الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 165 |  |  |
| | | from | 165 | | |
Abstract
Health risk prediction plays a crucial role in healthcare by leveraging advanced machine learning techniques to analyze large datasets and identify individuals at higher risk for developing certain health conditions. By utilizing a combination of patient data, including medical history, genetic information, lifestyle factors, and biomarkers, predictive models can generate personalized risk scores. These risk scores help healthcare professionals prioritize interventions and preventive measures, enabling early detection, timely treatment, and improved patient outcomes. Moreover, health risk prediction empowers individuals to make informed decisions about their health, leading to proactive measures and lifestyle modifications that can mitigate potential risks and promote overall well-being. Recent studies have proven that data analytics may assist in predicting events before they occur, which may impact the outcome of current situations. In the medical sector, it can be utilized for predicting the likelihood of getting a health condition such as chronic kidney disease (CKD) or Ischemic stroke outcome. In an attempt to save the lives of many patients, this research was divided into two parts: the first part focused on discovering chronic kidney disease before it occurs, while the second part focused on discovering the outcome of an ischemic stroke. For the first part, this research aims to develop a CKD prediction framework, which forecasts CKD occurrence over a specific time using deep learning and deep ensemble learning approaches. While a great deal of research focuses on disease detection, few studies contribute to disease prediction before it may occur. However, the performance of previous work was not competitive. This research tackles the under-explored area of early CKD prediction through a high-performing deep learning and ensemble framework. We bridge the gap between existing detection methods and preventive interventions by developing and comparing deep learning models like CNN, LSTM, and LSTM-BLSTM for 6–12-month CKD prediction; addressing data imbalance, feature selection, and optimizer optimization; and building an ensemble model combining the best individual models (CNN-Adamax, LSTM-Adam, and LSTM-BLSTM-Adamax). As for the second part, Ischemic stroke was considered as it can cause damage to health, leading to a loss of brain functions such as speech, movement, perception, and memory, and can result in disability or death if not diagnosed and treated promptly. Early prediction of the outcome of the first stroke, such as disability or death, can help many patients by administering appropriate medications to save their lives. Additionally, early prediction of a recurrent stroke within 14 days of the initial stroke can contribute to preventing its recurrence. This research first proposes a modified Manta-Ray Foraging Optimizer (MMRFO) to enhance the characteristics of the MRFO technique. This technique is based on inserting the Hill Climbing technique into the original MRFO to enhance the exploitation phase, which is responsible for finding the promising zone in the search area. After that, the developed method is used to establish the best hyperparameters of the Vision Transformer (ViT) model to predict the outcome of the stroke before its occurrence. The ASCII encoder method is proposed to convert the categorical data to numerical values. Harris Hawk Optimization technique (HHO) is applied in the feature selection phase to select the most vital features that can characterize the stroke. A comparative study has been performed to confirm the effectiveness of the proposed methodology.