Investigating the Unfavored Factors That Interfere MALDI-TOF Based AI in Predicting Antibiotic Resistance

Combining Matrix-Assisted Laser Desorption/Ionization Time-of-Flight (MALDI-TOF) spectra data and artificial intelligence (AI) has been introduced for rapid prediction on antibiotic susceptibility test (AST) of S. aureus. Based on the AI predictive probability, the cases with probabilities between low and high cut-offs are defined as “grey zone”. We aimed to investigate the underlying reasons of unconfident (grey zone) or wrong predictive AST. A total 479 S. aureus isolates were collected, analyzed by MALDI-TOF, and AST prediction, standard AST were obtained in a tertiary medical center. The predictions were categorized into the correct prediction group, wrong prediction group, and grey zone group. We analyzed the association between the predictive results and the demographic data, spectral data, and strain types. For MRSA, larger cefoxitin zone size was found in the wrong prediction group. MLST of the MRSA isolates in the grey zone group revealed that uncommon strain types composed 80%. Amid MSSA isolates in the grey zone group, the majority (60%) was composed of over 10 different strain types. In predicting AST based on MALDI-TOF AI, uncommon strains and high diversity would contribute to suboptimal predictive performance.

LIME-Based Data Selection Method for SAR Images Generation Using GAN

Deep learning has obtained remarkable achievements in computer vision, especially image and video processing. However, in synthetic aperture radar (SAR) image recognition, the application of DNNs is usually restricted due to data insufficiency. To augment datasets, generative adversarial networks (GANs) are usually used to generate numerous photo-realistic SAR images. Although there are many pixel-level metrics to measure GAN’s performance from the quality of generated SAR images, there are few measurements to evaluate whether the generated SAR images include the most representative features of the target. In this case, the classifier probably categorizes a SAR image into the corresponding class based on “wrong” criterion, i.e., “Clever Hans”. In this paper, local interpretable model-agnostic explanation (LIME) is innovatively utilized to evaluate whether a generated SAR image possessed the most representative features of a specific kind of target. Firstly, LIME is used to visualize positive contributions of the input SAR image to the correct prediction of the classifier. Subsequently, these representative SAR images can be selected handily by evaluating how much the positive contribution region matches the target. Experimental results demonstrate that the proposed method can ally “Clever Hans” phenomenon greatly caused by the spurious relationship between generated SAR images and the corresponding classes.

Iterative Named Entity Recognition with Conditional Random Fields

Named entity recognition (NER) constitutes an important step in the processing of unstructured text content for the extraction of information as well as for the computer-supported analysis of large amounts of digital data via machine learning methods. However, NER often relies on domain-specific knowledge, being conducted manually in a time- and human-resource-intensive process. These can be reduced with statistical models performing NER automatically. The current work investigates whether Conditional Random Fields (CRF) can be efficiently trained for NER in German texts, by means of an iterative procedure combining self-learning with a manual annotation–active learning–component. The training dataset increases continuously with the iterative procedure. Whilst self-learning did not markedly improve the performance of the CRF for NER, the manual annotation of sentences with the lowest probability of correct prediction clearly improved the model F1-score and simultaneously reduced the amount of manual annotation required to train the model. A model with an F1-score of 0.885 was able to be trained in 11.4 h.

