Araştırma Makalesi
BibTex RIS Kaynak Göster

OVER TÜMÖR ULTRASON GÖRÜNTÜLERİ ÜZERİNDE FARKLI BÖLÜTLEME MODELLERİNİN KARŞILAŞTIRMASI

Yıl 2025, Cilt: 28 Sayı: 3, 1486 - 1498, 03.09.2025

İbrahim Aruk

https://doi.org/10.17780/ksujes.1689777

Öz

Kanser, dünya genelinde başlıca ölüm nedenlerinden biri olup, erken teşhis ve etkin tedavi edilmesi, hastalığın kontrol altına alınmasında hayati bir rol oynamaktadır. Over kanseri, kadınlarda en ölümcül jinekolojik kanser türlerinden biri olarak öne çıkmaktadır. Erken evrelerde genellikle açık bir belirti göstermemesi, teşhisin ileri evrelere geçmesine ve tedavi olanağının azalmasına neden olmaktadır. Bu çalışmada, düşük kaliteli ultrason görüntüleri kullanılarak over tümörlerinin otomatik bölütlenmesi için derin öğrenme tabanlı modellerin performansı kapsamlı bir şekilde değerlendirilmiştir. U-Net, Nested U-Net, DeepLabv3+, SegNet ve FCN modelleri, OTU_2D veri seti üzerinde karşılaştırmalı olarak analiz edilmiştir. DeepLabv3+, %96,37 DSC ve %93,76 IoU skorlarıyla en yüksek bölütleme doğruluğu sağlarken, FCN modeli %96,34 DSC ve %93,82 IoU skorlarıyla benzer bir başarı sergilemiştir. Nested U-Net ve U-Net modelleri, lezyon sınır hassasiyeti ve küçük yapıların doğru şekilde bölütlenmesinde etkili olmuştur. SegNet mimarisi ise diğer modellere kıyasla daha sınırlı bir performans göstermiştir.

