Klasifikasi NSCLC dengan Arsitektur DenseNet dan GLCM Untuk Deteksi Dini Kanker Paru-Paru Pada Citra CT-Scan

Maulana, Irgi Azarya Putra (2024) Klasifikasi NSCLC dengan Arsitektur DenseNet dan GLCM Untuk Deteksi Dini Kanker Paru-Paru Pada Citra CT-Scan. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5023201023-Undergraduate_Thesis.pdf] Text
5023201023-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2026.

Download (1MB) | Request a copy

Abstract

Kanker paru-paru atau kanker pulmoner merupakan penyakin yang memiliki variasi keganasan tergantung kondisi. Kanker paru-paru terjadi ketika sel-sel di dalam paru-paru mengalami pertumbuhan yang tidak terkendali dan menjadi ganas. Ada dua jenis kanker paru-paru utama yaitu Non-Small Cell Lung Cancer (NSCLC) dan Small Cell Lung Cancer (SCLC). NSCLC adalah jenis kanker paru-paru yang paling umum, mencakup sekitar 85% dari semua kasus kanker paru-paru. NSCLC terbagi menjadi beberapa subjenis, termasuk Pulmonary Adenocarcinoma (ADC), dan Pulmonary Squamous Cell Carcinoma (SqCC). Sedangkan SCLC lebih jarang terjadi dan tumbuh lebih cepat daripada NSCLC. SCLC juga lebih cenderung menyebar ke bagian tubuh lain pada saat diagnosis dibandingkan NSCLC. Skrining menggunakan CT scan dosis rendah, yang diizinkan saat ini, seringkali memiliki tingkat sensitivitas yang rendah dan tingkat positif palsu yang tinggi. Lebih dari 90% dari hasil positif sebenarnya tidak menunjukkan adanya kanker. Selain itu, saat ini tidak ada biomarker tambahan yang dapat meningkatkan sensitivitas skrining CT dosis rendah, terutama pada pasien yang memiliki nodul paru-paru yang tidak jelas. Maka dari itu pengembangan machine learning untuk pendiagnosaan kanker paru-paru memudahkan pendiagnosaan dan meningkatkan efisiensi dalam pendiagnosaan non-invasif. Menggunakan metode Gray Level Co-occurance Matrix (GLCM) dan Convolutional Neural Network (CNN) menggunakan arsitektur Densely Connected Convolutional Network (DenseNet) yang digabungkan untuk klasifikasi tipe berdasarkan tekstur yang dilihat dari keabuan dan bentuk serta ukuran nodul.
=================================================================================================================================
Lung cancer, or pulmonary cancer, is a disease with varying degrees of malignancy depending on the condition. Lung cancer occurs when cells within the lungs experience uncontrolled growth and become malignant. There are two primary types of lung cancer: Non-Small Cell Lung Cancer (NSCLC) and Small Cell Lung Cancer (SCLC). NSCLC is the most common type of lung cancer, accounting for approximately 85% of all lung cancer cases. NSCLC is further divided into subtypes, including pulmonary adenocarcinoma (ADC) and pulmonary squamous cell carcinoma (SqCC). In contrast, SCLC is less common and tends to grow more rapidly than NSCLC, with a higher tendency to spread to other parts of the body at the time of diagnosis. Screening using low-dose CT scans, as currently permitted, often has low sensitivity and a high rate of false positives. More than 90% of positive results do not actually indicate the presence of cancer. Additionally, there are currently no additional biomarkers available to improve the sensitivity of low-dose CT screening, particularly for patients with unclear lung nodules. Therefore, the development of machine learning for lung cancer diagnosis would facilitate non-invasive diagnosis and improve diagnostic efficiency. Using methods such as the Gray Level Co-occurrence Matrix (GLCM) and Convolutional Neural Network (CNN) with the DenseNet architecture combined for classification based on texture, grayscale, shape, and size of nodules.

Item Type: Thesis (Other)
Uncontrolled Keywords: Kanker paru- paru, Non-Small Cell Lung Cancer (NSCLC), Gray Level Co-occurance Matrix (GLCM), Convolutional Neural Network (CNN), Densely Connected Convolutional Network (DenseNet), Klasifikasi, Small Cell Lung Cancer (SCLC), Classification
Subjects: R Medicine > R Medicine (General) > R858 Deep Learning
T Technology > T Technology (General)
T Technology > T Technology (General) > T57.5 Data Processing
T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models.
T Technology > T Technology (General) > T59.7 Human-machine systems.
T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User: Irgi Azarya Putra Maulana
Date Deposited: 22 Aug 2024 04:00
Last Modified: 22 Aug 2024 04:00
URI: http://repository.its.ac.id/id/eprint/112781

Actions (login required)

View Item View Item