Training Database Research Articles

Explainable machine learning for predicting neurological outcome in hemorrhagic and ischemic stroke patients in critical care.

The objective of this study is to develop accurate machine learning (ML) models for predicting the neurological status at hospital discharge of critically ill patients with hemorrhagic and ischemic stroke and identify the risk factors associated with the neurological outcome of stroke, thereby providing healthcare professionals with enhanced clinical decision-making guidance. Data of stroke patients were extracted from the eICU Collaborative Research Database (eICU-CRD) for training and testing sets and the Medical Information Mart for Intensive Care IV (MIMIC IV) database for external validation. Four machine learning models, namely gradient boosting classifier (GBC), logistic regression (LR), multi-layer perceptron (MLP), and random forest (RF), were used for prediction of neurological outcome. Furthermore, shapley additive explanations (SHAP) algorithm was applied to explain models visually. A total of 1,216 hemorrhagic stroke patients and 954 ischemic stroke patients from eICU-CRD and 921 hemorrhagic stroke patients 902 ischemic stroke patients from MIMIC IV were included in this study. In the hemorrhagic stroke cohort, the LR model achieved the highest area under curve (AUC) of 0.887 in the test cohort, while in the ischemic stroke cohort, the RF model demonstrated the best performance with an AUC of 0.867 in the test cohort. Further analysis of risk factors was conducted using SHAP analysis and the results of this study were converted into an online prediction tool. ML models are reliable tools for predicting hemorrhagic and ischemic stroke neurological outcome and have the potential to improve critical care of stroke patients. The summarized risk factors obtained from SHAP enable a more nuanced understanding of the reasoning behind prediction outcomes and the optimization of the treatment strategy.

Performance optimization of cement calciner based on CFD simulation and machine learning algorithm

In order to improve the combustion efficiency and decomposition rate of the cement calciner and reduce pollutant emission, a performance optimization method based on Computational Fluid Dynamics (CFD) numerical simulation integrated with machine learning is proposed. The Multiphase Particle-in-cell (MP-PIC) method and the chemical reaction models are employed to simulate the coal combustion and CaCO3 decomposition process, whose calculation results are combined with the industrial practical data of the cement plant, so as to establish a more comprehensive training database. On this basis, a novel Topology Particle Swarm Optimization algorithm integrating with Convolutional Neural Network and Long Short-Term Memory (RITPSO–CNN–LSTM) algorithm model is established to predict combustion efficiency, decomposition rate, and NOx emission, respectively. Results show that compared with two other relative basic algorithm models, the prediction error of the proposed algorithm model for the three targets is minimal with the average relative error of 0.045 %, 0.038 %, and 0.021 %, respectively. The addition of CFD simulation data makes the prediction model more applicable with higher stability and accuracy. Based on the prediction results, Grey Wolf Optimizer (GWO) algorithm is employed to optimize operating parameters, and finally the average optimization amount of combustion efficiency, decomposition rate, and NOx emission are 2.17 %, 2.24 %, and 6.15 ppm, respectively, which meet the optimization requirements.

Generative Convective Parametrization of a Dry Atmospheric Boundary Layer

AbstractTurbulence parametrizations will remain a necessary building block in kilometer‐scale Earth system models. In convective boundary layers, where the mean vertical gradients of conserved properties such as potential temperature and moisture are approximately zero, the standard ansatz which relates turbulent fluxes to mean vertical gradients via an eddy diffusivity has to be extended by mass‐flux parametrizations for the typically asymmetric up‐ and downdrafts in the atmospheric boundary layer. We present a parametrization for a dry and transiently growing convective boundary layer based on a generative adversarial network. The training and test data are obtained from three‐dimensional high‐resolution direct numerical simulations. The model incorporates the physics of self‐similar layer growth following from the classical mixed layer theory of Deardorff by a renormalization. This enhances the training data base of the generative machine learning algorithm and thus significantly improves the predicted statistics of the synthetically generated turbulence fields at different heights inside the boundary layer, above the surface layer. Differently to stochastic parametrizations, our model is able to predict the highly non‐Gaussian and transient statistics of buoyancy fluctuations, vertical velocity, and buoyancy flux at different heights thus also capturing the fastest thermals penetrating into the stabilized top region. The results of our generative algorithm agree with standard two‐equation mass‐flux schemes. The present parametrization provides additionally the granule‐type horizontal organization of the turbulent convection which cannot be obtained in any of the other model closures. Our proof of concept‐study also paves the way to efficient data‐driven convective parametrizations in other natural flows.

