Prediksi Kelayakan Donor Berdasarkan Riwayat Kesehatan Berbasis Algoritma Machine Learning
DOI:
https://doi.org/10.25139/htc.v8i2.11204Keywords:
Kelayakan Donor, Machine Learning, Random Forest, Dataset Sintesis, Unit Transfusi darahAbstract
Ketersediaan darah yang aman merupakan komponen krusial dalam pelayanan kesehatan. Proses penentuan kelayakan donor darah secara manual di Unit Transfusi Darah (UTD) masih bergantung pada penelitian subjektif dan pemeriksana fisiologis sederhana yang rentan terhadap kesalahan manusia. Pendekatan berbasis Machine Learning (ML) dapatn menstandarkan dan mengobjektifkan proses ini. Peneitian ini bertujuan mengevaliasi kinerja algoritma ML dalam memprediksikan kelayakan donor berdasarkan dataset simulasi. Tiga model klasifikasi yaitu, Decision Tree (DT), Random Forest (RF), dan Logistic Regression (LR) dilatih menggunakan dataset sintetis yang terdiri dari 2.000 data calon donor yang mencakup parameter fisiologis dan riwayat kesehatan. Hasil evaluasi menunjukkan bahwa model Random Forest mencapai performa 92%, diikuti Decision Tree 87%, dan Logistic Regression 83%. Analisis feature importance menegaskan bahwa kadar hemoglobin, riwayat penyakit, dan status penggunaan obat adalah variabel paling dominan yang mempengaruhi kelayakn donor. Hal ini sesuai dengan pedoman skrining dari WHO dan PMI. Temuan ini memberikan bukti konsep bahwa sistem berbasis ML dapat menjadi decision support system yang efisien dan akurat untuk mendukung proses skrining awal di UTD. Penggunaan data simulasi berhasil memitigasi kendala etika dan privasi data riil, sekaligus membuka jalan bagi penelitian lanjutan menggunakan data lapangan.
References
[2] World Health Organization (WHO), Blood Donor Selection: Guidelines on Assessing Donor Suitability for Blood Donation. Geneva, Switzerland: WHO, 2020.
[3] Palang Merah Indonesia (PMI), Standar Nasional Unit Transfusi Darah Indonesia. Jakarta, Indonesia: PMI, 2021.
[4] L. Zhang et al., “Predictive analytics for blood donor eligibility using machine learning algorithms,” Transfus. Med. Rev., vol. 37, no. 1, pp. 12–24, 2023.
[5] J. Wu et al., “Predicting willingness to donate blood based on machine learning: Two donor recruitments during COVID-19,” BMC Med. Inform. Decis. Mak., vol. 22, p. 136, 2022.
[6] V. Shelake et al., “Predictive modeling for blood donor eligibility using ensemble machine learning techniques,” Int. J. Adv. Sci. Eng. Inf. Technol., vol. 15, no. 1, pp. 100–108, 2025. (Disintesis dari IJASEIT, 2024 dan Shelake et al., 2025)
[7] T. Budi and R. Handayani, “Penerapan algoritma Decision Tree untuk prediksi kelayakan donor darah di PMI,” J. Teknol. Kesehatan, vol. 10, no. 2, pp. 45–53, 2022.
[8] G. Tadesse et al., “Predicting donor suitability using physiological parameters and ML techniques,” Comput. Biol. Med., vol. 150, pp. 106085, 2022.
[9] M. H. Vinkenoog et al., “Explainable haemoglobin deferral predictions using machine learning models: Interpretation and consequences for the blood supply,” Vox Sang., vol. 117, no. 8, pp. 896–908, 2022.
[10] D. Sharma et al., “Machine learning in blood donation management: Opportunities and challenges,” Health Inform. J., vol. 29, no. 3, pp. 1460–1475, 2023.
[11] J. Liu and X. Chen, “Ethical implications of synthetic data in healthcare research,” BMC Med. Ethics, vol. 22, no. 1, p. 65, 2021.
[12] P. Kaur and A. Singh, “Synthetic data generation for medical decision support systems,” Artif. Intell. Med., vol. 118, pp. 102135, 2021.
[13] Y. Zhao et al., “Decision tree and logistic regression in medical prediction: A comparative study,” J. Data Sci., vol. 20, no. 4, pp. 512–528, 2022.
[14] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, 2nd ed. New York, NY, USA: Springer, 2021.
[15] M. A. Rahman et al., “Application of machine learning in health diagnostics: A review,” J. Biomed. Inform., vol. 130, pp. 104045, 2022.
[16] S. García et al., “Machine learning experimental design in healthcare,” Expert Syst. Appl., vol. 141, pp. 112957, 2020.
[17] N. V. Chawla and D. A. Davis, “Bringing big data to personalized healthcare: Machine learning and applications,” IEEE Intell. Syst., vol. 35, no. 2, pp. 70–76, 2020.
[18] S. Bhattacharya et al., “AI-driven frameworks for healthcare management,” Front. Artif. Intell., vol. 4, 2021, Art. no. 657903.
