Comparative Evaluation of U-Net-Based Models for Lung Lesion Segmentation in CT Images

Abstract

Lung cancer remains one of the leading causes of cancer-related deaths worldwide, and early detection plays a crucial role in improving treatment success and patient survival. In this study, the problem of lesion segmentation in lung computed tomography (CT) images was addressed, and the performance of different U-Net-based deep learning architectures was comparatively evaluated. The dataset used in this study consisted of 22,782 CT slices obtained from 223 patients. Among these images, 3,584 slices contained lesions, while 19,198 slices were healthy images. To investigate the impact of class imbalance in the dataset on model performance, three different experimental scenarios were applied: using only lesion-containing images, using all images together, and using a balanced training dataset. Segmentation performance was analyzed using both image-based and patient-based evaluation approaches. The results indicate that data distribution has a significant impact on segmentation performance. In particular, balancing the training dataset improved the results, especially in patient-based evaluations.

Keywords

• Lung cancer
• computed tomography
• deep learning
• image segmentation
• U-Net
• medical image analysis

Supporting Institution

TUBİTAK

Project Number

125E062

Ethical Statement

This study was approved by the Scientific Research and Publication Ethics Committee of İnönü University (Date: 30.07.2024, Session No: 13, Decision No: 2024/6289).

Thanks

This study was supported by the Scientific and Technological Research Council of Türkiye (TÜBİTAK) under the 1001 Scientific and Technological Research Projects Support Program (Project No: 125E062).

BT Görüntülerinde Akciğer Lezyonu Segmentasyonu için U-Net Tabanlı Modellerin Karşılaştırmalı Değerlendirilmesi

Abstract

Akciğer kanseri dünya genelinde kanser kaynaklı ölümlerin başlıca nedenlerinden biridir ve hastalığın erken evrede tespit edilmesi tedavi başarısı açısından büyük önem taşımaktadır. Bu çalışmada akciğer bilgisayarlı tomografi (BT) görüntülerinde lezyon segmentasyonu problemi ele alınmış ve farklı U-Net tabanlı derin öğrenme mimarilerinin performansları karşılaştırmalı olarak incelenmiştir. Çalışmada toplam 223 hastaya ait 22.782 BT kesitinden oluşan bir veri kümesi kullanılmıştır. Veri kümesinde 3.584 kesit lezyon içerirken 19.198 kesit sağlıklı görüntülerden oluşmaktadır. Veri kümesindeki sınıf dengesizliğinin model performansı üzerindeki etkisini incelemek amacıyla üç farklı deneysel senaryo uygulanmıştır: yalnızca lezyon içeren görüntülerin kullanılması, tüm görüntülerin birlikte kullanılması ve eğitim veri setinde dengeli veri kullanımı. Segmentasyon performansı görüntü bazlı ve hasta bazlı değerlendirme yaklaşımları kullanılarak analiz edilmiştir. Elde edilen sonuçlar veri dağılımının segmentasyon performansı üzerinde önemli bir etkiye sahip olduğunu göstermiştir. Özellikle eğitim veri setinde dengeli veri kullanımının hasta bazlı değerlendirmelerde performansı artırdığı belirlenmiştir.

Keywords

• Akciğer kanseri
• bilgisayarlı tomografi
• derin öğrenme
• görüntü segmentasyonu
• U-Net
• medikal görüntü analizi

Supporting Institution

TUBİTAK

Project Number

125E062

Ethical Statement

Bu çalışma, İnönü Üniversitesi Bilimsel Araştırma ve Yayın Etiği Kurulu tarafından onaylanmıştır (Tarih: 30.07.2024, Oturum No: 13, Karar No: 2024/6289).

Thanks

Bu çalışma, Türkiye Bilimsel ve Teknolojik Araştırma Kurumu (TÜBİTAK) tarafından 1001 Bilimsel ve Teknolojik Araştırma Projelerini Destekleme Programı kapsamında desteklenmiştir (Proje No: 125E062).

References

1. Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229-63. doi:10.3322/caac.21834
2. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016. p. 770-8. doi:10.1109/CVPR.2016.90
3. Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JHM, Beasley MB, et al. The 2015 World Health Organization Classification of Lung Tumors: Impact of Genetic, Clinical and Radiologic Advances since the 2004 Classification. Journal of Thoracic Oncology. 2015;10(9):1243-60. doi:10.1097/JTO.0000000000000630
4. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209-49. doi:10.3322/caac.21660
5. Riquelme D, Akhloufi MA. Deep Learning for Lung Cancer Nodules Detection and Classification in CT Scans. AI. 2020;1(1):28-67. doi:10.3390/ai1010003
6. Wang X, Wang L, Zheng P. SC-Dynamic R-CNN: A Self-Calibrated Dynamic R-CNN Model for Lung Cancer Lesion Detection. Comput Math Methods Med. 2022;2022:9452157. doi:10.1155/2022/9452157
7. Hanbay K, Çalışan M, Özdemir TB. SAR Ship Detection Using Image Histograms and Machine Learning Approach. Turkish Journal of Nature and Science. 2024;13(3):171-175. doi:10.46810/tdfd.1528267
8. Gu Y, Lu X, Yang L, Zhang B, Yu D, Zhao Y, et al. Automatic lung nodule detection using a 3D deep convolutional neural network combined with a multi-scale prediction strategy in chest CTs. Comput Biol Med. 2018;103:220-31. doi:10.1016/j.compbiomed.2018.10.011

