Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Journal of Engineering Science and Technology Management
Insanul Kamil, M. Arib
Unknown Affiliation
Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Control & Systems Engineering Electrical & Electronics Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
DICOM Image Analysis for Lung Cancer Detection Using Convolutional Neural Network (CNN) Insanul Kamil, M. Arib; Mulyadi, Romi
Journal of Engineering Science and Technology Management (JES-TM) Vol. 5 No. 2 (2025): September 2025
Publisher : Journal of Engineering Science and Technology Management

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/jestm.v5i2.290

Abstract

Lung cancer remains the leading cause of cancer-related deaths worldwide, with the highest burden in Asia, including Indonesia. Early detection is critical, yet access to radiology services is often limited by infrastructure, cost, and a shortage of trained specialists. Recent advances in artificial intelligence, particularly Convolutional Neural Networks (CNNs), offer promising solutions for automated image-based diagnosis. This study aims to analyze the effectiveness of CNN in detecting lung cancer from CT scan images in DICOM format. A dataset consisting of lung CT images from Kaggle and local hospitals was preprocessed through Gaussian blur filtering, segmentation, and pixel normalization before model training. Images were classified into two categories: cancer and non-cancer. The CNN architecture was trained and validated with an 80:20 split ratio, and model performance was assessed using accuracy, precision, recall, and F1-score. The experimental results show that the proposed CNN model achieved an accuracy of 88.27%, precision of 88.96%, recall of 97.43%, and an F1-score of 92.98%. The high recall value indicates the model’s strong ability to minimize false negatives, which is essential for clinical application. Performance graphs demonstrated stable accuracy and loss across training and validation sets, suggesting minimal overfitting.In conclusion, the developed CNN model demonstrates strong potential as a supportive diagnostic tool for early lung cancer detection, particularly in resource-limited healthcare settings. Its integration into radiology workflows may accelerate screening processes and improve clinical decision-making
Co-Authors Romi Mulyadi