Neural Model Research Articles

Predicting which patients with cancer will see a psychiatrist or counsellor from their initial oncology consultation document using natural language processing.

Patients with cancer often have unmet psychosocial needs. Early detection of who requires referral to a counsellor or psychiatrist may improve their care. This work used natural language processing to predict which patients will see a counsellor or psychiatrist from a patient's initial oncology consultation document. We believe this is the first use of artificial intelligence to predict psychiatric outcomes from non-psychiatric medical documents. This retrospective prognostic study used data from 47,625 patients at BC Cancer. We analyzed initial oncology consultation documents using traditional and neural language models to predict whether patients would see a counsellor or psychiatrist in the 12 months following their initial oncology consultation. Here, we show our best models achieved a balanced accuracy (receiver-operating-characteristic area-under-curve) of 73.1% (0.824) for predicting seeing a psychiatrist, and 71.0% (0.784) for seeing a counsellor. Different words and phrases are important for predicting each outcome. These results suggest natural language processing can be used to predict psychosocial needs of patients with cancer from their initial oncology consultation document. Future research could extend this work to predict the psychosocial needs of medical patients in other settings.

A deep neural network for vehicle detection in aerial images

This research paper highlights the significance of vehicle detection in aerial images for surveillance systems, focusing on deep learning methods that outperform traditional approaches. However, the challenge of high computation complexity due to diverse vehicle appearances persists. The motivation behind this study is to highlight the crucial role of vehicle detection in aerial images for surveillance systems, emphasizing the superior performance of deep learning methods compared to traditional approaches. To address this, a lightweight deep neural network-based model is developed, striking a balance between accuracy and efficiency enabling real-time operation. The model is trained and evaluated on a standardized dataset, with extensive experiments demonstrating its ability to achieve accurate vehicle detection with significantly reduced computation costs, offering a practical solution for real-world aerial surveillance scenarios.

Automated real-time prediction of geological formation tops during drilling operations: an applied machine learning solution for the Norwegian Continental Shelf

AbstractAccurate prediction of geological formation tops is a crucial task for optimizing hydrocarbon exploration and production activities. This research investigates and conducts a comprehensive comparative analysis of several advanced machine learning approaches tailored for the critical application of geological formation top prediction within the complex Norwegian Continental Shelf (NCS) region. The study evaluates and benchmarks the performance of four prominent machine learning models: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forest ensemble method, and Multi-Layer Perceptron (MLP) neural network. To facilitate a rigorous assessment, the models are extensively evaluated across two distinct datasets - a dedicated test dataset and a blind dataset independent for validation. The evaluation criteria revolve around quantifying the models' predictive accuracy in successfully classifying multiple geological formation top types. Additionally, the study employs the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm as a baseline benchmarking technique to contextualize the relative performance of the machine learning models against a conventional clustering approach. Leveraging two model-agnostic feature importance analysis techniques - Permutation Feature Importance (PFI) and Shapley Additive exPlanations (SHAP), the investigation identifies and ranks the most influential input variables driving the predictive capabilities of the models. The comprehensive analysis unveils the MLP neural network model as the top-performing approach, achieving remarkable predictive accuracy with a perfect score of 0.99 on the blind validation dataset, surpassing the other machine learning techniques as well as the DBSCAN benchmark. However, the SVM model attains superior performance on the initial test dataset, with an accuracy of 0.99. Intriguingly, the PFI and SHAP analyses converge in consistently pinpointing depth (DEPT), revolution per minute (RPM), and Hook-load (HKLD) as the three most impactful parameters influencing model predictions across the different algorithms. These findings underscore the potential of sophisticated machine learning methodologies, particularly neural network-based models, to significantly enhance the accuracy of geological formation top prediction within the geologically complex NCS region. However, the study emphasizes the necessity for further extensive testing on larger datasets to validate the generalizability of the high performance observed. Overall, this research delivers an exhaustive comparative evaluation of state-of-the-art machine learning techniques, offering critical insights to guide the optimal selection, development, and real-world deployment of accurate and reliable predictive modeling strategies tailored for hydrocarbon exploration and reservoir characterization endeavors in the NCS. Graphical abstract

Experimental validation of ventilated double-skin façades aided by neural networks and thermal modelling for heating demand reduction