Decision support for preventing safety violations

Aim. The aim of the paper is to examine the experience of reducing the effect of the human factor on business processes, to develop the structure and software of the decisionsupport system for preventing safety violations by train drivers using machine learning and to analyse the findings. Methods. The study presented in the paper uses machine learning, statistical analysis and expert analysis. In terms of machine learning, the following methods were used: logistical regression, random forests, gradient boosting over decision trees with frequency-domain representation of categorical features, neural networks. Results. A set of indicators characterizing a train driver’s operation were identified and are to be used as part of the system under development. The term “train driver’s reliability” was defined as the ability not to violate train traffic safety over a certain number of trips. Algorithms were designed and examined for predicting violations in a train driver’s operation that are used in defining reliability groups and lists of preventive measures recommended for the reduction of the number of safety violations in a train driver’s operation. Major violations with proven guilt of the driver that may be committed within the following 3, 7, 10, 20, 30, 60 days were chosen as attributes for the purpose of safety violation prediction. Analysis of the results on the test sample revealed that the model based on gradient boosting over decision trees with frequency-domain representation of categorical features shows the best results for binary classification on the prediction horizon of 30 and 60 days. The developed algorithm made a correct prediction in 76% of cases with the threshold value of 0.7 and horizon of 30 days and in 82% of cases with the threshold value of 0.9 and horizon of 60 days. The solution of the problem can be found in the integration of different approaches to predicting safety violations in a train driver’s operation. Additionally, 10 of the most significant indicators of a train driver’s operation were identified with the best of the considered models, i.e., gradient boosting over decision trees with frequency-domain representation of categorical features. Conclusion. The paper presents an overview of methods and systems of assessing human reliability and the effect of the human factor on the safety of transportation systems. It allowed choosing the most promising directions and methods of predictive analysis of a train driver’s operation, including methods of machine learning. The resulting set of indicators of a train driver’s operation that take into consideration the changes in the quality of such operation allowed obtaining initial data for training the models implemented as part of the system under development. The implemented models enabled the aggregation of information on train drivers and adoption of targeted and temporary preventive measures recommended for improving driver reliability. The resulting approach to the definition of preventive measures has been implemented in three depots of JSC RZD in trial operation mode.

Computational Studies of Au(I) and Au(III) Anticancer MetalLodrugs: A Survey

Owing to the growing hardware capabilities and the enhancing efficacy of computational methodologies, computational chemistry approaches have constantly become more important in the development of novel anticancer metallodrugs. Besides traditional Pt-based drugs, inorganic and organometallic complexes of other transition metals are showing increasing potential in the treatment of cancer. Among them, Au(I)- and Au(III)-based compounds are promising candidates due to the strong affinity of Au(I) cations to cysteine and selenocysteine side chains of the protein residues and to Au(III) complexes being more labile and prone to the reduction to either Au(I) or Au(0) in the physiological milieu. A correct prediction of metal complexes’ properties and of their bonding interactions with potential ligands requires QM computations, usually at the ab initio or DFT level. However, MM, MD, and docking approaches can also give useful information on their binding site on large biomolecular targets, such as proteins or DNA, provided a careful parametrization of the metal force field is employed. In this review, we provide an overview of the recent computational studies of Au(I) and Au(III) antitumor compounds and of their interactions with biomolecular targets, such as sulfur- and selenium-containing enzymes, like glutathione reductases, glutathione peroxidase, glutathione-S-transferase, cysteine protease, thioredoxin reductase and poly (ADP-ribose) polymerase 1.

Carbonate Karstified Oil Fields Geological Prediction and Dynamic Simulation Through Equivalent Relative Permeability Curves

Karst systems heterogeneity may become a nightmare for reservoir modelers in predicting presence, spatial distribution, impact on formation petrophysical characteristics, and particularly in dynamic behaviour prediction. Moreover, the very high resolution required to describe in detail the phenomena does not reconcile with the geo-cellular model resolution typically used for reservoir simulation. The scope of the work is to present an effective approach to predict karst presence and model it dynamically. Karst presence recognition started from the analysis of anomalous well behaviour and potential sources of precursors (logs, drilling evidence, etc.) to derive concepts for karst reservoir model. This first demanding step implies then characterizing each cell classified as karstified in terms of petrophysical parameters. In a two-phase flow, karst brings to fast travelling of water which leaves the matrix almost unswept. This feature was characterized through dedicated fine simulations, leading to an upscaling of relative permeability curves for a single porosity formulation. The workflow was applied to a carbonate giant field with a long production history under waterflood development. Firstly, a machine learning algorithm was trained to recognize karst features based on log response, seismic attributes, and well dynamic evidence, then a karst probability volume was generated and utilized to predict the karst presence in the field. Karst characterization just in terms of porosity and permeability is sufficient to model the reservoir when still in single phase, however it fails to reproduce observed water production. Karst provides a high permeability path for water transport: classical history match approaches, such as the introduction of permeability multipliers, proved to be ineffective in reproducing the water breakthrough timing and growth rate. In fact, the reservoir consists of two systems, matrix, and karst: however, the karst is less known and laboratory analysis shows relative permeability only for the matrix medium. The introduction of equivalent or pseudo-relative permeability curves, accounting for both the media, was crucial for correct modelling of the reservoir underlying dynamics, allowing a proper reproduction of water breakthrough timing and water cut (WCT) trends. The implementation of a dedicated pseudo relative permeability curve dedicated to karstified cells allowed to replicate early water arrival, thus bringing to a correct prediction of oil and water rates, also highlighting the presence of bypassed oil associated with water circuiting, particularly in presence of highly karstified cells.

