Journals / Gazi University Journal of Science Part A: Engineering and Innovation / 2026 / Volume: 13 Issue: 1

Residual U-Net and Tversky Loss for Multi-Class Anatomical Segmentation in Chest X-Ray Images

Pages
—
Publication Date
2026-03-31
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results show that the Tversky loss alone increased the Dice score from 0.296 to 0.548, while residual connections increased it to 0.444.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study presents a deep learning based method for the simultaneous segmentation of five anatomical structures in chest X-ray images, namely the left lung, right lung, heart, left clavicle, and right clavicle, using the Japanese Society of Radiological Technology (JSRT) dataset.
📊
Key Findings & Data:In the initial configuration, a baseline U-Net model trained with the Cross-Entropy loss achieved low validation loss values; however, the regional overlap metrics did not reach satisfactory levels, and noticeable performance degradation was observed particularly on small anatomical structures.
💡
Conclusion & Impact:The results show that the Tversky loss alone increased the Dice score from 0.296 to 0.548, while residual connections increased it to 0.444.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Chest X-Ray · Segmentation · U-Net

Abstract

This study presents a deep learning based method for the simultaneous segmentation of five anatomical structures in chest X-ray images, namely the left lung, right lung, heart, left clavicle, and right clavicle, using the Japanese Society of Radiological Technology (JSRT) dataset. In the initial configuration, a baseline U-Net model trained with the Cross-Entropy loss achieved low validation loss values; however, the regional overlap metrics did not reach satisfactory levels, and noticeable performance degradation was observed particularly on small anatomical structures. To systematically examine the effects of residual connections and the Tversky loss function, four model configurations were evaluated: (i) U-Net with Cross-Entropy, (ii) U-Net with Tversky, (iii) Residual U-Net with Cross-Entropy, and (iv) Residual U-Net with Tversky. The results show that the Tversky loss alone increased the Dice score from 0.296 to 0.548, while residual connections increased it to 0.444. The configuration combining both components achieved the highest performance, reaching an average Dice score of 0.826 and a Jaccard score of 0.704 on the test set. Dice values reached the range of 0.86–0.88 for the lung regions, while scores of 0.696 and 0.817 were obtained for the heart and right clavicle, respectively. In contrast, low performance was observed for left clavicle segmentation across all configurations (maximum Dice: 0.108), which is attributed to class imbalance, anatomical variation, and low contrast. Overall, the findings indicate that pixel-wise Cross-Entropy loss does not directly optimize regional overlap, whereas the combined use of residual learning and the Tversky loss provides a more stable and accurate solution for multi-class chest anatomy segmentation.

🔖

Cite This Article

Arslan, B. (2026). Residual U-Net and Tversky Loss for Multi-Class Anatomical Segmentation in Chest X-Ray Images. Gazi University Journal of Science Part A: Engineering and Innovation, Volume: 13 Issue: 1. https://doi.org/10.54287/gujsa.1844215

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (19)

  1. Candemir, S., Jaeger, S., Palaniappan, K., Musco, J. P., Singh, R. K., Xue, Z., Karargyris, A., Antani, S., Thoma, G., & McDonald, C. J. (2014). Lung segmentation in chest radiographs using anatomical atlases with nonrigid registration. IEEE Transactions on Medical Imaging, 33(2), 577–590. https://doi.org/10.1109/TMI.2013.2290491
  2. Gaggion, N., Mansilla, L., Milone, D. H., & Ferrante, E. (2021). Hybrid graph convolutional neural networks for landmark-based anatomical segmentation. In Medical Image Computing and Computer Assisted Intervention – MICCAI 2021 (pp. 600–610). Springer.
  3. Gaggion, N., Mansilla, L., Mosquera, C., Milone, D. H., & Ferrante, E. (2023). Improving anatomical plausibility in medical image segmentation via hybrid graph neural networks: Applications to chest X-ray analysis. IEEE Transactions on Medical Imaging, 42(2), 546–556. https://doi.org/10.1109/TMI.2022.3224660
  4. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
  5. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (pp. 770-778). https://doi.org/10.1109/CVPR.2016.90
  6. Jaccard, P. (1912). The distribution of the flora in the alpine zone. New Phytologist, 11(2), 37-50. https://doi.org/10.1111/j.1469-8137.1912.tb05611.x
  7. Maas, A. L., Hannun, A. Y., & Ng, A. Y. (2013). Rectifier nonlinearities improve neural network acoustic models. In: Proceedings of the 30th International Conference on Machine Learning (ICML) Workshop on Deep Learning for Audio, Speech, and Language Processing.
  8. Milletari, F., Navab, N., & Ahmadi, S. A. (2016). V-Net: Fully convolutional neural networks for volumetric medical image segmentation. In: Proceedings of the 4th International Conference on 3D Vision (3DV), (pp. 565-571). https://doi.org/10.1109/3DV.2016.79
  9. Ngo, T. A., & Carneiro, G. (2015). Lung segmentation in chest radiographs using distance regularized level set and deep-structured learning and inference. In: IEEE International Conference on Image Processing (ICIP), pp. 2140-2143. https://doi.org/10.1109/ICIP.2015.7351179
  10. Novikov, A. A., Lenis, D., Major, D., Hladůvka, J., Wimmer, M., & Bühler, K. (2018). Fully convolutional architectures for multiclass segmentation in chest radiographs. IEEE Transactions on Medical Imaging, 37(8), 1865–1876. https://doi.org/10.1109/TMI.2018.2806086
  11. Odena, A., Dumoulin, V., & Olah, C. (2016). Deconvolution and checkerboard artifacts. Distill, 1(10). https://doi.org/10.23915/distill.00003
  12. Orenc, S., Ozerdem, M. S., Acar, E., & Yilmaz, M. (2025). Automatic segmentation of chest X-ray images via deep-improved various U-Net techniques. Digital Health, 11, 1–15. https://doi.org/10.1177/20552076251366855
  13. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In: Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), (pp. 234-241). https://doi.org/10.1007/978-3-319-24574-4_28
  14. Salehi, S. S. M., Erdogmus, D., & Gholipour, A. (2017). Tversky loss function for image segmentation using 3D fully convolutional deep networks. In: Proceedings of the 10th International Workshop on Machine Learning in Medical Imaging (MLMI), (pp. 379-387). https://doi.org/10.1007/978-3-319-67389-9_44
  15. Shiraishi, J., Katsuragawa, S., Ikezoe, J., Matsumoto, T., Kobayashi, T., Komatsu, K., Matsui, M., Fujita, H., Kodera, Y., & Doi, K. (2000). Development of a digital image database for chest radiographs with and without a lung nodule. American Journal of Roentgenology, 174(1), 71–74. https://doi.org/10.2214/ajr.174.1.1740071
  16. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15(1), 1929-1958.
  17. Ulyanov, D., Vedaldi, A., & Lempitsky, V. (2016). Instance normalization: The missing ingredient for fast stylization. https://arxiv.org/abs/1607.08022
  18. van Ginneken, B., Stegmann, M. B., & Loog, M. (2006). Segmentation of anatomical structures in chest radiographs using supervised methods: A comparative study on a public database. Medical Image Analysis, 10(1), 19-40. https://doi.org/10.1016/j.media.2005.02.002
  19. Wang, Y., Guo, Y., Wang, Z., Yu, L., Yan, Y., & Gu, Z. (2024). Enhancing semantic segmentation in chest X-ray images through image preprocessing: ps-KDE for pixel-wise substitution by kernel density estimation. PLOS One, 19(6), e0299623. https://doi.org/10.1371/journal.pone.0299623