Buildings are a significant contributor to global energy consumption, particularly in cold regions where heating is a primary concern. Double-Skin Façades have emerged as a promising passive building technology by reducing heating demand and enhancing thermal comfort. This study introduces a novel approach using a neural network-aided thermal model to improve the assessment of Double-Skin Façades. Annual real-time data was collected from an existing façade before conducting a sensitivity analysis to identify critical variables impacting heat loss through the façade. These variables were subsequently utilized to develop a heat transfer model for the existing façade. After validation against experimentally collected data, the model was updated for evaluating the proposed Double-Skin Façades. Evaluation under various conditions, including closed, natural ventilation, and mechanical ventilation scenarios, demonstrated the potential of Double-Skin Façades in reducing heat loss in cold regions. The developed heat transfer model showed significant accuracy, with minimal average errors of 0.20 % in winter and 0.32 % in summer. In winter, closed Double-Skin Façades significantly reduced heat loss by up to 24.65 % compared to Single-Skin Façades. In summer, mechanically ventilated Double-Skin Façades reduced heat gain by up to 33.7 % for a cavity airflow of 1.6 m/s.

Diabetic retinopathy detection using Bilayered Neural Network classification model with resubstitution validation

In recent years, eye diseases in diabetic patients are one of the most common has been diabetic retinopathy (DR). which leads to complete blindness in advanced stages. Diabetes affects the blood vessels in the retina and causes vision loss. One of the ways to decrease the risk of this issue is to detect diabetic retinopathy in its early stages. This study describes a computer-aided screening system (DREAM) that uses a neural network classification model in machine learning to assess fundus images with different illumination and fields of vision and provide a severity grade for diabetic retinopathy. Moreover, the methodology of this study based on:•Enhancement techniques have been used on dataset images, histogram equalization, noise reduction and image scaling,•vSLAM has been selected as feature extraction,•Bilayered Neural Network under resubstitution validation used as a classification model.Finally, after testing on the DR severity grading system is tested on 6332 images of detection of diabetic retinopathy images, the result of the ROC curve is 0.985 on image dataset and obtains accuracy reached 98.5%. The classification result has been founded under MATLAB platform, beside it that be work with real time analysis and detection when patient eyes analysis.

Neural network-based surrogate modeling and optimization of a multigeneration system

Multi-Objective Optimization (MOO) poses a computational challenge, particularly when applied to physics-based models. As a result, only up to three objectives are typically involved in simulation-based optimization. To go beyond this number, Surrogate Models (SMs) need to replace such high-fidelity models. In this exploratory study, the objectives are to perform comprehensive regression surrogate modeling and to conduct MOO for a Multi-Generation System (MGS). The most suitable SM was chosen among four neural-network models: Artificial Neural Network (ANN), Convolutional Neural Network (CNN), Long-Short Term Memory (LSTM), and an ensemble model developed through brute-force search using the three aforementioned models. The final model was found to be superior to others, achieving R2 values ranging from 0.9830 to 0.9999. Next, an optimization problem with six conflicting objectives was defined and performed at four distinct values of Direct Normal Irradiation (DNI), a time-dependent feature. This aimed to provide multi-criteria decision-making information based on atmospheric transparency. As a result, new understandings were gained: (I) exergy efficiency, production cost, and freshwater production rate were found to be highly influenced by DNI, and (II) the critical range of operation was observed within the DNI interval of 100 to 400 W/m2. Furthermore, we compared the result of the six-objective optimization with that of the bi-objective optimization obtained in our simulation-based study and found that all objectives showed improvements ranging from 1.9% to 12.7%. Finally, based on the findings obtained in the present study, some practical recommendations were put forward for applying the proposed methodology to similar MGSs.