Development of secondary vortex structures in rotor wakes

The blade tip vortex system is a crucial feature in the wake of helicopter rotors, and its correct prediction represents a major challenge in the numerical simulation of rotor flows. A common phenomenon in modern high-fidelity CFD simulations is the breakdown of the primary vortex system in hover due to secondary vortex braids. Since they are strongly influenced by the numerical settings, the degree to which these secondary vortex structures actually physically occur is still discussed and needs experimental validation. In the current work, the development of secondary vortex structures in the wake of a two-bladed rotor in hover conditions was investigated by combining stereoscopic particle image velocimetry measurements in different measurement planes and high-fidelity simulations. Secondary vortex structures were detected and quantified at different axial locations in the wake by applying an identical scheme to the measured and simulated velocity data. In agreement, it was found that the number of secondary vortices is maximum at a distance of $$0.8\,R$$ 0.8 R below the rotor. The more intense secondary vortex structures were quantitatively well captured in the simulation, whereas in the experiment a larger number of weaker vortices were detected. No distinct preferential direction of rotation was found for the secondary vortices, but they tended to develop in vortex pairs with alternating sense of rotation. A clustered occurrence of secondary vortices was observed close to the primary tip vortices, where the rolled-up blade shear layer breaks down into coherent vortex structures. Graphical abstract

Effect of drill pipe orbital motion on non-Newtonian fluid flow in an eccentric wellbore: a study with computational fluid dynamics

To ensure an effective drilling operation of an explored well, the associated hydraulics program should be established carefully based on the correct prediction of a drilling fluid’s pressure drop and velocity field. For that, the impact of the drill string orbital motion should be considered by drilling engineers since it has an important influence on the flow of drilling fluid and cuttings transport process. In the present investigation, the finite volume method coupled with the sliding mesh approach is used to analyze the influence of the inner cylinder orbital motion on the flow of a power-law fluid (Ostwald-de Waele) in an annular geometry. The findings indicate that the orbital motion positively affects the homogeneity of the power-law axial velocity through the entire eccentric annulus; however, this impact diminishes as the diameter ratio increases. In addition, higher torque is induced when the orbital motion occurs, especially for high values of eccentricity and diameter ratio; nonetheless, a slight decrease in torque is recorded when the fluid velocity increases.

An Explainable Machine Learning Model for Material Backorder Prediction in Inventory Management

Global competition among businesses imposes a more effective and low-cost supply chain allowing firms to provide products at a desired quality, quantity, and time, with lower production costs. The latter include holding cost, ordering cost, and backorder cost. Backorder occurs when a product is temporarily unavailable or out of stock and the customer places an order for future production and shipment. Therefore, stock unavailability and prolonged delays in product delivery will lead to additional production costs and unsatisfied customers, respectively. Thus, it is of high importance to develop models that will effectively predict the backorder rate in an inventory system with the aim of improving the effectiveness of the supply chain and, consequentially, the performance of the company. However, traditional approaches in the literature are based on stochastic approximation, without incorporating information from historical data. To this end, machine learning models should be employed for extracting knowledge of large historical data to develop predictive models. Therefore, to cover this need, in this study, the backorder prediction problem was addressed. Specifically, various machine learning models were compared for solving the binary classification problem of backorder prediction, followed by model calibration and a post-hoc explainability based on the SHAP model to identify and interpret the most important features that contribute to material backorder. The results showed that the RF, XGB, LGBM, and BB models reached an AUC score of 0.95, while the best-performing model was the LGBM model after calibration with the Isotonic Regression method. The explainability analysis showed that the inventory stock of a product, the volume of products that can be delivered, the imminent demand (sales), and the accurate prediction of the future demand can significantly contribute to the correct prediction of backorders.