, and Boron as representative of a system with strongly directional components in the interactions. The AGL is compared with unsupervised filter methods and found to perform consistently better in reducing the number of features needed to reproduce the reference simulation data at a similar level of accuracy as the starting feature set. In particular, our results show the importance of taking into account model predictions in feature selection for interatomic potentials.

Amplifying Generative AI's Impact on Creative Content: Maximizing Neural Network Potential

In a world increasingly driven by technology, the field of creative content production stands on the initiative of a revolution. This paper explores the cutting-edge advancements in Generative Artificial Intelligence (GAI) & its profound impact on creative content creation. By leveraging the power of neural networks, GAI has the potential to transform how we generate, consume, and interact with creative works across various mediums. The paper begins by explaining the core principles of GAI, simplifying complex concepts for readers. It explores the fascinating world of neural networks, demonstrating how these intricate systems learn and replicate creative patterns and styles similar to human thinking. But the fascination doesn't end there. As the story progresses, the paper reveals the diverse benefits of GAI for creative professionals and enthusiasts alike. From transforming content creation processes to enabling personalized experiences and simplifying automation, the possibilities seem limitless. However, every technological wonder comes with its own set of challenges and considerations. This paper navigates through the ethical maze surrounding GAI, addressing concerns such as prejudices, copyright issues, and the intricate balance between human and artificial creativity. It advocates for responsible and moral practices in the use of GAI, ensuring that this potent tool is employed for the betterment of society. Yet, the most compelling feature lies in the introduction of a groundbreaking innovation: the "Investigate" algorithm. This algorithm, a culmination of advanced techniques including reinforcement learning and evolutionary optimization, promises to enhance GAI's impact by expanding the boundaries of neural network potential even further. It intrigues readers with the promising potential of a future where AI continuously improves itself, leading to the creation of increasingly sophisticated and innovative creative content. In conclusion, this paper calls upon readers to embark on a journey of discovery and innovation, inviting them to explore further into the realm of GAI's impact on creative content production. Through the examination of neural network potential and the moral considerations that come with it, we stand on the threshold of a new era in creativity and innovation.

A physics-inspired neural network for short-wave radiation parameterization

Abstract Radiation parameterization schemes are crucial components of weather and climate models, but they are also known to be computationally intensive. In recent decades, researchers have attempted to emulate these schemes using neural networks, with more attention to convolutional neural networks. However, in this paper, we explore the potential of recurrent neural networks (RNNs) for predicting solar heating rates. Our architecture was trained and tested using long-term hindcast data from the Pechora Sea region, with the conventional RRTMG scheme serving as a shortwave parameterization. Our findings show that the RNN offers rapid learning, fast inference, and excellent data fitting. We also present preliminary results demonstrating the use of RNNs for operational weather forecasting, which achieved a significant speedup in parameterization and reduced the overall forecast time by 40 %.

Detecting Parkinson’s disease from shoe-mounted accelerometer sensors using convolutional neural networks optimized with modified metaheuristics

Neurodegenerative conditions significantly impact patient quality of life. Many conditions do not have a cure, but with appropriate and timely treatment the advance of the disease could be diminished. However, many patients only seek a diagnosis once the condition progresses to a point at which the quality of life is significantly impacted. Effective non-invasive and readily accessible methods for early diagnosis can considerably enhance the quality of life of patients affected by neurodegenerative conditions. This work explores the potential of convolutional neural networks (CNNs) for patient gain freezing associated with Parkinson’s disease. Sensor data collected from wearable gyroscopes located at the sole of the patient’s shoe record walking patterns. These patterns are further analyzed using convolutional networks to accurately detect abnormal walking patterns. The suggested method is assessed on a public real-world dataset collected from parents affected by Parkinson’s as well as individuals from a control group. To improve the accuracy of the classification, an altered variant of the recent crayfish optimization algorithm is introduced and compared to contemporary optimization metaheuristics. Our findings reveal that the modified algorithm (MSCHO) significantly outperforms other methods in accuracy, demonstrated by low error rates and high Cohen’s Kappa, precision, sensitivity, and F1-measures across three datasets. These results suggest the potential of CNNs, combined with advanced optimization techniques, for early, non-invasive diagnosis of neurodegenerative conditions, offering a path to improve patient quality of life.