Neural stochastic agent‐based limit order book simulation with neural point process and diffusion probabilistic model

SummaryModern financial exchanges use an electronic limit order book (LOB) to store bid and ask orders for a specific financial asset. As the most fine‐grained information depicting the demand and supply of an asset, LOB data are essential in understanding market dynamics. Therefore, realistic LOB simulations offer a valuable methodology for explaining the empirical properties of markets. Mainstream simulation models include agent‐based models (ABMs) and stochastic models (SMs). However, ABMs tend not to be grounded on real historical data, whereas SMs tend not to enable dynamic agent‐interaction. More recently, deep generative approaches have been successfully implemented to tackle these issues, whereas its black‐box essence hampers the explainability and transparency of the system. To overcome these limitations, we propose a novel hybrid neural stochastic agent‐based model (NS‐ABM) for LOB simulation that incorporates a neural stochastic trader in agent‐based simulation, characterised by (1) representing the aggregation of market events' logic by a neural stochastic background trader that is pre‐trained on historical LOB data through a neural point process model; (2) learning the complex distribution for order‐related attributes conditioned on various market indicators through a non‐parametric diffusion probabilistic model; and (3) embedding the background trader in a multi‐agent simulation platform to enable interaction with other strategic trading agents. We instantiate this hybrid NS‐ABM model using the ABIDES platform. We first run the background trader in isolation and show that the simulated LOB can recreate a comprehensive list of stylised facts that demonstrate realistic market behaviour. We then introduce a population of ‘trend’ and ‘value’ trading agents, which interact with the background trader. We show that the stylised facts remain and we demonstrate order flow impact and financial herding behaviours that are in accordance with empirical observations of real markets.

A cooperative neural dynamic model for solving general convex nonlinear optimization problems with fuzzy parameters and an application in manufacturing systems

SummaryIn the presented study, the solution of the fuzzy nonlinear optimization problems (FNLOPs) is calculated using a recurrent neural network (RNN) model. Since there is a few research for solving FNLOP by RNN's, we give a new approach to solve the problem. By reducing the original program to an interval problem and then weighting problem, the Karush–Kuhn–Tucker (KKT) conditions are given. Moreover, we use the KKT conditions into a RNN as an important tool to solve the problem. Besides, the global convergence properties and the Lyapunov stability of the dynamic model are studied in this study. In the final step, some illustrative examples are considered to establish the obtained results. Reported results are compared with some others network models.

Human motion activity recognition and pattern analysis using compressed deep neural networks

ABSTRACT This work presents an on-device machine learning model with the ability to identify different mobility gestures called human activity recognition (HAR), which includes running, walking, squatting, jumping, and others. The data is collected through an Arduino Nano 33 BLE Sense board with a sampling rate of 119 Hz, which is embedded with an Inertial Measurement Unit (IMU) sensor. The same board is used as a microcontroller to identify human gestures by developing an end-to-end edge computing application. A deep neural network model is trained and then compressed for deployment on the board to create a self-contained, embedded device capable of identifying the type of gesture performed. Three deep learning models, namely Multi-Layer Perceptron (MLP), Convolutional Neural Network – Long Short Term Memory (CNN-LSTM) & CNN-Gated Recurrent Unit (CNN-GRU), are evaluated in the identification of the mobility gestures. The observed accuracy of the models is 96%, 97.1% and 97.8%, respectively, MLP, CNN-LSTM & CNN-GRU across the different gesture categories. The study shows the utility of embedded devices with deep neural network-based models, which can provide low cost, minimal power usage, and meet data privacy requirements in HAR.

Multi-step planning with learned effects of partial action executions

In this paper, we propose a novel affordance model, which combines object, action, and effect information in the latent space of a predictive neural network architecture that is built on Conditional Neural Processes. Our model allows us to make predictions of intermediate effects expected to be obtained during action executions and make multi-step plans that include partial actions. We first compared the prediction capability of our model using an existing interaction data set and showed that it outperforms a recurrent neural network-based model in predicting the effects of lever-up actions. Next, we showed that our model can generate accurate effect predictions for other actions, such as push and grasp actions. Our system was shown to generate successful multi-step plans to bring objects to desired positions using the traditional A* search algorithm. Furthermore, we realized a continuous planning method and showed that the proposed system generated more accurate and effective plans with sequences of partial action executions compared to plans that only consider full action executions using both planning algorithms.

