Implementasi Artificial Intelligence dalam Sistem Prediksi Penyakit Berbasis Data Rekam Medis Digital

Abstract

The advancement of Artificial Intelligence (AI) has significantly impacted the healthcare sector, particularly in processing digital medical records to support faster and more accurate disease diagnosis. This study aims to analyze the implementation of AI in disease prediction systems based on digital medical record data and evaluate its effectiveness in assisting clinical decision-making. A quantitative experimental approach was employed using patient datasets that include age, gender, medical history, laboratory results, blood pressure, blood glucose levels, and other health indicators. The data were processed using machine learning algorithms, namely Decision Tree, Random Forest, and Artificial Neural Network (ANN), to compare prediction accuracy. The research stages included data collection, data cleaning, transformation, and model training using classification techniques. System performance was evaluated using confusion matrix, accuracy, precision, recall, and F1-score. The results indicate that ANN achieved the highest accuracy at 94%, followed by Random Forest at 91% and Decision Tree at 86%. The implementation of AI enhances early disease risk identification, improves healthcare efficiency, and reduces diagnostic errors. However, challenges remain, including data security, infrastructure requirements, and data quality. Therefore, sustainable development requires regulatory support, cybersecurity, and system integration.