Machine Learning Based Predictions of Airfoil Aerodynamic Coefficients for Reynolds Number Extrapolations

This study investigates the application of various machine learning (ML) algorithms for predicting two critical aerodynamic coefficients, i.e. the maximum lift coefficient (C l max ) and the minimum drag coefficient (C d min ), for wind turbine airfoils at any given Reynolds number. We propose to use clustering techniques to group similar airfoil shapes and use the created partitions to predict unseen airfoil properties utilizing their similarity. Here, we also represent airfoils in the PARSEC low dimensional space, rather than high dimensional airfoil points space, to remedy the small number of training data. For this purpose, an extended experimental airfoil database is created and used for training models based on five different ML algorithms. We observe that the Decision Tree Ensemble (DTE), Random Forest (RF) and multi-layer perceptron (MLP) models emerge as the most effective predictors for C l max and C d min . Testing these two ML models on three additional airfoil cases not included in the training database shows that the C l max prediction performance is generally reasonable, with error levels being around 5% on average. In contrast, the prediction error levels for C d min are usually higher, with an average of around 15%.

Use of Artificial Intelligence to Monitor the Reliability of Removable Load-Handling Devices

Introduction. The malfunction of removable load-handling devices (RLHD) poses significant production risks. That is why research in this field is relevant. The problem has often become a topic of scientific investigation. The authors propose using artificial intelligence more extensively to monitor the state of RLHD. This paper presents a study on how to improve the machine vision model to better identify the absence of locks on RLHD hooks. A probable occurrence of such an issue in production is noted. A storage and monitoring system for RLHD condition is proposed. The aim of this study is to demonstrate the potential for further training of neural networks to significantly enhance the efficiency of RLHD monitoring, ensuring their safe use.Materials and Methods. The work is based on the results of a survey conducted at the LLC “KZ Rostselmash” plant from 2022 to 2023, involving 144 RLHD. Mathematical statistics methods were used to process the data. A neural network model previously trained using the YOLO computer vision algorithm was studied. It was retrained taking into account the norms of the rejection of RLHD, specified in federal rules and standards. Images of RLHD with defects and missing parts were collected from these sources and used to create a training database. The database was expanded by augmentation. The Roboflow platform was used for work.Results. The array of images used for further training of the neural network was divided into three samples: training (88%), validation (8%) and test (4%). These samples were used to train and validate its results. The training was completed after 260 epochs, with a steady increase in accuracy. The neural network model of computer vision obtained in this way automatically detected a common defect in the RLHD hook — the absence of a lock. Its performance was assessed using three indicators: average accuracy (94%), prediction accuracy (88.8%) and response (89.2%). The neural network could receive images from a video camera in real-time and recognize hook defects. During the RLHD inspection at the Rostselmash plant, a grab for lifting engines was found to have all three hooks defective — without locks. To avoid such situations, at the end of work, it was recommended to place the RLHD on a special stand equipped with a microcontroller device that could monitor for a number of potential issues using radio frequency identification.Discussion and Conclusions. The main goal of this proposed solution is to detect and address signs of non-compliance with the established standards. This task can be implemented in facilities that use lifting equipment. In this case, the timely noticed RLHD defects will allow preventing production incidents. As a result, material damage can be reduced and injury statistics improved.

