<![CDATA[Abstract: Chronic Kidney Disease (CKD) is a disease characterized by a gradual decline in kidney function and can cause serious complications if not detected early. The disease detection process generally requires an examination of several clinical parameters so that a system is needed that can help the detection process in a faster and more structured manner. This study aims to design and build a website-based kidney failure disease detection system using the logistic regression machine learning method. The data used came from the Cut Meutia Hospital in North Aceh with eight medical parameters, namely age, blood pressure, hemoglobin, leukocytes, erythrocytes, urea, creatinine, and random blood glucose. The research stages include data collection, preprocessing, model training, evaluation using confusion matrix, and implementation of models into website-based systems using Python and Flask. The results of the study showed that the system built was able to detect kidney failure disease with an accuracy value of 95.6%, precision of 98.7%, recall of 95.1%, and F1-score of 96.9%. The results of the evaluation showed that the model had a good ability to classify patient data into CKD and normal categories with a high level of accuracy and detection ability. The system built is expected to be used as an early detection tool for kidney failure to support the initial screening process and provide information on detection results in a faster and more structured manner. Keywords: Chronic Kidney Failure, Machine Learning, Logistic Regression, Early Detection, Website. Abstrak: Penyakit Gagal Ginjal Kronis (Chronic Kidney Disease/CKD) merupakan penyakit yang ditandai dengan penurunan fungsi ginjal secara bertahap dan dapat menyebabkan komplikasi serius apabila tidak terdeteksi sejak dini. Proses deteksi penyakit umumnya memerlukan pemeriksaan beberapa parameter klinis sehingga diperlukan suatu sistem yang dapat membantu proses deteksi secara lebih cepat dan terstruktur. Penelitian ini bertujuan untuk merancang dan membangun sistem deteksi penyakit gagal ginjal berbasis website menggunakan machine learning metode regresi logistik. Data yang digunakan berasal dari RSU Cut Meutia Aceh Utara dengan delapan parameter medis, yaitu umur, tekanan darah, hemoglobin, leukosit, eritrosit, ureum, kreatinin, dan glukosa darah acak. Tahapan penelitian meliputi pengumpulan data, preprocessing, pelatihan model, evaluasi menggunakan confusion matrix, serta implementasi model ke dalam sistem berbasis website menggunakan Python dan Flask. Hasil penelitian menunjukkan bahwa sistem yang dibangun mampu melakukan deteksi penyakit gagal ginjal dengan nilai accuracy sebesar 95,6%, precision sebesar 98,7%, recall sebesar 95,1%, dan F1-score sebesar 96,9%. Hasil evaluasi menunjukkan bahwa model memiliki kemampuan yang baik dalam mengklasifikasikan data pasien ke dalam kategori CKD dan normal dengan tingkat ketepatan serta kemampuan deteksi yang tinggi. Sistem yang dibangun diharapkan dapat digunakan sebagai alat bantu deteksi dini penyakit gagal ginjal untuk mendukung proses skrining awal serta memberikan informasi hasil deteksi secara lebih cepat dan terstruktur. Kata Kunci: Penyakit Gagal Ginjal Kronis, Machine Learning, Regresi Logistik, Deteksi Dini, Website.]]>
