ABSTRACT:

Electronic Health Records (EHR) have become a cornerstone in modern healthcare, providing comprehensive patient data that can significantly enhance clinical decision-making and patient outcomes. However, the vast amount of data stored in EHR systems poses challenges for effective utilization in predictive analytics. This paper explores the application of machine learning (ML) techniques to predict patient survival status using data from EHR systems. We compare the performance of four classifiers: Random Forest, XGBoost, K-Nearest Neighbors (KNN), and Decision Tree.

The dataset used in this study includes patient demographics, tumor characteristics, biopsy results, treatment details, and survival status. After preprocessing and feature selection, the models are trained and evaluated on their accuracy in predicting survival. Our results indicate that the XGBoost classifier achieves the highest accuracy, making it the most suitable model for this task.
Furthermore, we developed a graphical user interface (GUI) using Tkinter, which allows healthcare professionals to input patient data and receive real-time survival predictions. This tool aims to assist clinicians in making informed decisions quickly and effectively.

INTRODUCTION:

The integration of Electronic Health Records (EHR) in modern healthcare systems has significantly transformed patient care by providing comprehensive and accurate health information. EHR systems store vast amounts Includes details about patients, which includes demographics, medical information, assessment, plans of care, dates of immunizations allergies, radiography photographs, and the results of tests performed in laboratories.

These records are vital for clinicians assist in improving patient outcomes and make knowledgeable decisions. However, the sheer volume and complexity of data present challenges in efficiently utilizing this information for predictive analytics and personalized treatment.
Predictive modeling in healthcare aims to analyze historical data to predict future outcomes. One critical application is predicting patient survival rates, which can guide treatment plans and resource allocation. Machine learning (ML) techniques, including ensemble methods and decision trees, have shown promise in making accurate predictions by learning patterns and correlations within the data.