Identifikasi Diabetes Mellitus Melalui Urine Menggunakan Deret Sensor Gas dan Metode Probabilistic Neural Network

Devi, Intan Rolystiana (2025) Identifikasi Diabetes Mellitus Melalui Urine Menggunakan Deret Sensor Gas dan Metode Probabilistic Neural Network. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6022231041-Master_Thesis.pdf Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

Diabetes Mellitus merupakan salah satu penyakit kronis yang belum dapat disembuhkan. Diabetes Mellitus disebabkan oleh tingginya kadar glukosa dalam urine yang tidak diserap oleh ginjal. Jika penyakit ini tidak segera diatasi, maka akan menimbulkan komplikasi dalam tubuh bahkan dapat mengancam nyawa. Pemeriksaan secara invasive dengan mengambil sampel darah menimbulkan rasa nyeri dan tidak nyaman pada penderita, sehingga perlu adanya terobosan untuk beralih ke pemeriksaan secara non invasive, salah satunya dengan menggunakan sampel urine yang relatif lebih cepat, aman dan nyaman. Pada penelitian ini dikembangkan sistem untuk identifikasi diabetes mellitus menggunakan tujuh sensor gas semikonduktor dan Arduino Mega sebagai mikrokontroller untuk mengubah nilai sinyal analog menjadi data digital yang dikirimkan ke komputer. Metode Probabilistic Neural Network (PNN) digunakan untuk mengenali pola dan mengidentifikasi urine sehat dan diabetes. Hasil percobaan akhir secara realtime menunjukkan bahwa metode PNN dapat mengidentifikasi urine kategori sehat dan diabetes dengan tingkat keberhasilan 87,5%.
==================================================================================================================================
Diabetes Mellitus is a chronic disease for which there is currently no cure. This condition is primarily characterized by elevated glucose levels in the urine that are not absorbed by the kidneys. If left untreated, this disease can lead to severe complications and life-threatening. Invasive diagnostic procedures, such as blood sample collection, cause pain and discomfort for patients. Consequently, there is a need for breakthroughs transition to non-invasive examinations, with urine samples offering a relatively faster, safer, and more convenient alternative. In this study, a system was developed for Diabetes Mellitus identification, utilizing seven semiconductor gas sensors and an Arduino Mega as a microcontroller. Arduino converts analog signal values into digital data, then transmitted to a computer. The Probabilistic Neural Network (PNN) method was employed for pattern recognition and identification of healthy and diabetic urine samples. The final real-time experimental results demonstrated that the PNN method can accurately classify healthy and diabetic urine with a success rate of 87.5%.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Penyakit, Urine, Diabetes Mellitus, Probabilistic Neural Network, Sensor Gas, disease, urine, diabetes mellitus, gas sensor, probabilistic neural network
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Intan Rolystiana Devi
Date Deposited:	23 Jul 2025 00:51
Last Modified:	23 Jul 2025 00:51
URI:	http://repository.its.ac.id/id/eprint/120568

Actions (login required)

View Item