9. Lakshmanaprabu SK, Mohanty SN, Shankar K, Arunkumar N, Ramirez G. Optimal deep learning model for classification of lung cancer on CT images. Future Generation Computer Systems. 2019;92:374-82. doi:10.1016/j.future.2018.10.009
10. Paing MP, Hamamoto K, Tungjitkusolmun S, Visitsattapongse S, Pintavirooj C. Automatic detection of pulmonary nodules using three-dimensional chain coding and optimized random forest. Applied Sciences. 2020;10(7). doi:10.3390/app10072346
11. Perez G, Arbelaez P. Automated lung cancer diagnosis using three-dimensional convolutional neural networks. Med Biol Eng Comput. 2020;58(8):1803-15. doi:10.1007/s11517-020-02197-7
12. Naqi SM, Sharif M, Jaffar A. Lung nodule detection and classification based on geometric fit in parametric form and deep learning. Neural Comput Appl. 2020;32(9):4629-47. doi:10.1007/s00521-018-3773-x
13. Khan MA, Rubab S, Kashif A, Sharif MI, Muhammad N, Shah JH, et al. Lungs cancer classification from CT images: An integrated design of contrast based classical features fusion and selection. Pattern Recognit Lett. 2020 Jan 1;129:77-85. doi:10.1016/j.patrec.2019.11.014
14. Suresh S, Mohan S. ROI-based feature learning for efficient true positive prediction using convolutional neural network for lung cancer diagnosis. Neural Comput Appl. 2020;32(20):15989-6009. doi:10.1007/s00521-020-04787-w
15. Jain S, Indora S, Atal DK. Lung nodule segmentation using Salp Shuffled Shepherd Optimization Algorithm-based Generative Adversarial Network. Comput Biol Med. 2021;137:104811. doi:10.1016/j.compbiomed.2021.104811
16. Shimazaki A, Ueda D, Choppin A, Yamamoto A, Honjo T, Shimahara Y, et al. Deep learning-based algorithm for lung cancer detection on chest radiographs using the segmentation method. Sci Rep. 2022 Dec 1;12(1). doi:10.1038/s41598-021-04667-w PubMed PMID: 35031654.
17. Swain AK, Swetapadma A, Rout JK, Balabantaray BK. Classification of non-small cell lung cancer types using sparse deep neural network features. Biomed Signal Process Control. 2024 Jan 1;87. doi:10.1016/j.bspc.2023.105485
18. Xiao Z, Liu B, Geng L, Zhang F, Liu Y. Segmentation of lung nodules using improved 3D-UNet neural network. Symmetry (Basel). 2020;12(11):1-15. doi:10.3390/sym12111787
19. Kumar V, Prabha C, Sharma P, Mittal N, Askar SS, Abouhawwash M. Unified deep learning models for enhanced lung cancer prediction with ResNet-50-101 and EfficientNet-B3 using DICOM images. BMC Med Imaging. 2024 Dec 1;24(1). doi:10.1186/s12880-024-01241-4 PubMed PMID: 38500083.
20. Cifci MA. SegChaNet: A Novel Model for Lung Cancer Segmentation in CT Scans. Appl Bionics Biomech. 2022;2022. doi:10.1155/2022/1139587
21. Alshayeji MH, Sa'ed Abed. Lung cancer classification and identification framework with automatic nodule segmentation screening using machine learning. Applied Intelligence. 2023;53:19724-41. doi:10.1007/s10489-023-04552-1
22. Wu Y, Liu X, Shi Y, Chen X, Wang Z, Xu Y, et al. S3TU-Net: Structured Convolution and Superpixel Transformer for Lung Nodule Segmentation [Internet]. 2024 Nov 19. Available from: http://arxiv.org/abs/2411.12547
23. Rikhari H, Baidya Kayal E, Ganguly S, Sasi A, Sharma S, Antony A, et al. Improving lung nodule segmentation in thoracic CT scans through the ensemble of 3D U-Net models. Int J Comput Assist Radiol Surg. 2024 Oct 1;19(10):2089-99. doi:10.1007/s11548-024-03222-y PubMed PMID: 39044036.
24. Akila Agnes S, Arun Solomon A, Karthick K. Wavelet U-Net++ for accurate lung nodule segmentation in CT scans: Improving early detection and diagnosis of lung cancer. Biomed Signal Process Control. 2024 Jan 1;87. doi:10.1016/j.bspc.2023.105509
25. Vijayakumar SR, Aarthy S, Deepa D, Suresh P. Sustainable framework for automated segmentation and prediction of lung cancer in CT image using CapsNet with U-net segmentation. Biomed Signal Process Control. 2025 Jan 1;99. doi:10.1016/j.bspc.2024.106873
26. Setio AAA, Traverso A, de Bel T, Berens MSN, Bogaard C van den, Cerello P, et al. Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge. Med Image Anal. 2017 Dec 1;42:1-13. doi:10.1016/j.media.2017.06.015 PubMed PMID: 28732268.
27. Armato SG, McLennan G, Bidaut L, McNitt-Gray MF, Meyer CR, Reeves AP, et al. The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): A Completed Reference Database of Lung Nodules on CT Scans. Med Phys. 2011 Feb 24;38(2):915-31. doi:10.1118/1.3528204
28. Survival of Patients with Stage I Lung Cancer Detected on CT Screening. New England Journal of Medicine. 2006 Oct 26;355(17):1763-71. doi:10.1056/NEJMoa060476
29. Taha AA, Hanbury A. Metrics for evaluating 3D medical image segmentation: Analysis, selection, and tool. BMC Med Imaging. 2015 Aug 12;15(1). doi:10.1186/s12880-015-0068-x PubMed PMID: 26263899.