•  
  •  
 

Abstract

This study, carried out over six months at an Indonesian hospital, explores the benefits of standardizing medical record data and integrating health information systems for healthcare delivery. Utilizing a quantitative research approach, it focuses on the impact of precise data mining extraction on data analysis and the advantages of an integrated system for accessing patient records. Advanced data mining methods were employed for feature extraction, selection, and dataset reduction to enhance data classification accuracy. Findings revealed a direct correlation between the accuracy of data extraction and the reliability of data classification, highlighting the significant role of dataset reduction in improving analysis precision. The introduction of the XRP Model, a new predictive tool for assessing disease likelihood, marked a notable advancement, demonstrating high accuracy rates in predicting diabetes and heart disease (96.8% and 88%, respectively). The model's consistent performance across various outcome scenarios underscores its potential in healthcare decision-making. This research evidences the value of advanced data mining and dataset reduction in refining data classification, thus facilitating better healthcare decisions. The XRP Model's success in disease prediction suggests considerable benefits for healthcare services, offering insights crucial for the development and optimization of health information systems. These findings have the potential to influence healthcare policy and practice, advocating for a new standard in healthcare data management.

Keywords

Disease prediction, Feature reduction, Feature selection, Machine learning, Medical record

Subject Area

Chemistry

First Page

63

Last Page

71

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Receive Date

9-27-2023

Revise Date

3-15-2024

Accept Date

3-17-2024

Download

Share

COinS