EAI Endorsed Transactions
on Pervasive Health and Technology Review Article
1
Lung Cancer Detection in CT Images Using Deep
Learning Techniques: A Survey Review
C. Usharani
1,*
, B. Revathi
2
, A. Selvapandian
3
and S. K. Kezial Elizabeth
4
1,2
Ramco Institute of Technology, Rajapalayam, India
3
PSNA College of Engineering and Technology, Dindigul, India
4
Mangayarkarasi College of Engineering, Madurai, India
Abstract
INTRODUCTION: The Computed Tomography (CT) imaging-based Lung cancer detection is crucial for early diagnosis.
This survey paper presents an overview of the techniques and advancements in CT-based lung cancer detection. It covers
the fundamentals of CT imaging, including principles, types, and protocols.
OBJECTIVES: The paper explores image processing techniques for pre-processing, such as noise reduction, enhancement,
and segmentation.
METHODS: Additionally, it discusses feature extraction methods, including shape, texture, and intensity-based features,
as well as Deep Learning (DL) and Machine Learning (ML) methods for automated classification.
RESULTS: Computerised systems and their integration is examined with CT imaging along with performance evaluation
metrics. The survey concludes by addressing challenges, limitations, and future directions. The imaging modalities and
artificial intelligence techniques are used to improve lung cancer detection.
CONCLUSION: This comprehensive survey aims to provide a concise understanding of CT-based lung cancer detection
for researchers and healthcare professionals.
Keywords: Computed Tomography, Lung cancer, Machine Learning, Deep Learning, image processing
Received on 04 December 2023, accepted on 23 February 2024, published on 01 March 2024
Copyright © 2024 C. Usharani et al., licensed to EAI. This is an open access article distributed under the terms of the CC BY-NC-
SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as
the original work is properly cited.
doi: 10.4108/eetpht.10.5265
*
Corresponding author. Email: cusha91@gmail.com
1. Introduction
One of the most prevalent and deadly forms of
cancer worldwide is Lung cancer. It poses a significant
public health challenge, accounting for substantial numerous
deaths globally due to a cancer [1]. Early detection helps to
provide a timely intervention and effective treatment
strategies. The CT imaging has emerged as a vital tool in the
lung cancer detection, offering detailed and cross-sectional
images of the lungs with high resolution and accuracy [2, 3].
Over the years, advancements in CT technology and image
analysis techniques have greatly improved the capabilities of
lung cancer detection. CT imaging provides radiologists
with the ability to visualize and characterize lung nodules,
facilitating early diagnosis and subsequent treatment
planning. The ability to precisely identify and classify these
nodules is paramount in determining their malignancy and
guiding appropriate clinical decisions.
The process of lung cancer detection using CT
imaging involves various stages, including image
acquisition, pre-processing, classification and feature
extraction. An Image pre-processing method used to
minimise noise and enhance images, play a crucial role in
improving the quality of CT images, thereby enhancing the
detection of lung nodules [4]. Feature extraction methods
extract quantitative measurements from the CT images,
capturing relevant information about the shape, texture, and
intensity characteristics of the nodules. These features serve
as discriminative markers for distinguishing between benign
and malignant nodules [5, 6]. Furthermore, ML and DL
algorithms have shown remarkable potential in automating
EAI Endorsed Transactions on
Pervasive Health and Technology
| Volume 10 | 2024 |