ABSTRACT
Cardiovascular diseases (CVDs) remain a major global health challenge and a leading cause of mortality, highlighting the need for improved predictive models. We introduce an innovative agent-based dynamic simulation technique that enhances our AI models' capacity to predict CVD progression. This method simulates individual patient responses to various cardiovascular risk factors, improving prediction accuracy and detail. Also, by incorporating an ensemble learning model and interface of web application in the context of CVD prediction, we developed an AI dashboard-based model to enhance the accuracy of disease prediction and provide a user-friendly app. The performance of traditional algorithms was notable, with Ensemble learning and XGBoost achieving accuracies of 91% and 95%, respectively. A significant aspect of our research was the integration of these models into a streamlit-based interface, enhancing user accessibility and experience. The streamlit application achieved a predictive accuracy of 97%, demonstrating the efficacy of combining advanced AI techniques with user-centered web applications in medical prediction scenarios. This 97% confidence level was evaluated by Brier score and calibration curve. The design of the streamlit application facilitates seamless interaction between complex ML models and end-users, including clinicians and patients, supporting its use in real-time clinical settings. While the study offers new insights into AI-driven CVD prediction, we acknowledge limitations such as the dataset size. In our research, we have successfully validated our predictive proposed methodology against an external clinical setting, demonstrating its robustness and accuracy in a real-world fixture. The validation process confirmed the model's efficacy in the early detection of CVDs, reinforcing its potential for integration into clinical workflows to aid in proactive patient care and management. Future research directions include expanding the dataset, exploring additional algorithms, and conducting clinical trials to validate our findings. This research provides a valuable foundation for future studies, aiming to make significant strides against CVDs.
ABSTRACT
Early pregnancy loss (EPL) is a prevalent health concern with significant implications globally for gestational health. This research leverages machine learning to enhance the prediction of EPL and to differentiate between typical pregnancies and those at elevated risk during the initial trimester. We employed different machine learning methodologies, from conventional models to more advanced ones such as deep learning and multilayer perceptron models. Results from both classical and advanced machine learning models were evaluated using confusion matrices, cross-validation techniques, and analysis of feature significance to obtain correct decisions among algorithmic strategies on early pregnancy loss and the vitamin D serum connection in gestational health. The results demonstrated that machine learning is a powerful tool for accurately predicting EPL, with advanced models such as deep learning and multilayer perceptron outperforming classical ones. Linear discriminant analysis and quadratic discriminant analysis algorithms were shown to have 98 % accuracy in predicting pregnancy loss outcomes. Key determinants of EPL were identified, including levels of maternal serum vitamin D. In addition, prior pregnancy outcomes and maternal age are crucial factors in gestational health. This study's findings highlight the potential of machine learning in enhancing predictions related to EPL that can contribute to improved gestational health outcomes for mothers and infants.
ABSTRACT
This research project explored into the intricacies of road traffic accidents severity in the UK, employing a potent combination of machine learning algorithms, econometric techniques, and traditional statistical methods to analyse longitudinal historical data. Our robust analysis framework includes descriptive, inferential, bivariate, multivariate methodologies, correlation analysis: Pearson's and Spearman's Rank Correlation Coefficient, multiple logistic regression models, Multicollinearity Assessment, and Model Validation. In addressing heteroscedasticity or autocorrelation in error terms, we've advanced the precision and reliability of our regression analyses using the Generalized Method of Moments (GMM). Additionally, our application of the Vector Autoregressive (VAR) model and the Autoregressive Integrated Moving Average (ARIMA) models have enabled accurate time series forecasting. With this approach, we've achieved superior predictive accuracy and marked by a Mean Absolute Scaled Error (MASE) of 0.800 and a Mean Error (ME) of -73.80 compared to a naive forecast. The project further extends its machine learning application by creating a random forest classifier model with a precision of 73%, a recall of 78%, and an F1-score of 73%. Building on this, we employed the H2O AutoML process to optimize our model selection, resulting in an XGBoost model that exhibits exceptional predictive power as evidenced by an RMSE of 0.1761205782994506 and MAE of 0.0874235576229789. Factor Analysis was leveraged to identify underlying variables or factors that explain the pattern of correlations within a set of observed variables. Scoring history, a tool to observe the model's performance throughout the training process was incorporated to ensure the highest possible performance of our machine learning models. We also incorporated Explainable AI (XAI) techniques, utilizing the SHAP (Shapley Additive Explanations) model to comprehend the contributing factors to accident severity. Features such as Driver_Home_Area_Type, Longitude, Driver_IMD_Decile, Road_Type, Casualty_Home_Area_Type, and Casualty_IMD_Decile were identified as significant influencers. Our research contributes to the nuanced understanding of traffic accident severity and demonstrates the potential of advanced statistical, econometric, machine learning techniques in informing evidence based interventions and policies for enhancing road safety.
