Machine Learning Applications in Healthcare: Predictive Models for Disease Diagnosis
Development and Validation of ML Models for Early Disease Detection
Penulis: Prof. Dr. Mada Rahman, S.Kom., M.T., Dr. Siti Nurhaliza, S.T., M.Kom., Dr. Yuki Tanaka, Ph.D.
Penulis Korespondensi: Prof. Dr. Mada Rahman, S.Kom., M.T.
Jurnal: Journal of Biomedical Informatics
Penerbit: Elsevier
Abstrak
Abstract
This study presents the development and validation of machine learning models for predictive disease diagnosis in healthcare settings. We compare multiple algorithms including deep learning, ensemble methods, and traditional statistical approaches.
Methodology:
We developed predictive models using electronic health records from 50,000 patients. Algorithms tested include Random Forest, SVM, Neural Networks, and Gradient Boosting. Models were validated using 10-fold cross-validation and tested on independent datasets.
Results:
- Overall prediction accuracy: 94.2%
- Early detection rate improved by 35%
- False positive rate reduced to 3.1%
- Model deployment reduced diagnosis time by 60%
The study demonstrates the potential of ML models in improving healthcare outcomes through early disease detection.
Kata Kunci
machine learning, healthcare, predictive models, disease diagnosis, medical informatics
Bidang Subjek
Medical Informatics
Informasi Tambahan
Jurnal:
Lihat Jurnal
Machine Learning, Healthcare, Disease Diagnosis
Kategori: Jurnal Internasional
Diterbitkan: 29/12/2025 01:22:49