Deep learning algorithms for lung cancer detection in CT images: A narrative review

Read full text Summarise with VIRAT
Authors: Michał Gałuszewski, Tomasz Kotowski, Natalia Wizner, Ewelina Porzycka, Tomasz Biedulski, Michał Azierski
Cited by: 0
References: 37

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Lung cancer remains a leading cause of cancer-related mortality, primarily due to late-stage diagnosis. Deep learning (DL) algorithms, particularly Convolutional Neural Networks (CNNs), have emerged as powerful tools for the early detection of pulmonary nodules in computed tomography (CT) images, demonstrating high sensitivity and specificity. This review explores the effectiveness and accuracy of these algorithms, highlighting their ability to surpass traditional diagnostic methods. However, the widespread clinical implementation of DL faces significant challenges, including the need for large, annotated datasets, the “black box” nature of models which limits their interpretability, and high implementation costs. This paper discusses potential solutions to these obstacles, such as explainable AI (XAI) methodologies like SHAP and LIME, the development of unified datasets, and the integration of hybrid intelligence systems. Furthermore, we explore future directions, including the application of edge computing to enable real-time analysis and enhance data privacy. Despite the existing hurdles, the continued advancement of DL technologies holds the promise of revolutionizing lung cancer diagnostics, leading to earlier detection and improved patient outcomes.

Fields of Study & Subjects

AI in cancer detection Lung Cancer Diagnosis and Treatment Radiomics and Machine Learning in Medical Imaging

Publication details

Journal / source
Polish Journal of Public Health
Volume
136
Pages
24–24
Publication date
25 March 2026
Language
EN