Lung Cancer Classification from CT-Scan Images Using an Enhanced VGG16 Model

Title

Lung Cancer Classification from CT-Scan Images Using an Enhanced VGG16 Model

Creator

Cheekolu, Karunakar; Babu Kumar, S.; Raju, G.

Description

Lung cancer has been one of the most common and deadly types of cancer around the globe, for which early detection is quite crucial for patient survival. In this research work, a deep learning-based method for four-class classification of chest CT-scan images, such as Squamous Cell Carcinoma, Large Cell Carcinoma, Adenocarcinoma, and Normal, is presented. With a modified VGG16 architecture, adding Squeeze-and-Excitation (SE) blocks and residual connections, the enhanced SERES_VGG16 model enhances feature representation and classification accuracy. The dataset we used here contains preprocessed chest CT-scan images divided into a training set, validation set, and test set. It is trained with augmentation techniques in the data to improve generalization. Its performance is evaluated using measures of standard performances, such as F1-score, recall, precision, accuracy and confusion matrices. The model achieved over 95% accuracy, class-wise precision ranging from 94 to 99%, recall ranging from 88 to 99%, F1-score from 93 to 96%. The presented approach reached over 95% accuracy on the test set and can be a trusted second opinion for radiologists to assist with early and accurate lung cancer subtype classification. However, this study is constrained by the small size of the dataset and the lack of other clinical parameters like genetic information. Future studies will concentrate on expanding the dataset and integrating multi-modal clinical information for enhanced robustness. This work in this study justifies the importance of deep learning in the classification of the medical images and points out further ways toward improving automated diagnostic systems. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

Source

Lecture Notes in Networks and Systems;Volume;1614 LNNS;pp.467-480

Date

01-01-2026

Publisher

Springer Science and Business Media Deutschland GmbH

Subject

CT scans; Lung cancer; Residual blocks; Squeeze-and-Excitation; VGG

Coverage

Cheekolu K., Department of Computer Science and Engineering, CHRIST University, Bengaluru, India; Babu Kumar S., Department of Computer Science and Engineering, CHRIST University, Bengaluru, India; Raju G., Department of Computer Science and Engineering, CHRIST University, Bengaluru, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISSN: 23673370; ISBN: 978-981952877-6;

Format

online

Language

English

Type

Conference paper

Identifier

https://doi.org/10.1007/978-981-95-2878-3_33

https://www.scopus.com/pages/publications/105029561054?origin=resultslist