Anahtar Kelimeler

Görüntü bölütleme Over tümör Tıbbi görüntü analizi

Kaynakça

  • Afify, H. M., Mohammed, K. K., & Hassanien, A. E. (2021). An improved framework for polyp image segmentation based on SegNet architecture. International Journal of Imaging Systems and Technology, 31(3), 1741-1751. https://doi.org/10.1002/ima.22568.
  • Badrinarayanan, V., Kendall, A., & Cipolla, R. (2017). SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(12), 2481-2495. https://doi.org/10.1109/TPAMI.2016.2644615.
  • Bi, L., Feng, D., & Kim, J. (2018). Dual-Path Adversarial Learning for Fully Convolutional Network (FCN)-Based Medical Image Segmentation. The Visual Computer, 34(6), 1043-1052. https://doi.org/10.1007/s00371-018-1519-5.
  • Bierig, S. M., & Jones, A. (2009). Accuracy and cost comparison of ultrasound versus alternative imaging modalities, including CT, MR, PET, and angiography. Journal of Diagnostic Medical Sonography, 25(3), 138-144. https://doi.org/10.1177/8756479309336240.
  • Bui, H. S., Tran, S. H., Nguyen, T. B., Tran, T. H., Vu, H., & Le, T. L. (2024, 3-6 Dec. 2024). Marker-Aware Ovarian Tumor Segmentation from Ultrasound Images. 2024 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), https://doi.org/10.1109/APSIPAASC63619.2025.10848960.
  • Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., & Adam, H. (2018). Encoder-decoder with atrous separable convolution for semantic image segmentation. Proceedings of the European conference on computer vision (ECCV), https://doi.org/10.48550/arXiv.1802.02611.
  • El-khatib, M., Popescu, D., Teodor, O., & Ichim, L. (2024). Intelligent system based on multiple networks for accurate ovarian tumor semantic segmentation. Heliyon, 10(17). https://doi.org/10.1016/j.heliyon.2024.e37386.
  • Maheswari, P., Raja, P., Karkee, M., Raja, M., Baig, R. U., Trung, K. T., & Hoang, V. T. (2025). Performance analysis of modified DeepLabv3+ architecture for fruit detection and localization in apple orchards. Smart Agricultural Technology, 10, 100729. https://doi.org/10.1016/j.atech.2024.100729.
  • Nguyen, T.-N., Tran, T. D., & Cuong, P. V. (2024). Segmentation of Concrete Surface Cracks Using DeeplabV3+ Architecture. Proceedings of the Third International Conference on Sustainable Civil Engineering and Architecture, Singapore, https://doi.org/10.1007/978-981-99-7434-4.
  • Pham, T.-L., Le, V.-H., Tran, T.-H., & Vu, D. H. (2023). Comprehensive Study on Semantic Segmentation of Ovarian Tumors from Ultrasound Images. The 12th Conference on Information Technology and Its Applications, Cham, https://doi.org/10.1007/978-3-031-36886-8_22.
  • Pham, T.-L., Pham, G.-M., Nguyen, T.-D., Le, V.-H., Le, T.-L., Vu, D.-H.,…Tran, T.-H. (2024). Data Augmentation and Assessment for Enhanced Ovarian Tumor Classification. 2024 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), https://doi.org/10.1109/APSIPAASC63619.2025.10848688.
  • Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Cham, https://doi.org/10.1007/978-3-319-24574-4_28.
  • Shelhamer, E., Long, J., & Darrell, T. (2017). Fully Convolutional Networks for Semantic Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(4), 640-651. https://doi.org/10.1109/TPAMI.2016.2572683.
  • Siahpoosh, M., Imani, M., & Ghassemian, H. (2024, 6-7 March 2024). Ovarian Tumor Ultrasound Image Segmentation with Deep Neural Networks. 2024 13th Iranian/3rd International Machine Vision and Image Processing Conference (MVIP), https://doi.org/10.1109/MVIP62238.2024.10491156.
  • Siegel, R. L., Kratzer, T. B., Giaquinto, A. N., Sung, H., & Jemal, A. (2025). Cancer statistics, 2025. Ca, 75(1), 10. https://doi.org/10.3322/caac.21871.
  • Ta, N. A., Ngo, V. A., Pham, T. L., Vu, D. H., Le, T. L., Vu, H.,…Tran, T. H. (2024, 31 July-2 Aug. 2024). Weighted Fusion Architecture for Improved Ovarian Tumor Segmentation from Ultrasound Images. 2024 Tenth International Conference on Communications and Electronics (ICCE), http://doi.org/10.1109/ICCE62051.2024.10634737.
  • Wang, Z., Gao, X., Wu, R., Kang, J., & Zhang, Y. (2022). Fully automatic image segmentation based on FCN and graph cuts. Multimedia Systems, 28(5), 1753-1765. https://doi.org/10.1007/s00530-022-00945-3.
  • Zhang, H., Zhu, C., Lian, X., & Hua, F. (2023). A Nested Attention Guided UNet++ Architecture for White Matter Hyperintensity Segmentation. IEEE Access, 11, 66910-66920. https://doi.org/10.1109/ACCESS.2023.3281201.
  • Zhao, Q., Lyu, S., Bai, W., Cai, L., Liu, B., Cheng, G.,…Chen, L. (2022). MMOTU: a multi-modality ovarian tumor ultrasound image dataset for unsupervised cross-domain semantic segmentation. arXiv preprint arXiv:2207.06799. https://doi.org/10.48550/arXiv.2207.06799.
  • Zhou, Z., Rahman Siddiquee, M. M., Tajbakhsh, N., & Liang, J. (2018). Unet++: A nested u-net architecture for medical image segmentation. Deep learning in medical image analysis and multimodal learning for clinical decision support: 4th international workshop, DLMIA 2018, and 8th international workshop, ML-CDS 2018, held in conjunction with MICCAI 2018, Granada, Spain, September 20, 2018, proceedings 4, https://doi.org/10.1007/978-3-030-00889-5_1.

