Klasifikasi Kanker Paru pada Citra CT Scan menggunakan Extreme Learning Machine dan Histogram Equalization

Omar Pahlevi (Universitas Bina Sarana Informatika)
Yopi Handrianto (Universitas Bina Sarana Informatika)
Dewi Ayu Nur Wulandari (Universitas Bina Sarana Informatika)
Henny Leidiyana (Universitas Bina Sarana Informatika)
Luci Kanti Rahayu (Universitas Bina Sarana Informatika)

Jurnal Informatika dan Rekayasa Perangkat Lunak
Vol. 7 No. 1 (2025): Maret

Lung cancer is one of the deadliest types of cancer worldwide, making early detection crucial to improving patient survival rates. One of the primary methods for detecting lung cancer is through Computed Tomography (CT) scan images. However, automated analysis of these images faces challenges due to image quality being affected by noise and low contrast. This study aims to develop a lung cancer classification model from CT scan images using the Extreme Learning Machine (ELM) algorithm and Gray Level Co-occurrence Matrix (GLCM) feature extraction, supported by Histogram Equalization techniques to enhance image quality. Histogram Equalization is employed to improve image contrast, facilitating the extraction of texture features from GLCM, such as contrast, homogeneity, energy, and entropy. ELM was chosen for its speed and accuracy in handling complex medical image classification tasks. The study results demonstrate that the proposed model successfully enhances classification performance with an accuracy of 91.06%. The combination of ELM and Histogram Equalization techniques produces an efficient and accurate classification system for detecting lung cancer from CT scan images.