Meta-Learning Enhanced Trade Forecasting: A Neural Framework Leveraging Efficient Multicommodity STL Decomposition

In the dynamic global trade environment, accurately predicting trade values of diverse commodities is challenged by unpredictable economic and political changes. This study introduces the Meta-TFSTL framework, an innovative neural model that integrates Meta-Learning Enhanced Trade Forecasting with efficient multicommodity STL decomposition to adeptly navigate the complexities of forecasting. Our approach begins with STL decomposition to partition trade value sequences into seasonal, trend, and residual elements, identifying a potential 10-month economic cycle through the Ljung–Box test. The model employs a dual-channel spatiotemporal encoder for processing these components, ensuring a comprehensive grasp of temporal correlations. By constructing spatial and temporal graphs leveraging correlation matrices and graph embeddings and introducing fused attention and multitasking strategies at the decoding phase, Meta-TFSTL surpasses benchmark models in performance. Additionally, integrating meta-learning and fine-tuning techniques enhances shared knowledge across import and export trade predictions. Ultimately, our research significantly advances the precision and efficiency of trade forecasting in a volatile global economic scenario.

A Novel Hybrid Model Combining BPNN Neural Network and Ensemble Empirical Mode Decomposition

Neural network models have been successfully used to predict stock prices, weather, and traffic patterns. Due to the sensitivity of the data, it is very effective in identifying and maintaining long-term dependencies in time series. The back propagation neural network (BPNN) model works well in regression and classification applications, such as predicting stock prices and sales volumes. BPNN needs to sort out the mapping between inputs and outputs before continuous values. BPNN neural network model is integrated with ensemble empirical mode decomposition (EEMD), and a new hybrid neural network prediction model is constructed. Integrating ensemble empirical mode decomposition, collecting and preprocessing sequence features, reducing noise, improving robustness, and then training neural networks with returned feature vectors instead. In the international gold price series forecasting, the R2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$R^2$$\\end{document} of the new hybrid model is 1.85%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} better than the existing EEMD-LSTM model, 3.8%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} and 5.44%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\%$$\\end{document} better than the independent BPNN and long short-term memory network (LSTM) neural network models, respectively. Compared with LSTM, the BPNN plays the performance of EEMD better, reduces the error to a certain extent, and improves the prediction accuracy.

Two Proposed Models for Face Recognition: Achieving High Accuracy and Speed with Artificial Intelligence

In light of the development in computer science and modern technologies, the impersonation crime rate has increased. Consequently, face recognition technology and biometric systems have been employed for security purposes in a variety of applications including human-computer interaction, surveillance systems, etc. Building an advanced sophisticated model to tackle impersonation-related crimes is essential. This study proposes classification Machine Learning (ML) and Deep Learning (DL) models, utilizing Viola-Jones, Linear Discriminant Analysis (LDA), Mutual Information (MI), and Analysis of Variance (ANOVA) techniques. The two proposed facial classification systems are J48 with LDA feature extraction method as input, and a one-dimensional Convolutional Neural Network Hybrid Model (1D-CNNHM). The MUCT database was considered for training and evaluation. The performance, in terms of classification, of the J48 model reached 96.01% accuracy whereas the DL model that merged LDA with MI and ANOVA reached 100% accuracy. Comparing the proposed models with other works reflects that they are performing very well, with high accuracy and low processing time.

Measuring Color Index of Transformer Oil–Enabling Single-Wavelength Spectroscopy With Artificial Neural Network-Fuzzy Logic Model

Conventionally, the color index of transformer oil is determined by a color comparator based on the American Society for Testing and Materials (ASTM) D 1500 standard. The equipment requires humans to operate, which leads to human error and limited number of samples tested per day. This work proposes the utilization of single-wavelength spectroscopy with 405 nm laser diode using artificial neural network (ANN) to determine the color index of transformer oil. Two ANN models were developed using data collected from 50 oil samples with different optical pathlengths of 1 to 10 mm, and laser output powers of 1 to 15 mW. The first model classified the input into different color indices and another model correlated the input parameters through regression analysis to determine the color index. A hybrid ANN-fuzzy logic model was also developed to improve the color index prediction. The root-mean-squared error (RMSE) obtained from the prediction by ANN regressor and ANN classifier are 0.5602 and 0.6416, respectively. The hybrid ANN-fuzzy logic model improves the RMSE especially for optical pathlengths <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$&lt;$</tex-math> </inline-formula> 5 mm, which is required for measuring samples with high color index. This proposed method reduces the dependency on complex optoelectronic hardware to obtain highly accurate results. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —Unlike the conventional testing method for color index of transformer oil that requires human observation, the findings of this study enables the possibility of compact and smart portable device through the utilization of single wavelength spectroscopy with machine learning models. With no human involvement, more computational power with lesser hardware dependency, the maintenance cost and error can be reduced. This proposed method can potentially be applied to measure the color of other amber-colored liquid products such as olive oil, honey and others.