Validation of a ray-tracing-based guided Lamb wave propagation methodology in aerostructures

Validation of a ray-tracing-based guided Lamb wave propagation methodology in aerostructures

Assessment of Roughness Characterization Methods for Data-Driven Predictions

A comparative analysis is undertaken to explore the impact of various roughness characterization methods as input variables on the performance of data-driven predictive models for estimating the roughness equivalent sand-grain size ks\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$k_s$$\\end{document}. The first type of model, denoted as ENNPS\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {ENN}_\ ext {PS}$$\\end{document}, incorporates the roughness height probability density function (p.d.f.) and power spectrum (PS), while the second type of model, ENNPA\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {ENN}_\ ext {PA}$$\\end{document}, utilizes a finite set of 17 roughness statistical parameters as input variables. Furthermore, a simplified parameter-based model, denoted as ENNPAM\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {ENN}_\ ext {PAM}$$\\end{document}, is considered, which features only 6 input roughness parameters. The models are trained based on identical databases and evaluated using roughness samples similar to the training databases as well as an external testing database based on literature. While the predictions based on p.d.f. and PS achieves a stable error level of around 10% among all considered testing samples, a notable deterioration in performance is observed for the parameter-based models for the external testing database, indicating a lower extrapolating capability to diverse roughness types. Finally, the sensitivity analysis on different types of roughness confirms an effective identification of distinct roughness effects by ENNPAM\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {ENN}_\ ext {PAM}$$\\end{document}, which is not observed for ENNPA\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {ENN}_\ ext {PA}$$\\end{document}. We hypothesize that the successful training of ENNPAM\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {ENN}_\ ext {PAM}$$\\end{document} is attributed to the enhanced training efficiency linked to the lower input dimensionality.

Machine-Learning-Driven High-Throughput Screening for High-Energy Density and Stable NASICON Cathodes.

The Na super ionic conductor (NASICON), which has outstanding structural stability and a high operating voltage, is an appealing material for overcoming the limits of low specific energy and larger volume distortion of sodium-ion batteries. In this study, to discover ideal NASICON cathode materials, a screening platform based on density functional theory (DFT) calculations and machine learning (ML) is developed. A training database was generated utilizing the previous 124 545 electrode databases, and a test set of 3126 potential NASICON structures [NaxMyM'1-y(PO4)3] with 27 dopants at the metal site and 6 dopants at the polyanion central site was constructed. The developed ML surrogate model identifies 796 materials that satisfy the following criteria: formation energy of <0.0 eV/atom, energy above hull of ≤0.025 eV/atom, volume change of ≤4%, and theoretical capacity of ≥50 mAh/g. The thermodynamically stable configurations of doped NASICON structures were then selected using machine learning interatomic potential (MLIP), enabling rapid consideration of various dopant site configurations. DFT calculations are followed on 796 screened materials to obtain energy density, average voltage, and volume change. Finally, 50 candidates with an average voltage of ≥3.5 V are identified. The suggested platform accelerates the exploration for optimal NASICON materials by narrowing the focus on materials with desired properties, saving considerable resources.

Enhancing wrist arthroscopy: artificial intelligence applications for bone structure recognition using machine learning

IntroductionWrist arthroscopy is a rapidly expanding surgical discipline, but has a long and challenging learning curve. One of its difficulties is distinguishing the various anatomical structures during the procedure.Although artificial intelligence has made significant progress in recent decades, its potential as a valuable tool in surgery training is largely untapped. Materials and methodsThe objective of this study was to develop an algorithm that could accurately recognize the anatomical bone structures of the wrist during arthroscopy. We prospectively included 20 wrist arthroscopies: 10 in patients and 10 in cadavers. For each surgery, we extracted and labeled images of the various carpal bones. These images were used to create a database for training, validating and testing a structure recognition algorithm. The primary criterion used was a Dice loss detection and categorization score for structures of interest, with a threshold greater than 80%. ResultsThe database contained 511 labeled images (4,088 after data augmentation). We developed a Deeplabv3+ classification algorithm with a U-Net architecture. After training and testing our algorithm, we achieved an average Dice loss score of 89% for carpal bone recognition. ConclusionThis study demonstrated reliable detection of different carpal bones during arthroscopic wrist surgery using artificial intelligence. However, some bones were detected more accurately than others, suggesting that additional algorithm training could further enhance performance. Application in real-life conditions could validate these results and potentially contribute to learning and improvement in arthroscopic wrist surgery. Level of evidenceIV.

