Deteksi Dini Karies Gigi Dengan Metode Fluoresen Optik Pada Saliva

Mayerd, Immanuel (2023) Deteksi Dini Karies Gigi Dengan Metode Fluoresen Optik Pada Saliva. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 07111940000147-Undergraduate_Thesis.pdf] Text
07111940000147-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2025.

Download (3MB) | Request a copy

Abstract

Menurut Riset Kesehatan Dasar Nasional, dari populasi penduduk Indonesia tahun 2018 penderita penyakit karies mencapai 80%, dengan 90% lebih penderitanya adalah anak-anak. Tindakan pencegahan berupa deteksi dini sebelum karies semakin parah sangatlah penting. Metode deteksi dini sekarang ini hanya tersedia di rumah sakit dan tempat praktik. Selain itu, metode-metode tersebut hanya bisa dilakukan oleh tenaga profesional (dokter gigi). Penelitian ini menawarkan ide dengan menggunakan metode Fluoresen Optik. Prinsipnya adalah dengan memanfaatkan biomarker Matriks Metalloproteinase-8 (MMP-8) pada saliva, yang dapat bereaksi secara inhibit dengan kurkumin yang memiliki sifat fluoresen. Kandungan MMP-8 akan semakin bertambah dengan bertambahnya jumlah karies gigi ataupun tingkat keparahannya. Dengan menggunakan kurkumin yang bersifat fluoresen, reaksi antara kurkumin dan MMP-8 akan dimanfaatkan. Prinsipnya adalah sampel yang berupa campuran saliva dan kurkumin akan dikenai cahaya UV yang kemudian akan menghasilkan emisi cahaya fluoresen yang dapat diakuisisi oleh sensor AS7262 untuk kemudian dikirim ke komputer untuk dianalisis dan dilatih tiga jenis algoritma yakni ANN, KNN dan SVM. Klasifikasi yang digunakan menggunakan standar American Dental Association (ADA) dan klasifikasi biner (positif/negatif). Model dengan akurasi terbaik dihasilkan oleh algoritma dengan klasifikasi positif/negatif. Model ANN, KNN dam SVM untuk klasifikasi biner memiliki akurasi validasi yang masing-masing nilainya adalah 90%, 80% dan 60%. Sensitivitas pengklasifikasian pada metode ini cukup baik dibandingkan dengan metode konvensional yang ada, namun masih ada ruang untuk meningkatkan hasil dari metode ini.
================================================================================================================================
According to the National Basic Health Research, 80% of Indonesia's population suffers from caries disease in 2018, with more than 90% of sufferers being children. Preventive action in the form of early detection before caries gets worse is critical. Early detection methods are currently only available in hospitals and doctor's practices. In addition, these methods can only be done by professionals (dentists). This study offers an idea using the Optical Fluorescent method. The principle is to utilize Matrix Metalloproteinase-8 (MMP-8) biomarkers in saliva, which can react inhibitive with curcumin which has fluorescent properties. The content of MMP-8 will increase with the increasing number of dental caries or their severity. By using curcumin which is fluorescent, the reaction between curcumin and MMP-8 will be utilized. The principle is that a sample which is a mixture of saliva and curcumin will be exposed to UV light which will then produce fluorescent light emission, which can be acquired by the AS7262 sensor and then sent to a computer to be analyzed and trained on three types of algorithms namely ANN, KNN and SVM. The classification uses the American Dental Association (ADA) standard and binary classification (positive/negative). The model with the best accuracy is generated by an algorithm with a positive/negative classification. The ANN, KNN, and SVM models for binary classification have validation accuracy of 90%, 80%, and 60%, respectively. The classification sensitivity of this method is quite good compared to existing conventional methods, but there is still room to improve the results of this method.

Item Type: Thesis (Other)
Uncontrolled Keywords: Fluoresen, Karies, Kurkumin, MMP-8, Caries, Curcumin, Fluorescent
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Q Science > QD Chemistry > QD96F56 Fluorescence spectroscopy
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User: Mayerd Immanuel
Date Deposited: 25 Jul 2023 05:44
Last Modified: 25 Jul 2023 05:44
URI: http://repository.its.ac.id/id/eprint/99369

Actions (login required)

View Item View Item