EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Optimized Lung Nodule Classification Using CLAHE-Enhanced CT Imaging and Swin Transformer-Based Deep Feature Extraction

Dorsaf Hrizi iD, Khaoula Tbarki iD, Sadok Elasmi

Journal of Imaging · 04.10.2025 · 11 (10) · 346

DOI: 10.3390/jimaging11100346 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 04.10.2025. Verfasst von Dorsaf Hrizi und 2 weiteren Autor:innen. Beteiligte Einrichtung: COSIM Laboratory, Higher School of Communication of Tunis, University of Carthage, Ariana 2083, Tunisia. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Lung cancer remains one of the most lethal cancers globally. Its early detection is vital to improving survival rates. In this work, we propose a hybrid computer-aided diagnosis (CAD) pipeline for lung cancer classification using Computed Tomography (CT) scan images. The proposed CAD pipeline integrates ten image preprocessing techniques and ten pretrained deep learning models for feature extraction including convolutional neural networks and transformer-based architectures, and four classical machine learning classifiers. Unlike traditional end-to-end deep learning systems, our approach decouples feature extraction from classification, enhancing interpretability and reducing the risk of overfitting. A total of 400 model configurations were evaluated to identify the optimal combination. The proposed approach was evaluated on the publicly available Lung Image Database Consortium and Image Database Resource Initiative dataset, which comprises 1018 thoracic CT scans annotated by four thoracic radiologists. For the classification task, the dataset included a total of 6568 images labeled as malignant and 4849 images labeled as benign. Experimental results show that the best performing pipeline, combining Contrast Limited Adaptive Histogram Equalization, Swin Transformer feature extraction, and eXtreme Gradient Boosting, achieved an accuracy of 95.8%.

Schlagwörter

Stichwörter der Autor:innen

contrast limited adaptive histogram equalizationdeep learningeXtreme Gradient Boostingimage preprocessinglung cancerswin transformertransfer learning

Autor:innen und Institutionen

  • COSIM Laboratory, Higher School of Communication of Tunis, University of Carthage, Ariana 2083, Tunisia.
  • National Institute of Technology and Science of Kef, University of Jendouba, El Kef 7100, Tunisia.
  • LR-SITI, National Engineering School of Tunis (ENIT), University of Tunis El Manar, Tunis 2092, Tunisia.

ORCID-Profile: Dorsaf Hrizi, Khaoula Tbarki

Zitieren

APA

Hrizi, D., Tbarki, K., & Elasmi, S. (2025). Optimized Lung Nodule Classification Using CLAHE-Enhanced CT Imaging and Swin Transformer-Based Deep Feature Extraction. Journal of Imaging, 11(10), 346. https://doi.org/10.3390/jimaging11100346

Vancouver

Hrizi D, Tbarki K, Elasmi S. Optimized Lung Nodule Classification Using CLAHE-Enhanced CT Imaging and Swin Transformer-Based Deep Feature Extraction. Journal of Imaging. 2025;11(10):346. doi:10.3390/jimaging11100346

BibTeX
@article{Hrizi2025,
  author = {Hrizi, Dorsaf and Tbarki, Khaoula and Elasmi, Sadok},
  title = {Optimized Lung Nodule Classification Using CLAHE-Enhanced CT Imaging and Swin Transformer-Based Deep Feature Extraction},
  journal = {Journal of Imaging},
  year = {2025},
  volume = {11},
  number = {10},
  pages = {346},
  doi = {10.3390/jimaging11100346}
}
Kostenlos starten →