Deep learning-based modeling of the cyclic behavior of replaceable fuse buckling-restrained braces (BRBs)

This paper presents a novel approach for predicting the behavior of buckling restrained braces (BRBs) using long short-term memory (LSTM) and residual neural network (ResNet) models. Nonlinear finite element analysis (NLFEA) usually requires 3D modeling of complex parts and assigning various material and simulation properties before conducting the analysis, which can be time-consuming and computationally costly. LSTM networks are particularly well-suited for modeling sequential data, such as time series, and ResNet is a discriminative deep learning-based model that can handle deeper architectures without the issue of vanishing gradient. Developing a general LSTM model requires a comprehensive database for training, cross-validation, and blind testing. Even if such a database were available, concatenating data from different BRB specimens would not be feasible when modeling a data series. Such a method would intertwine the features of the individual specimens, rendering the data useless for machine-learning purposes. On the other hand, a well-trained LSTM can predict cyclic behavior for specimens that resemble the training data. As such, a group of well-trained models can produce superior predictions when used on individually familiar data. Hence, there is a need for a prescreening classification stage. The framework presented herein utilizes an AI classifier (ResNet) that automatically recognizes the BRB specimen type of the input data and directs it to the appropriate LSTM model for prediction. The presented framework demonstrates excellent predictions with an accuracy greater than 99%. It is customizable, adaptable, and scalable to include different BRB specimens. The input data consisted of the displacement of the BRB specimens based on the AISC 341–10 qualification recommendations, while the output was the corresponding force (hysteretic response to cyclic loading). The proposed framework can be valuable in earthquake engineering applications as it demonstrates superior predictions.

Deep learning approaches to recover the plasma current density profile from the safety factor based on Grad–Shafranov solutions across multiple tokamaks

Many magnetohydrodynamic stability analyses require generation of a set of equilibria with a fixed safety factor q-profile while varying other plasma parameters. A neural network (NN)-based approach is investigated that facilitates such a process. Both multilayer perceptron (MLP)-based NN and convolutional neural network (CNN) models are trained to map the q-profile to the plasma current density J-profile, and vice versa, while satisfying the Grad–Shafranov radial force balance constraint. When the initial target models are trained, using a database of semi-analytically constructed numerical equilibria, an initial CNN with one convolutional layer is found to perform better than an initial MLP model. In particular, a trained initial CNN model can also predict the q- or J-profile for experimental tokamak equilibria. The performance of both initial target models is further improved by fine-tuning the training database, i.e. by adding realistic experimental equilibria with Gaussian noise. The fine-tuned target models, referred to as fine-tuned MLP and fine-tuned CNN, well reproduce the target q- or J-profile across multiple tokamak devices. As an important application, these NN-based equilibrium profile convertors can be utilized to provide a good initial guess for iterative equilibrium solvers, where the desired input quantity is the safety factor instead of the plasma current density.

An Artificial Intelligence-Based Intrusion Detection System using Optimization and Deep Learning

The Intrusion Detection System (IDS) was developed to resolve the malicious activities on the networking. However, the existing IDS models face issues in achieving high accuracy, and it is prone to detection error. Hence, to address these challenges, an innovative Vulture-based Deep Belief System (VbDBNS) for enhancing the performance of detecting intrusion activity by monitoring their behaviour and features. This study utilized the NSL-KDD database for training and validating the system. Henceforth, pre-processing is employed to standardize the dataset using Min-Max normalization, and 1-N encoding, which transforms the categorical data variables. Moreover, feature extraction was utilized to capture the most significant 42 attributes from the pre-processed database. Further, a classification module was designed using VbDBN to categorize normal and malicious data in which the fitness of vulture was updated, which enables it to monitor and detect the intrusion continuously. The experimental results of the study illustrate that the designed methodology achieved better outcomes than the conventional techniques in terms of recall, accuracy, precision, F1-score, etc.

