Pengembangan Microwave Sensor Untuk Deteksi Kadar Glukosa Secara Non-Invasif Pada Penderita Diabetes

Ayun, Syahdifa Aisyah Qurrata (2025) Pengembangan Microwave Sensor Untuk Deteksi Kadar Glukosa Secara Non-Invasif Pada Penderita Diabetes. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5023211032-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (19MB)

Abstract

Pemantauan kadar glukosa darah secara rutin merupakan aspek krusial dalam manajemen penyakit diabetes melitus. Sayangnya, metode pemantauan yang bersifat invasif seperti tusukan jarum secara berkala sering menimbulkan ketidaknyamanan, nyeri, serta risiko infeksi bagi pasien. Penelitian ini berfokus pada pengembangan sistem pemantauan glukosa darah secara non-invasif berbasis sensor microwave dengan pendekatan resonansi elektromagnetik. Sistem yang dikembangkan terdiri dari antena metamaterial jenis Complementary Split Ring Resonator (CSRR) yang terintegrasi dengan radar ADALM-PLUTO sebagai perangkat transmisi dan penerima gelombang, serta perangkat lunak untuk akuisisi sinyal dan pemrosesan data. Sinyal I/Q yang diperoleh dari interaksi medan elektromagnetik dengan larutan glukosa dianalisis untuk mengekstraksi fitur-fitur penting seperti frekuensi resonansi, gamma spectrum, amplitudo rata-rata, dan statistik frekuensi instan. Selanjutnya, fitur-fitur tersebut digunakan sebagai masukan model machine learning berbasis algoritma Random Forest untuk memetakan kadar glukosa darah secara prediktif. Pengujian dilakukan menggunakan larutan glukosa buatan dengan rentang konsentrasi hingga 527 mg/dL, dan hasil prediksi divalidasi menggunakan glukometer standar sebagai acuan pembanding. Hasil menunjukkan bahwa sistem mampu mendeteksi perubahan kadar glukosa melalui pergeseran frekuensi resonansi. Selain itu, perangkat lunak digunakan untuk mengotomatisasi seluruh alur pemrosesan sinyal, mulai dari ekstraksi fitur hingga klasifikasi, dengan korelasi signifikan antara beberapa fitur seperti gamma peak dan mean amplitude terhadap kadar glukosa. Dengan kemampuan ini, sistem menunjukkan potensi kuat sebagai prototipe perangkat pemantauan kadar glukosa darah yang nyaman digunakan secara non-invasif, serta menjanjikan untuk dikembangkan ke arah aplikasi klinis atau pemantauan mandiri.

===========================================================================================================================================

Routine monitoring of blood glucose levels is a crucial aspect in the management of diabetes mellitus. Unfortunately, conventional invasive monitoring methods, such as regular finger-prick tests, often cause discomfort, pain, and an increased risk of infection for patients. This study focuses on the development of a non-invasive blood glucose monitoring system based on microwave sensors using an electromagnetic resonance approach. The proposed system consists of a metamaterial antenna in the form of a Complementary Split Ring Resonator (CSRR) that integrated with the ADALM-PLUTO radar device for microwave transmission and reception, as well as a software platform for signal acquisition and data processing. The acquired I/Q signals, resulting from the interaction between electromagnetic waves and glucose solutions, are analyzed to extract essential features such as resonance frequency, gamma spectrum, average amplitude, and instantaneous frequency statistics. These features are then used as input for a machine learning model based on the Random Forest algorithm to predict glucose concentrations. The system was tested on artificial glucose solutions with concentrations up to 527 mg/dL, and the prediction results were validated using a commercial glucometer as the reference standard. The results demonstrate that the system is capable of detecting changes in glucose levels through shifts in resonance frequency. Additionally, the software component successfully automates the entire signal processing pipeline, from feature extraction to classification, and yields significant correlations between features such as gamma peak and mean amplitude with glucose concentration. With these capabilities, the system shows strong potential as a prototype for a comfortable non-invasive blood glucose monitoring device, promising for further development toward clinical applications.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Glukosa darah, non-invasif, sensor microwave, CSRR, Blood glucose, non-invasive, microwave sensor, CSRR
Subjects:	R Medicine > R Medicine (General) > R856.2 Medical instruments and apparatus. T Technology > T Technology (General) > T59.7 Human-machine systems. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK6564 Radio transmitter-receivers
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Syahdifa Aisyah Qurrata Ayun
Date Deposited:	04 Aug 2025 01:20
Last Modified:	04 Aug 2025 01:20
URI:	http://repository.its.ac.id/id/eprint/125624

Actions (login required)

View Item