Implementasi Knowledge-based Decision Making untuk Memprediksi Survival Rate Pasien Pediatri Continuous Renal Replacement Therapy

Ilhama, Zulfan Zidni (2025) Implementasi Knowledge-based Decision Making untuk Memprediksi Survival Rate Pasien Pediatri Continuous Renal Replacement Therapy. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5026211099-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Continuous Renal Replacement Therapy (CRRT) merupakan terapi pengganti fungsi ginjal yang digunakan untuk pasien pediatri kritis di ruang Pediatric Intensive Care Unit (PICU). Tingkat mortalitas pasien pediatri yang menjalani CRRT mencapai 60%, sehingga diperlukan sistem pendukung keputusan medis yang mampu memprediksi survival rate secara cepat dan obyektif. Penelitian ini mengembangkan sistem prediksi survival rate pasien pediatri CRRT menggunakan pendekatan Knowledge-Based Decision Making (KBDM). Sistem mengintegrasikan pengetahuan eksplisit dari meta-analisis jurnal medis ke dalam logika inferensi yang mempertimbangkan variabel klinis kuantitatif dan kualitatif. Metode weighted average digunakan untuk menentukan batas referensi nilai survivor dan non-survivor. Aplikasi berbasis web dikembangkan menggunakan framework Streamlit dengan fitur prediksi survival rate, autentikasi, koneksi database, dan fitur unduh data. Hasil pengujian membuktikan bahwa aplikasi berjalan dengan baik dan optimal berdasarkan aspek fungsionalitas, reliabilitas, skalabilitas, dan kompatibilitas. Dari segi fungsionalitas semua fitur dapat diakses tanpa adanya bug, dari segi reliabilitas skor prediksi terbukti konsisten, dari segi skalabilitas aplikasi dapat menangani multiuser secara stabil, dan aplikasi kompatibel untuk diakses melalui berbagai perangkat. Dengan demikian, sistem prediksi survival rate pasien pediatri Continuous Renal Replacement Therapy berbasis Knowledge-Based Decision Making ini berhasil diimplementasikan ke dalam aplikasi web.
===================================================================================================================================
Continuous renal replacement therapy (CRRT) is a form of renal replacement therapy used for critically ill paediatric patients in the paediatric intensive care unit (PICU). As the mortality rate for paediatric patients undergoing CRRT is as high as 60%, there is a need for a medical decision support system that can quickly and objectively predict survival rates. This study has developed such a system for predicting the survival rates of paediatric CRRT patients, using a Knowledge-Based Decision Making (KBDM) approach. This system incorporates explicit knowledge from medical journal meta-analyses into an inference logic that considers both quantitative and qualitative clinical variables. The weighted average method was used to determine the reference limits of survivor and non-survivor values. A web-based application was developed using the Streamlit framework, incorporating features such as survival rate prediction, authentication, database connection and data download. Test results prove that the application runs optimally in terms of functionality, reliability, scalability and compatibility. All features can be accessed without any bugs; the prediction score is consistent; the application can handle multiple users stably; and the application can be accessed via various devices. Thus, the survival rate prediction system for paediatric continuous renal replacement therapy based on knowledge-based decision making has been successfully implemented as a web application.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Continuous Renal Replacement Therapy, Knowledge-Based Decision Making, Survival Rate, Pediatri, Sistem Pendukung Keputusan, Aplikasi Web, Paediatrics, Decision Support System.
Subjects:	R Medicine > RJ Pediatrics
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Zulfan Zidni Ilhama
Date Deposited:	28 Jul 2025 05:12
Last Modified:	28 Jul 2025 05:12
URI:	http://repository.its.ac.id/id/eprint/122609

Actions (login required)

View Item