Deteksi Dini Risiko Kesehatan Ibu Hamil Menggunakan Algoritma XGBoost dan Analisis Feature Importance
Keywords:
Deteksi Dini, Kehamilan, Machine Learning, XGBoost, Feature ImportanceAbstract
Risiko kesehatan pada ibu hamil merupakan salah satu kondisi yang perlu dideteksi sejak dini karena berkaitan dengan kesehatan ibu dan janin. Penelitian ini bertujuan untuk membangun model klasifikasi risiko kesehatan ibu hamil menggunakan algoritma Extreme Gradient Boosting (XGBoost) berdasarkan data kondisi kesehatan ibu hamil. Dataset yang digunakan terdiri dari 12 variabel klinis, yaitu Usia (Age), Tekanan Darah Sistolik (Systolic Blood Pressure), Tekanan Darah Diastolik (Diastolic Blood Pressure), Kadar Gula Darah (Blood Sugar), Suhu Tubuh (Body Temperature), Indeks Massa Tubuh (Body Mass Index), Riwayat Komplikasi (Previous Complications), Riwayat Diabetes Sebelumnya (Preexisting Diabetes), Diabetes Kehamilan (Gestational Diabetes), Kesehatan Mental (Mental Health), Denyut Jantung (Heart Rate), Dan Kategori Risiko (Risk Level). Target Klasifikasi Terdiri Dari Dua Tingkat Risiko Kehamilan, Yaitu Low Risk (Risiko Rendah) Dan High Risk (Risiko Tinggi). Tahapan penelitian meliputi pengumpulan data, preprocessing data, pembagian dataset, pemodelan menggunakan XGBoost, evaluasi model, dan analisis feature importance. Hasil penelitian menunjukkan bahwa model XGBoost menghasilkan performa klasifikasi yang sangat baik dengan accuracy sebesar 99.16%, precision sebesar 0.979, recall sebesar 0.986, dan F1-Score sebesar 0.99. Hasil analisis feature importance menunjukkan bahwa variabel Blood Sugar, Gestational Diabetes, dan Preexisting Diabetes merupakan faktor yang paling berpengaruh terhadap tingkat risiko kesehatan ibu hamil. Berdasarkan hasil penelitian, algoritma XGBoost layak digunakan sebagai bagian dari sistem deteksi dini risiko kesehatan ibu hamil untuk membantu proses pengambilan keputusan dan pencegahan komplikasi secara lebih cepat dan tepat.
References
[1]B. P. Statistik, “Profil Kesehatan Ibu dan Anak 2024,” Badan Pusat Statistik, Jakarta, 2024. [Daring]. Tersedia pada: https://www.bps.go.id/id/publication/2024/12/31/a919c55a72b74e33d011b0dc/profil-kesehatan-ibu-dan-anak-2024.html?utm_source=copilot.com
[2]S. T. Salsabila, R. Widiasih, dan L. Mamuroh, “Jurnal Keperawatan Karya Bhakti DETERMINAN TERJADINYA KOMPLIKASI KEHAMILAN : ANALISIS DATA INDONESIAN DEMOGRAPHIC AND HEALTH SURVEYS ( IDHS ) 2017 Pendahuluan,” J. Keperawatan Karya Bhakti, vol. 11, no. 1, hal. 67–79, 2025, [Daring]. Tersedia pada: https://journal.akperkbn.com/index.php/jkkb/article/view/194
[3]M. R. Sarker, A. E. Skeith, H. Bacheller, A. B. Caughey, dan A. M. Valent, “585: Impact of delayed initiation of prenatal care on maternal outcomes,” Am. J. Obstet. Gynecol., vol. 220, no. 1, hal. S388–S389, 2019, doi: 10.1016/j.ajog.2018.11.607.