ВЫБОР МОДЕЛИ И АРХИТЕКТУРЫ НЕЙРОННОЙ СЕТИ ДЛЯ СИСТЕМЫ ТЕХНИЧЕСКОГО ЗРЕНИЯ ДЕТЕКЦИИ КОРНЕПЛОДОВ САХАРНОЙ СВЁКЛЫ

The use of a technical vision system, aimed at ensuring greater safety of sugar beet roots during harvesting and storage by analyzing raw materials arriving at factories, is considered among the methods of long-term storage. The process of distribution of trucks at the receiving point of sugar factories is considered by humans in the work. The purpose of the work is to develop a method for detecting sugar beet roots on an image of the surface of a truck embankment at the receiving point of sugar factories. A neural network was chosen as the raw material recognition method due to the extremely diverse nature of the resulting images, which recognition using a threshold transformation cannot cope with. Algorithms (options such as SSD, YOLO, R-CNN) and architectures (options such as AlexNet, VGGNet, GoogleNet, ResNet) of various neural networks are considered. The most suitable neural network for the required tasks has been identified. The optimal size of the database for training has been determined. As a result of training and testing the neural network, the need to create additional filters was identified. Filters have been created and described that clarify the work of the neural network: filtering out “stuck together” objects, objects with low class membership, objects with an impossible area (too small or large), as well as with an aspect ratio of less than 0.2 or more than 5. A neural network capable to find a sufficient number of sugar beet roots in incoming images and to meet the requirements for high-quality functioning of the technical vision system, was trained based on the results of the work performed.

Establishment of a Clinical Thinking Ability Training Database Based on Artificial Intelligence Technology: Analysis Study (Preprint)

Establishment of a Clinical Thinking Ability Training Database Based on Artificial Intelligence Technology: Analysis Study (Preprint)

Prototyping of a Machine Learning–Based Burnup Measurement Capability for Pebble Bed Reactor Fuel

The pebble bed reactor is a unique reactor design due to its capacity for continuous multipass circulation of the fuel elements, without causing interruption to reactor operation, with the assistance of the burnup measurement system. Such a system necessarily requires an accurate knowledge of the burnup of each fuel pebble upon ejection from the core so as to inform the reloading decision and to ensure that no pebble exceeds the regulated discharge burnup limit at any point following reinsertion into the reactor core. In this work, we conceptualize, develop, and demonstrate a machine learning–based fuel burnup prediction framework leveraging advanced modeling and simulation capabilities. At its core, machine learning regression models are learned from simulated data to establish the correlation among the irradiated fuel composition (hence burnup), the gamma leakage spectrum, and the gamma spectroscopy results. Sensitivity analysis is conducted to quantify the impact of unknown design parameters, such as fuel enrichment, and irradiation environment, including power density, temperature, and neighboring materials, on the prediction accuracy of various supervised regression algorithms. The effects of a short cooldown period on machine learning prediction accuracy are also investigated. A test data set is used to validate that the data generation methodology proposed in this work successfully results in a machine learning model capable of interpolating its prediction of burnup onto a much wider range of irradiation conditions than were explicitly represented in the training database. The inclusion of a cooldown period of just 2 h leads to a prediction root-mean-square error of <5 MWd/kgU when the fuel enrichment is known and <9 MWd/kgU otherwise.

Multilevel thermoplastic waste segregation and classification with AHGSO using federated learning framework

PurposeAn abundance of techniques has been presented so forth for waste classification but, they deliver inefficient results with low accuracy. Their achievement on various repositories is different and also, there is insufficiency of high-scale databases for training. The purpose of the study is to provide high security.Design/methodology/approachIn this research, optimization-assisted federated learning (FL) is introduced for thermoplastic waste segregation and classification. The deep learning (DL) network trained by Archimedes Henry gas solubility optimization (AHGSO) is used for the classification of plastic and resin types. The deep quantum neural networks (DQNN) is used for first-level classification and the deep max-out network (DMN) is employed for second-level classification. This developed AHGSO is obtained by blending the features of Archimedes optimization algorithm (AOA) and Henry gas solubility optimization (HGSO). The entities included in this approach are nodes and servers. Local training is carried out depending on local data and updations to the server are performed. Then, the model is aggregated at the server. Thereafter, each node downloads the global model and the update training is executed depending on the downloaded global and the local model till it achieves the satisfied condition. Finally, local update and aggregation at the server is altered based on the average method. The Data tag suite (DATS_2022) dataset is used for multilevel thermoplastic waste segregation and classification.FindingsBy using the DQNN in first-level classification the designed optimization-assisted FL has gained an accuracy of 0.930, mean average precision (MAP) of 0.933, false positive rate (FPR) of 0.213, loss function of 0.211, mean square error (MSE) of 0.328 and root mean square error (RMSE) of 0.572. In the second level classification, by using DMN the accuracy, MAP, FPR, loss function, MSE and RMSE are 0.932, 0.935, 0.093, 0.068, 0.303 and 0.551.Originality/valueThe multilevel thermoplastic waste segregation and classification using the proposed model is accurate and improves the effectiveness of the classification.

