<![CDATA[Lung cancer is one of the leading causes of death in the world, so early detection is very important to increase the chances of patient recovery. This study aims to develop a method for identifying lung cancer types using Convolutional Neural Network (CNN) combined with Regionprops segmentation technique to determine the position of cancer in CT scan images. The dataset used consists of 1,294 CT scan images classified into three classes, namely Benign, Malignant, and Normal, with variations in the ratio of training and testing data: 80:20, 70:30, 60:40, 50:50, and 40:60. The CNN method is used to perform classification, while the Regionprops segmentation technique is applied to determine the position of the cancer. The results showed that the model with a data ratio of 80:20 achieved the highest accuracy of 99.54%, indicating a very good generalization ability of the model. The Regionprops segmentation technique successfully separated the nodule area in the CT scan image clearly, thus providing more detailed information regarding the position of the cancer. The conclusion of this study shows that the combination of CNN and Regionprops segmentation methods is effective in detecting and analyzing lung cancer and has the potential to be used as a diagnostic tool in the medical field. This study recommends further testing with a larger dataset and optimization of model parameters to improve classification and segmentation performance.]]>
Copyrights © 2025