COMPARISON OF DIFFERENT SEGMENTATION MODELS ON OVARIAN TUMOUR ULTRASOUND IMAGES

Yıl 2025, Cilt: 28 Sayı: 3, 1486 - 1498, 03.09.2025

İbrahim Aruk

https://doi.org/10.17780/ksujes.1689777

Öz

Cancer remains one of the leading causes of mortality worldwide, and early diagnosis along with effective treatment plays a crucial role in controlling the disease. Among gynecological cancers, ovarian cancer stands out as one of the most fatal types for women. The absence of noticeable symptoms in its early stages often delays diagnosis to advanced stages, significantly reducing treatment opportunities. This study comprehensively evaluates the performance of deep learning-based models for the automatic segmentation of ovarian tumors using low-quality ultrasound images. U-Net, Nested U-Net, DeepLabv3+, SegNet, and FCN models were comparatively analyzed on the OTU_2D dataset. DeepLabv3+ demonstrated the highest accuracy with 96.37% DSC and 93.76% IoU scores, while the FCN model achieved comparable performance with 96.34% DSC and 93.82% IoU scores. Nested U-Net and U-Net models were found to be effective in accurately segmenting boundaries and small structures, whereas the SegNet architecture exhibited relatively limited performance compared to the other models.

Anahtar Kelimeler

Image segmentation Ovarian tumour Medical image analysis

Kaynakça

  • Afify, H. M., Mohammed, K. K., & Hassanien, A. E. (2021). An improved framework for polyp image segmentation based on SegNet architecture. International Journal of Imaging Systems and Technology, 31(3), 1741-1751. https://doi.org/10.1002/ima.22568.
  • Badrinarayanan, V., Kendall, A., & Cipolla, R. (2017). SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(12), 2481-2495. https://doi.org/10.1109/TPAMI.2016.2644615.
  • Bi, L., Feng, D., & Kim, J. (2018). Dual-Path Adversarial Learning for Fully Convolutional Network (FCN)-Based Medical Image Segmentation. The Visual Computer, 34(6), 1043-1052. https://doi.org/10.1007/s00371-018-1519-5.
  • Bierig, S. M., & Jones, A. (2009). Accuracy and cost comparison of ultrasound versus alternative imaging modalities, including CT, MR, PET, and angiography. Journal of Diagnostic Medical Sonography, 25(3), 138-144. https://doi.org/10.1177/8756479309336240.
  • Bui, H. S., Tran, S. H., Nguyen, T. B., Tran, T. H., Vu, H., & Le, T. L. (2024, 3-6 Dec. 2024). Marker-Aware Ovarian Tumor Segmentation from Ultrasound Images. 2024 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), https://doi.org/10.1109/APSIPAASC63619.2025.10848960.
  • Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., & Adam, H. (2018). Encoder-decoder with atrous separable convolution for semantic image segmentation. Proceedings of the European conference on computer vision (ECCV), https://doi.org/10.48550/arXiv.1802.02611.
  • El-khatib, M., Popescu, D., Teodor, O., & Ichim, L. (2024). Intelligent system based on multiple networks for accurate ovarian tumor semantic segmentation. Heliyon, 10(17). https://doi.org/10.1016/j.heliyon.2024.e37386.
  • Maheswari, P., Raja, P., Karkee, M., Raja, M., Baig, R. U., Trung, K. T., & Hoang, V. T. (2025). Performance analysis of modified DeepLabv3+ architecture for fruit detection and localization in apple orchards. Smart Agricultural Technology, 10, 100729. https://doi.org/10.1016/j.atech.2024.100729.
  • Nguyen, T.-N., Tran, T. D., & Cuong, P. V. (2024). Segmentation of Concrete Surface Cracks Using DeeplabV3+ Architecture. Proceedings of the Third International Conference on Sustainable Civil Engineering and Architecture, Singapore, https://doi.org/10.1007/978-981-99-7434-4.