Performance of a Region of Interest–based Algorithm in Diagnosing International Society of Urological Pathology Grade Group ≥2 Prostate Cancer on the MRI-FIRST Database—CAD-FIRST Study

Background and objectiveProstate multiparametric magnetic resonance imaging (MRI) shows high sensitivity for International Society of Urological Pathology grade group (GG) ≥2 cancers. Many artificial intelligence algorithms have shown promising results in diagnosing clinically significant prostate cancer on MRI. To assess a region-of-interest–based machine-learning algorithm aimed at characterising GG ≥2 prostate cancer on multiparametric MRI. MethodsThe lesions targeted at biopsy in the MRI-FIRST dataset were retrospectively delineated and assessed using a previously developed algorithm. The Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2) score assigned prospectively before biopsy and the algorithm score calculated retrospectively in the regions of interest were compared for diagnosing GG ≥2 cancer, using the areas under the curve (AUCs), and sensitivities and specificities calculated with predefined thresholds (PIRADSv2 scores ≥3 and ≥4; algorithm scores yielding 90% sensitivity in the training database). Ten predefined biopsy strategies were assessed retrospectively. Key findings and limitationsAfter excluding 19 patients, we analysed 232 patients imaged on 16 different scanners; 85 had GG ≥2 cancer at biopsy. At patient level, AUCs of the algorithm and PI-RADSv2 were 77% (95% confidence interval [CI]: 70–82) and 80% (CI: 74–85; p = 0.36), respectively. The algorithm’s sensitivity and specificity were 86% (CI: 76–93) and 65% (CI: 54–73), respectively. PI-RADSv2 sensitivities and specificities were 95% (CI: 89–100) and 38% (CI: 26–47), and 89% (CI: 79–96) and 47% (CI: 35–57) for thresholds of ≥3 and ≥4, respectively. Using the PI-RADSv2 score to trigger a biopsy would have avoided 26–34% of biopsies while missing 5–11% of GG ≥2 cancers. Combining prostate-specific antigen density, the PI-RADSv2 and algorithm’s scores would have avoided 44–47% of biopsies while missing 6–9% of GG ≥2 cancers. Limitations include the retrospective nature of the study and a lack of PI-RADS version 2.1 assessment. Conclusions and clinical implicationsThe algorithm provided robust results in the multicentre multiscanner MRI-FIRST database and could help select patients for biopsy. Patient summaryAn artificial intelligence–based algorithm aimed at diagnosing aggressive cancers on prostate magnetic resonance imaging showed results similar to expert human assessment in a prospectively acquired multicentre test database.

Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method.

In recent years, the application of deep learning models for underwater target recognition has become a popular trend. Most of these are pure 1D models used for processing time-domain signals or pure 2D models used for processing time-frequency spectra. In this paper, a recent temporal 2D modeling method is introduced into the construction of ship radiation noise classification models, combining 1D and 2D. This method is based on the periodic characteristics of time-domain signals, shaping them into 2D signals and discovering long-term correlations between sampling points through 2D convolution to compensate for the limitations of 1D convolution. Integrating this method with the current state-of-the-art model structure and using samples from the Deepship database for network training and testing, it was found that this method could further improve the accuracy (0.9%) and reduce the parameter count (30%), providing a new option for model construction and optimization. Meanwhile, the effectiveness of training models using time-domain signals or time-frequency representations has been compared, finding that the model based on time-domain signals is more sensitive and has a smaller storage footprint (reduced to 30%), whereas the model based on time-frequency representation can achieve higher accuracy (1-2%).