[4]W. Apriliah, I. Kurniawan, M. Baydhowi, dan T. Haryati, “Prediksi Kemungkinan Diabetes pada Tahap Awal Menggunakan Algoritma Klasifikasi Random Forest,” Sist. Sist. Inf., vol. 10, hal. 163–171, 2021, [Daring]. Tersedia pada: https://sistemasi.ftik.unisi.ac.id/index.php/stmsi/article/view/1129/355
[5]S. Dharmawan, V. Fernandes, dan H. Halim, “PREDIKSI SERANGAN JANTUNG DENGAN MENGGUNAKAN METODE LOGISTIC REGRESSION CLASSIFIER DAN ADABOOST,” Comput. J. Comput. Sci. Inf. Syst., vol. 8, no. 1, hal. 96–103, 2024, doi: https://doi.org/10.24912/computatio.v8i1.15176.
[6]Z. Dafianto, Y. Yunefri, dan A. Digdaya, “Model Klasifikasi Kesehatan Karyawan Berbasis Naïve Bayes Classifier dan K-Nearest Neighbor,” in Prosiding-Seminar Nasional Teknologi Informasi & Ilmu Komputer (SEMASTER), 2024, vol. 3, no. 1, hal. 277–288. [Daring]. Tersedia pada: https://journal.unilak.ac.id/index.php/Semaster/article/view/24126
[7]T. Chen dan C. Guestrin, “XGBoost : A Scalable Tree Boosting System,” Washinton, 2016. doi: https://doi.org/10.1145/2939672.2939785.
[8]H. Kaneko, “Interpretation of Machine Learning Models for Data Sets with Many Features Using Feature Importance,” ACS OMEGA, vol. 8, no. 25, hal. 23218–23225, 2023, doi: 10.1021/acsomega.3c03722.
[9]M. N. Zanis dan G. Ghufron, “IMPLEMENTASI METODE NAIVE BAYES UNTUK KLASIFIKASI RISIKO PENYAKIT HIPERTENSI,” J. Nusant. Eng., vol. 8, no. 2, hal. 409–416, 2025, [Daring]. Tersedia pada: https://ojs.unpkediri.ac.id/index.php/noe/article/view/25207
[10]A. D. A. S. Surya dan Y. Yamasari, “Penerapan Algoritma Naïve Bayes ( NB ) untuk Klasifikasi Penyakit Jantung,” J. Informatics Comput. Sci., vol. 05, no. 3, hal. 447–455, 2024, doi: https://doi.org/10.26740/jinacs.v5n03.p447-455.
[11]V. Jackins, S. Vimal, M. Kaliappan, dan M. Y. Lee, “AI‑based smart prediction of clinical disease using random forest classifier and Naive Bayes,” J. Supercomput., vol. 77, no. 5, hal. 5198–5219, 2021, doi: 10.1007/s11227-020-03481-x.
[12]K. J. Olowe, N. L. Edoh, S. J. C. Zouo, dan J. Olamijuwon, “Comprehensive review of logistic regression techniques in predicting health outcomes and trends,” World J. Adv. Pharm. Life Sci., vol. 7, no. 2, hal. 016–026, 2024, [Daring]. Tersedia pada: https://zealjournals.com/wjapls/content/comprehensive-review-logistic-regression-techniques-predicting-health-outcomes-and-trends
[13]Z. Rafie, M. S. Talab, B. E. Z. Koor, A. Garavand, C. Salehnasab, dan M. Ghaderzadeh, “Leveraging XGBoost and explainable AI for accurate prediction of type 2 diabetes,” BMC Public Health, vol. 25, no. 3688, hal. 1–17, 2025, doi: https://doi.org/10.1186/s12889-025-24953-w.
[14]I. Shyu, C.-F. Liu, Y.-C. Tsai, Y.-S. Ma, T.-N. Kuo, dan S. Y.- Ling, “Machine learning predictive system to predict the risk of developing pre- eclampsia,” BMJ Heal. Care Informatics, vol. 32, no. 1, hal. 1–8, 2025, doi: 10.1136/bmjhci-2024-101151.