Neuroethology: Decoding the waggle dance

A new study combining high-speed video recordings and computational modeling has revealed an overlooked feature of the famous honeybee waggle dance, yielding the first biologically plausible neural circuit model of how the information transmitted via the waggle dance could be assimilated by the follower bees.

The reduced-order model of 5 × 5 fuel rod bundles

The fuel rod bundles are the core part of pressurized water reactors (PWRs), and its heat transfer characteristics directly impact the safety of PWRs. A computational fluid dynamics (CFD) model of 5 × 5 fuel rod bundles with a spacer grid is established, and the numerical simulation results are in excellent agreement with the experimental results. Then, the effects of four turbulence models, namely shear stress transport model, standard k–ε model, re-normalization group k–ε model, and realizable k–ε model on the thermal-hydraulic characteristics of the 5 × 5 fuel rod bundles are systematically investigated. Furthermore, two data-driven methods, namely proper orthogonal decomposition (POD) and dynamic mode decomposition (DMD), are used to analyze the flow fields of the 5 × 5 fuel rod bundles. The two methods can extract key modes or features to enhance the comprehension and description of the dynamic behaviors within the flow fields of 5 × 5 fuel rod bundles. Finally, two reduced-order models (ROMs), called the POD-radial basis function neural network surrogate model and DMD method, are constructed, which can enable rapid prediction of the flow fields for 5 × 5 fuel rod bundles with high accuracy. The CFD simulation results presented in this paper can provide valuable insights for studying the thermal-hydraulic characteristics of the 5 × 5 fuel rod bundles. The two ROMs proposed in this paper can significantly reduce the computational costs associated with studying the thermal-hydraulic characteristics of 5 × 5 fuel rod bundles.

Artificial neural network-substituted transition model for crossflow instability: Modeling strategy and application prospect

Data-driven approaches have made preliminary inroads into the area of transition–turbulence modeling, but are still in their infancy with regard to widespread industrial adoption. This paper establishes an artificial neural network (ANN)-based transition model to enhance the capacity of capturing the crossflow (CF) transition phenomena, which are frequently identified over a wide range of aerodynamic problems. By taking a new CF-extended shear stress transport (SST) transition-predictive (SST-γ) model as the baseline, a mapping from mean flow variables to transition intermittency factor (γ) is constructed by ANN algorithm at various Mach and Reynolds numbers of an infinite swept wing. Generalizability of the resulting ANN-based (SST-γANN) model is fully validated in the same infinite swept wing, an inclined 6:1 prolate spheroid, and a finite swept wing in extensive experiment regimes, together with two effective a priori analysis strategies. Furthermore, the calculation efficiency, grid dependence, and performance of the present model in non-typical transitional flow are also assessed to inspect its industrial feasibility, followed by the elucidation of rationality behind the preliminary success and transferability of present framework. The results manifest that the SST-γANN model aligns well with the benchmark SST-γ model, and both can capture the CF transition accurately compared with their experiment counterpart, completely breaking through the disability of original SST-γ model without CF correction. In addition, good properties of efficiency, robustness, and generalizability are achieved for the ANN-alternative transition model, together with the usability of present framework across various transitional flows.

Knowledge-enhanced Graph Topic Transformer for Explainable Biomedical Text Summarization.

Given the overwhelming and rapidly increasing volumes of the published biomedical literature, automatic biomedical text summarization has long been a highly important task. Recently, great advances in the performance of biomedical text summarization have been facilitated by pre-trained language models (PLMs) based on fine-tuning. However, existing summarization methods based on PLMs do not capture domain-specific knowledge. This can result in generated summaries with low coherence, including redundant sentences, or excluding important domain knowledge conveyed in the full-text document. Furthermore, the black-box nature of the transformers means that they lack explainability, i.e. it is not clear to users how and why the summary was generated. The domain-specific knowledge and explainability are crucial for the accuracy and transparency of biomedical text summarization methods. In this article, we aim to address these issues by proposing a novel domain knowledge-enhanced graph topic transformer (DORIS) for explainable biomedical text summarization. The model integrates the graph neural topic model and the domain-specific knowledge from the Unified Medical Language System (UMLS) into the transformer-based PLM, to improve the explainability and accuracy. Experimental results on four biomedical literature datasets show that our model outperforms existing state-of-the-art (SOTA) PLM-based summarization methods on biomedical extractive summarization. Furthermore, our use of graph neural topic modeling means that our model possesses the desirable property of being explainable, i.e. it is straightforward for users to understand how and why the model selects particular sentences for inclusion in the summary. The domain-specific knowledge helps our model to learn more coherent topics, to better explain the performance.