Cloud-Based Machine Learning Platform to Predict Clinical Outcomes at Home for Patients With Cardiovascular Conditions Discharged From Hospital: Clinical Trial.

Hospitalizations account for almost one-third of the US $4.1 trillion health care cost in the United States. A substantial portion of these hospitalizations are attributed to readmissions, which led to the establishment of the Hospital Readmissions Reduction Program (HRRP) in 2012. The HRRP reduces payments to hospitals with excess readmissions. In 2018, >US $700 million was withheld; this is expected to exceed US $1 billion by 2022. More importantly, there is nothing more physically and emotionally taxing for readmitted patients and demoralizing for hospital physicians, nurses, and administrators. Given this high uncertainty of proper home recovery, intelligent monitoring is needed to predict the outcome of discharged patients to reduce readmissions. Physical activity (PA) is one of the major determinants for overall clinical outcomes in diabetes, hypertension, hyperlipidemia, heart failure, cancer, and mental health issues. These are the exact comorbidities that increase readmission rates, underlining the importance of PA in assessing the recovery of patients by quantitative measurement beyond the questionnaire and survey methods. This study aims to develop a remote, low-cost, and cloud-based machine learning (ML) platform to enable the precision health monitoring of PA, which may fundamentally alter the delivery of home health care. To validate this technology, we conducted a clinical trial to test the ability of our platform to predict clinical outcomes in discharged patients. Our platform consists of a wearable device, which includes an accelerometer and a Bluetooth sensor, and an iPhone connected to our cloud-based ML interface to analyze PA remotely and predict clinical outcomes. This system was deployed at a skilled nursing facility where we collected >17,000 person-day data points over 2 years, generating a solid training database. We used these data to train our extreme gradient boosting (XGBoost)-based ML environment to conduct a clinical trial, Activity Assessment of Patients Discharged from Hospital-I, to test the hypothesis that a comprehensive profile of PA would predict clinical outcome. We developed an advanced data-driven analytic platform that predicts the clinical outcome based on accurate measurements of PA. Artificial intelligence or an ML algorithm was used to analyze the data to predict short-term health outcome. We enrolled 52 patients discharged from Stanford Hospital. Our data demonstrated a robust predictive system to forecast health outcome in the enrolled patients based on their PA data. We achieved precise prediction of the patients' clinical outcomes with a sensitivity of 87%, a specificity of 79%, and an accuracy of 85%. To date, there are no reliable clinical data, using a wearable device, regarding monitoring discharged patients to predict their recovery. We conducted a clinical trial to assess outcome data rigorously to be used reliably for remote home care by patients, health care professionals, and caretakers.

Predicting the formation of NADES using a transformer-based model

The application of natural deep eutectic solvents (NADES) in the pharmaceutical, agricultural, and food industries represents one of the fastest growing fields of green chemistry, as these mixtures can potentially replace traditional organic solvents. These advances are, however, limited by the development of new NADES which is today, almost exclusively empirically driven and often derivative from known mixtures. To overcome this limitation, we propose the use of a transformer-based machine learning approach. Here, the transformer-based neural network model was first pre-trained to recognize chemical patterns from SMILES representations (unlabeled general chemical data) and then fine-tuned to recognize the patterns in strings that lead to the formation of either stable NADES or simple mixtures of compounds not leading to the formation of stable NADES (binary classification). Because this strategy was adapted from language learning, it allows the use of relatively small datasets and relatively low computational resources. The resulting algorithm is capable of predicting the formation of multiple new stable eutectic mixtures (n = 337) from a general database of natural compounds. More importantly, the system is also able to predict the components and molar ratios needed to render NADES with new molecules (not present in the training database), an aspect that was validated using previously reported NADES as well as by developing multiple novel solvents containing ibuprofen. We believe this strategy has the potential to transform the screening process for NADES as well as the pharmaceutical industry, streamlining the use of bioactive compounds as functional components of liquid formulations, rather than simple solutes.