Klasifikasi Penyakit Berdasarkan Foto Rontgen Thorax Menggunakan Metode Support Vector Machine

Abiyyi, Muhammad Zuhdi Afi (2023) Klasifikasi Penyakit Berdasarkan Foto Rontgen Thorax Menggunakan Metode Support Vector Machine. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05211940000135-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy

Abstract

Pada tahun 2019 seluruh dunia terdampak pandemi Covid-19, Covid-19 disebabkan oleh SARS-Cov-2 yang merupakan virus jenis baru dari coronavirus atau kelompok virus yang menginfeksi sistem pernafasan. Virus ini akan memberikan dampak yang berbeda-beda pada setiap orang, mulai dari infeksi tanpa gejala, penyakit ringan seperti flu, hingga penyakit kritis seperti pneumonia, atau bahkan kematian. Keragaman ini menyebabkan para dokter khususnya spesialis paru memerlukan tindakan foto rontgen sebagai penunjang dalam proses penegakan diagnosis. Dan peningkatan jumlah pasien tidak sebanding dengan jumlah tenaga kerja yang memiliki kemampuan dalam menginterpretasikan foto rontgen, karena hanya dokter spesialis radiologi yang memiliki kemampuan tersebut, bahkan dokter spesialis paru pun belum tentu bisa menginterpretasikan foto rontgen. Penelitian Tugas Akhir ini dilakukan dengan metode principle component analysis sebagai salah satu cara dalam melakukan reduksi dimensi terhadap data yang diperoleh. Selain itu, penelitian ini juga menggunakan metode support vector machine sebagai salah satu cara dalam melakukan pengklasifikasian jenis-jenis penyakit berdasarkan foto rontgent. Model SVM terbaik yang didapatkan dari percobaan pada tugas akhir adalah model SVM dengan proporsi data 80% data training, 20% data testing, dan ukuran gambar 200x200. Model tersebut mampu menghasilkan akurasi sebesar 83.65%. Hasil dari penelitian ini diharapkan dapat membantu para tenaga medis maupun pasien dalam mengetahui atau mendiagnosis hasil foto rontgennya sedini mungkin dengan hasil yang tepat dan akurat.
=================================================================================================================================
In 2019 the whole world was affected by the Covid-19 pandemic, Covid-19 was caused by SARS-Cov-2 which is a new type of coronavirus or a group of viruses that infect the respiratory system. This virus will have a different impact on each person, ranging from asymptomatic infections, minor illnesses such as the flu, to critical illnesses such as pneumonia, or even death. This diversity causes doctors, especially pulmonary specialists, to require X-rays as a support in the process of making a diagnosis. And the increase in the number of patients is not proportional to the number of workers who have the ability to interpret X-rays, because only radiology specialists have this ability, even lung specialists may not necessarily be able to interpret X-rays. This Final Project research was conducted using the principle component analysis method as a way of reducing the dimensions of the data obtained. In addition, this study also used the support vector machine method as a way of classifying the types of diseases based on X-rays. The best SVM model obtained from the experiments in the final assignment is the SVM model with the proportion of data 80% training data, 20% testing data, and 200x200 image size. This model is capable of producing an accuracy of 83.65%. The results of this study are expected to assist medical personnel and patients in knowing or diagnosing X-ray results as early as possible with precise and accurate results.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Rontgen Thorax, Principle Component Analysis, Support Vector Machine
Subjects:	T Technology > T Technology (General) T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Zuhdi Afi Abiyyi
Date Deposited:	14 Jul 2023 14:07
Last Modified:	14 Jul 2023 14:07
URI:	http://repository.its.ac.id/id/eprint/98468

Actions (login required)

View Item