Artificial Neural Network Language Models Predict Human Brain Responses to Language Even After a Developmentally Realistic Amount of Training.

Artificial neural networks have emerged as computationally plausible models of human language processing. A major criticism of these models is that the amount of training data they receive far exceeds that of humans during language learning. Here, we use two complementary approaches to ask how the models' ability to capture human fMRI responses to sentences is affected by the amount of training data. First, we evaluate GPT-2 models trained on 1 million, 10 million, 100 million, or 1 billion words against an fMRI benchmark. We consider the 100-million-word model to be developmentally plausible in terms of the amount of training data given that this amount is similar to what children are estimated to be exposed to during the first 10 years of life. Second, we test the performance of a GPT-2 model trained on a 9-billion-token dataset to reach state-of-the-art next-word prediction performance on the human benchmark at different stages during training. Across both approaches, we find that (i) the models trained on a developmentally plausible amount of data already achieve near-maximal performance in capturing fMRI responses to sentences. Further, (ii) lower perplexity-a measure of next-word prediction performance-is associated with stronger alignment with human data, suggesting that models that have received enough training to achieve sufficiently high next-word prediction performance also acquire representations of sentences that are predictive of human fMRI responses. In tandem, these findings establish that although some training is necessary for the models' predictive ability, a developmentally realistic amount of training (∼100 million words) may suffice.

466 Development of Machine Learning Algorithms to Predict Symptomatic VTE at Time of Admission and Time of Discharge after Severe Traumatic Injury

OBJECTIVES/GOALS: Clinical indicators predictive of venous thromboembolism (VTE) in trauma patients at multiple time points are not well outlined, particularly at time of discharge. We aimed to describe and predict inpatient and post-discharge risk factors of VTE after trauma using a multi-variate regression model and best of class machine learning (ML) models. METHODS/STUDY POPULATION: In a prospective, case-cohort study, all trauma patients (pts) who arrived as level 1 or 2 trauma activations, from June 2018 to February 2020 were considered for study inclusion. A subset of pts who developed incident, first time, VTE and those who did not develop VTE within 90 days of discharge were identified. VTE were confirmed either by imaging or at autopsy during inpatient stay or post-discharge. Outcomes were defined as the development of symptomatic VTE (DVT and/or PE) within 90 days of discharge.A multi-variate Cox regression model and a best in class of a set of 5 different ML models (support-vector machine, random-forest, naives Bayes, logistic regression, neural network]) were used to predict VTE using models applied a) at 24 hours of injury date or b) on day of patient discharge. RESULTS/ANTICIPATED RESULTS: Among 393 trauma pts (ISS=12.0, hospital LOS=4.0 days, age=48 years, 71% male, 96% with blunt mechanism, mortality 2.8%), 36 developed inpatient VTE and 36 developed VTE after discharge. In a weighted, multivariate Cox model, any type of surgery by day 1, increased age per 10 years, and BMI per 5 points were predictors of overall symptomatic VTE (C-stat 0.738). Prophylactic IVC filter placement (4.40), increased patient age per 10 years, and BMI per 5 points were predictors of post-discharge symptomatic VTE (C-stat= 0.698). A neural network ML model predicted VTE by day 1 with accuracy and AUC of 0.82 and 0.76, with performance exceeding those of a Cox model. A naīve Bayesian ML model predicted VTE at discharge, with accuracy and AUC of 0.81 and 0.77 at time of discharge, with performance exceeding those of a Cox model. DISCUSSION/SIGNIFICANCE: The rate of inpatient and post-discharge VTEs remain high. Limitations: single institution study, limited number of patients, internal validation only, with the use of limited number of ML models. We developed and internally validated a ML based tool.Future work will focus on external validation and expansion of ML techniques.