  • Pham, T.-L., Le, V.-H., Tran, T.-H., & Vu, D. H. (2023). Comprehensive Study on Semantic Segmentation of Ovarian Tumors from Ultrasound Images. The 12th Conference on Information Technology and Its Applications, Cham, https://doi.org/10.1007/978-3-031-36886-8_22.
  • Pham, T.-L., Pham, G.-M., Nguyen, T.-D., Le, V.-H., Le, T.-L., Vu, D.-H.,…Tran, T.-H. (2024). Data Augmentation and Assessment for Enhanced Ovarian Tumor Classification. 2024 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), https://doi.org/10.1109/APSIPAASC63619.2025.10848688.
  • Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Cham, https://doi.org/10.1007/978-3-319-24574-4_28.
  • Shelhamer, E., Long, J., & Darrell, T. (2017). Fully Convolutional Networks for Semantic Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(4), 640-651. https://doi.org/10.1109/TPAMI.2016.2572683.
  • Siahpoosh, M., Imani, M., & Ghassemian, H. (2024, 6-7 March 2024). Ovarian Tumor Ultrasound Image Segmentation with Deep Neural Networks. 2024 13th Iranian/3rd International Machine Vision and Image Processing Conference (MVIP), https://doi.org/10.1109/MVIP62238.2024.10491156.
  • Siegel, R. L., Kratzer, T. B., Giaquinto, A. N., Sung, H., & Jemal, A. (2025). Cancer statistics, 2025. Ca, 75(1), 10. https://doi.org/10.3322/caac.21871.
  • Ta, N. A., Ngo, V. A., Pham, T. L., Vu, D. H., Le, T. L., Vu, H.,…Tran, T. H. (2024, 31 July-2 Aug. 2024). Weighted Fusion Architecture for Improved Ovarian Tumor Segmentation from Ultrasound Images. 2024 Tenth International Conference on Communications and Electronics (ICCE), http://doi.org/10.1109/ICCE62051.2024.10634737.
  • Wang, Z., Gao, X., Wu, R., Kang, J., & Zhang, Y. (2022). Fully automatic image segmentation based on FCN and graph cuts. Multimedia Systems, 28(5), 1753-1765. https://doi.org/10.1007/s00530-022-00945-3.
  • Zhang, H., Zhu, C., Lian, X., & Hua, F. (2023). A Nested Attention Guided UNet++ Architecture for White Matter Hyperintensity Segmentation. IEEE Access, 11, 66910-66920. https://doi.org/10.1109/ACCESS.2023.3281201.
  • Zhao, Q., Lyu, S., Bai, W., Cai, L., Liu, B., Cheng, G.,…Chen, L. (2022). MMOTU: a multi-modality ovarian tumor ultrasound image dataset for unsupervised cross-domain semantic segmentation. arXiv preprint arXiv:2207.06799. https://doi.org/10.48550/arXiv.2207.06799.
  • Zhou, Z., Rahman Siddiquee, M. M., Tajbakhsh, N., & Liang, J. (2018). Unet++: A nested u-net architecture for medical image segmentation. Deep learning in medical image analysis and multimodal learning for clinical decision support: 4th international workshop, DLMIA 2018, and 8th international workshop, ML-CDS 2018, held in conjunction with MICCAI 2018, Granada, Spain, September 20, 2018, proceedings 4, https://doi.org/10.1007/978-3-030-00889-5_1.
Toplam 20 adet kaynakça vardır.

Ayrıntılar

Birincil Dil Türkçe
Konular Görüntü İşleme, Yapay Görme
Bölüm Bilgisayar Mühendisliği
Yazarlar

İbrahim Aruk 0009-0003-7483-4542

Yayımlanma Tarihi 3 Eylül 2025
Gönderilme Tarihi 2 Mayıs 2025
Kabul Tarihi 16 Haziran 2025
Yayımlandığı Sayı Yıl 2025 Cilt: 28 Sayı: 3

Kaynak Göster

APA Aruk, İ. (2025). OVER TÜMÖR ULTRASON GÖRÜNTÜLERİ ÜZERİNDE FARKLI BÖLÜTLEME MODELLERİNİN KARŞILAŞTIRMASI. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 28(3), 1486-1498. https://doi.org/10.17780/ksujes.1689777
 Kapak Resmi İndir