[15]Y. Wahyuni, H. Hadiyanto, R. Sanjaya, dan N. Herdianto, “A Comparative Analysis and Interpretation of Random Forest , XGBoost , and SVM ( RBF ) Models with Explainable AI for Pregnancy Risk Classification,” SISFORMA J. Inf. Syst. (, vol. 12, no. 2, hal. 191–204, 2025, doi: 10.24167/sisforma.v12i2.14292.
[16]M. U. Mojumdar et al., “Maternal Health Risk Assessment Dataset.” Mendeley Data, 2024. doi: 10.17632/p5w98dvbbk.1.
[17]R. Mitchell dan E. Frank, “Accelerating the XGBoost algorithm using GPU computing,” PeerJ Comput. Sci., 2017, doi: 10.7717/peerj-cs.127.
[18]Y. Zhu, “Stock Price Prediction based on LSTM and XGBoost Combination Model,” Trans. Comput. Sci. Intell. Syst. Res., vol. 1, no. 1, hal. 1–16, 2023, [Daring]. Tersedia pada: https://wepub.org/index.php/TCSISR/article/view/90
[19]E. K. Sahin, “Assessing the predictive capability of ensemble tree methods for landslide susceptibility mapping using XGBoost , gradient boosting machine , and random forest,” SN Appl. Sci., vol. 2, no. 7, hal. 1–17, 2020, doi: 10.1007/s42452-020-3060-1.
[20]S. Fatima et al., “XGBoost and Random Forest Algorithms : An In- Depth Analysis,” Pakistan J. Sci. Res., vol. 3, no. 1, hal. 26–31, 2023, [Daring]. Tersedia pada: https://paas-pk.org/index.php/pjosr/article/view/1876
[21]C. A. Negrato, R. Mattar, dan M. B. Gomes, “Adverse pregnancy outcomes in women with diabetes,” Diabetol. Metab. Syndr., vol. 4, no. 41, hal. 1–6, 2012, [Daring]. Tersedia pada: https://pmc.ncbi.nlm.nih.gov/articles/PMC3514247/
[22]M. L. Geurtsen, E. E. L. Van Soest, E. Voerman, E. A. P. Steegers, V. W. V Jaddoe, dan R. Gaillard, “High maternal early-pregnancy blood glucose levels are associated with altered fetal growth and increased risk of adverse birth outcomes,” Diabetologia, vol. 62, hal. 1880–1890, 2019, [Daring]. Tersedia pada: https://pubmed.ncbi.nlm.nih.gov/31392381/
[23]F. K. Adli, “DIABETES MELITUS GESTASIONAL: DIAGNOSIS DAN FAKTOR RISIKO,” J. Med. Hutama, vol. 03, no. 01, hal. 1545–1551, 2021, [Daring]. Tersedia pada: https://jurnalmedikahutama.com/index.php/JMH/article/view/312
[24]O. Zapata-cortes, M. D. Arango-Serna, Andres, J. A. Zapata-Cortes, dan J. A. Restrepo-Carmona, “Machine Learning Models and Applications for Early Detection,” Sensors, vol. 24, no. 14, hal. 1–20, 2024, [Daring]. Tersedia pada: https://www.mdpi.com/1424-8220/24/14/4678
[25]R. Islam, M. Kabir, M. F. Mridha, S. Alfarhood, M. Safran, dan D. Che, “Deep Learning-Based IoT System for Remote Monitoring and Early Detection of Health Issues in Real-Time,” Sensors, vol. 23, no. 11, hal. 1–18, 2023, [Daring]. Tersedia pada: https://www.mdpi.com/1424-8220/23/11/5204
[26]M. M. Ahsan, S. A. Luna, dan Z. Siddique, “Machine-Learning-Based Disease Diagnosis: A Comprehensive Review,” vol. 10, no. 3, hal. 1–30, 2022, [Daring]. Tersedia pada: https://www.mdpi.com/2227-9032/10/3/541