Atomistic modeling of nucleation kinetics of Guinier–Preston zones in Al–Cu alloys: Two formation scenarios and prediction of the time-temperature-transformation diagram

Metastable precipitates of solute atoms, known as Guinier–Preston (GP) zones, play a significant role in the age hardening of Al–Cu alloys. In this study, we utilized the classical nucleation theory (CNT) to obtain time-temperature-transformation (TTT) diagrams for steady-state nucleation over a wide temperature range. The CNT model, which is based on atomistic simulations using artificial neural network potentials, was constructed to predict the nucleation kinetics of GP zones in the Al–Cu system with solute concentrations ranging from 1.3 to 2.5 at% (3.0 to 5.7 wt%). The nose temperature at which the steady-state nucleation time for GP zone formation is the shortest was calculated. The nose temperature increased and the minimum nucleation time decreased as the solute concentration increased. These predictions are comparable to previous theoretical results and are in good qualitative agreement with experimental observations. Furthermore, two formation scenarios of double-layer GP (GP2) zones, considering synchronous and asynchronous attachments of solute atoms to clusters, were compared in terms of nucleation efficiency. This provides new insights into the nucleation pathways of the GP2 zone in Al–Cu alloys.

Knowledge graph learning algorithm based on deep convolutional networks

Knowledge graphs (KGs) serve as invaluable tools for organizing and representing structural information, enabling powerful data analysis and retrieval. In this paper, we propose a novel knowledge graph learning algorithm based on deep convolutional neural networks (KGLA-DCNN) to enhance the classification accuracy of KG nodes. Leveraging the hierarchical and relational nature of KGs, our algorithm utilizes deep convolutional neural networks to capture intricate patterns and dependencies within the graph. We evaluate the effectiveness of KGLA-DCNN on two benchmark datasets, Cora and Citeseer, renowned for their challenging node classification tasks. Through extensive experiments, we demonstrate that our proposed algorithm significantly improves classification accuracy compared to state-of-the-art methods, showcasing its capability to leverage the rich structural information inherent in KGs. The results highlight the potential of deep convolutional neural networks in enhancing the learning and representation capabilities of knowledge graphs, paving the way for more accurate and efficient knowledge discovery in diverse domains.

AI-based diagnosis of chronic obstructive pulmonary disease from low-dose CT images

Background: Chronic obstructive pulmonary disease (COPD) is a group of diseases characterized by airflow blockage. It is one of the leading causes of global mortality and is primarily attributed to smoking. COPD patients are usually diagnosed by spirometry test. Although regarded as the gold standard for COPD diagnosis, the spirometry test carries contraindications, thus prompting the development of low-dose computed tomography (low-dose CT) scan as an alternative for COPD screening. However, a practical limitation of diagnosing COPD from CT images is its reliance on the expertise of a skilled radiologist. Objective: To address this limitation, we aimed to develop a deep-learning model for the automated classification of COPD and non-COPD from low-dose CT images. Materials and methods: We examined the potential of a convolutional neural network for identifying COPD. Our dataset consisted of 10,000 low-dose CT images obtained from a lung cancer screening program, involving both ex-smokers and current smokers deemed at high risk of lung cancer. Spirometry data served as the ground truth for defining COPD. We used 90% of the datasets for training and 10% for testing. Results: Our developed model achieved notable performance metrics: an area under the receiver operating characteristic curve (AUC) of 0.97, an accuracy of 0.89, a precision of 0.85, a recall of 0.96, and an F1-score of 0.90. Conclusion: Our study demonstrates the potential of deep learning models to augment clinical assessments and improve the diagnosis of COPD, thereby enhancing diagnostic accuracy and efficiency. The findings suggest the feasibility of integrating this technology into routine lung cancer screening programs for COPD detection.

Ab initio informed machine learning potential for tribochemistry and mechanochemistry: Application for eco–friendly gallate lubricant additive

Mechanochemistry and tribochemistry processes involve multiple physical/chemical interactions induced by extreme conditions including molecular confinement, high temperatures and mechanical stress applied. Simulating these processes by molecular dynamics is very challenging. While force fields fall short reproducing the enhanced reactivity arising by quantum effects, ab initio molecular dynamics is severely limited by the complexity of the systems of interest, their sizes, and the long-time scale on which relevant events take place. In this work using an active learning approach, a landmark deep neural network potential has been developed which reproduces the accuracy of ab initio interactions at the classical molecular dynamics computational cost and permits to successfully simulate the tribochemical processes occurring at the interface between butyl gallate molecules and iron substrates under tribological conditions. The simulations of the dynamics of the Fe-gallate system when sliding under an imposed external load reveal the key atomistic mechanisms underlying the formation of the friction reducing lubricant tribofilm and permit to characterize the tribological properties of the explored systems, clearly exposing the shortcoming of reactive force field based approaches. The successful development of neural network potentials, making it possible to push the limits of molecular dynamics marrying accuracy with system sizes and long time scales, paves the route toward a new area in computational tribochemistry.