İntrakraniyal Kanama Tespiti İçin Segmentasyon Algoritmalarının Karşılaştırmalı Bir Çalışması

Yıl 2024, Cilt: 3 Sayı: 2, 75 - 94, 12.06.2024

Murat Canayaz Aysel Milanlioglu Sanem Şehribanoğlu Abdulsabır Yalın Adem Yokuş

https://doi.org/10.62520/fujece.1423648

Öz

Tıp alanında segmentasyon özel bir öneme sahiptir. Segmentasyonun amaçlarından biri, herhangi bir organdaki hastalık tespiti sonrasında hastalıktan etkilenen bölgeyi görselleştirmektir. Son yıllarda, bu amaçla derin öğrenme modelleri ile etkili çalışmalar gerçekleştirilmiştir. Bu çalışmada, beyin parankimindeki kanama tespiti için 3 segmentasyon algoritması karşılaştırılmıştır. Bu algoritmalar, en bilinen U-net, LinkNet ve FPN algoritmalarıdır. Bu algoritmaların arka planında, derin öğrenme modellerinden oluşan 5 farklı ana yapı kullanılmıştır. Bu ana yapılar, Resnet34, ResNet50, ResNet169, EfficientNetB0 ve EfficientNet B1'dir. Çalışma için orijinal bir veri kümesi oluşturulmuştur. Çalışmadaki veri kümesi uzmanlar tarafından doğrulanmıştır. Çalışmada, tıp alanındaki en yaygın metrikler olan Dice katsayısı ve Jaccard indeksi, değerlendirme metrikleri olarak seçilmiştir. Algoritmaların performans sonuçları göz önüne alındığında, eğitim verisi için FPN mimarisi 0.9495 Dice katsayısı değeri ile en iyi sonuçları verirken, test verisi için LinkNet 0.9244 Dice katsayısı ile en iyi sonuçları vermiştir. Ayrıca, kullanılan ana yapılar arasında EfficientNetB1 en iyi sonuçları sağlamıştır. Elde edilen sonuçlar incelendiğinde, mevcut çalışmalara göre daha iyi bir segmentasyon performansı elde edilmiştir.

Anahtar Kelimeler

Bölütleme, İntrakraniyal kanama, Derin öğrenme

A Comparative Study of Segmentation Algorithms for Intracerebral Hemorrhage Detection

Öz

Segmentation in the medical field has special importance. One of the purposes of segmentation is to visualize the area affected by the disease after disease detection in any organ. In recent years, efficient studies have been carried out for this purpose with deep learning models. In this study, three segmentation algorithms were compared for the detection of hemorrhage in brain parenchyma. These algorithms are the most familiar: U-net, LinkNet, and FPN algorithms. For the background of these algorithms, five backbones consisting of deep learning models were used. These backbones are Resnet34, ResNet50, ResNet169, EfficientNetB0, and EfficientNet B1. An original dataset was created for the study. The dataset in the study was verified by experts. In the study, the Dice coefficient and Jaccard index, which are the most common metrics in the medical field, were chosen as evaluation metrics. Considering the performance results of the algorithms, the FPN architecture with a 0.9495 Dice coefficient value for the training data and LinkNet with a 0.9244 Dice coefficient for the test data gave the best results. In addition, EfficientNetB1 provided the best results among the backbones used. When the results obtained were examined, better segmentation performance was obtained than in existing studies.

Anahtar Kelimeler

segmentation, deep learning, intracerebral hemorrhage

Kaynakça

• M.U. Rehman, S. Cho, J.H. Kim and K.T. Chong, "Bu-net: Brain tumor segmentation using modified u-net architecture", Elect., 9, 1-12, 2020.
• T. Lan, Y. Li, J.K. Murugi, Y. Ding and Z. Qin, "RUN:Residual U-Net for Computer-Aided Detection of Pulmonary Nodules without Candidate Selection", http://arxiv.org/abs/1805.11856, 2018.
• R.L. Araújo, F.H.D. de Araújo and R.R.V. Silva, Automatic segmentation of melanoma skin cancer using transfer learning and fine-tuning, Multimed. Syst., 2021.
• Q. Chen, P. Liu, J. Ni, Y. Cao, B. Liu and H. Zhang, "Pseudo-Labeling for Small Lesion Detection on Diabetic Retinopathy Images", Int. Jt. Conf. Neural Networks, IEEE, pp. 1–8, 2020.
• C.J. J van Asch, M.J. A Luitse, G.J. E Rinkel, I. van der Tweel, A. Algra and C.J. M Klijn, "Incidence, case fatality, and functional outcome of intracerebral haemorrhage over time, according to age, sex, and ethnic origin: a systematic review and meta-analysis", Lancet Neurol., 9, 167–176, 2010.
• M.T.C. Poon, A.F. Fonville and R. Al-Shahi Salman, "Long-term prognosis after intracerebral haemorrhage: systematic review and meta-analysis", J. Neurol. Neur., Psych., 85, 660– 667, 2014.
• M. Yamada, "Cerebral amyloid angiopathy: emerging concepts", J. Stroke. 17, 17–30, 2015.
• W.C. Ziai and J.R. Carhuapoma, "Intracerebral Hemorrhage., Continuum (Minneap. Minn)", 24, 1603–1622, 2018.
• V. Aiyagari, "The clinical management of acute intracerebral hemorrhage", Expert Rev. Neurother. 15, 1421–1432, 2015.
• A.I. Qureshi, S. Tuhrim, J.P. Broderick, H.H. Batjer, H. Hondo and D.F. Hanley, "Spontaneous intracerebral hemorrhage", N. Engl. J. Med. 344, 1450–1460, 2001.
• O. Flower and M. Smith, "The acute management of intracerebral hemorrhage", Curr. Opin. Crit. Care. 17, 106–114, 2011.
• J.C. 3rd Hemphill, S.M. Greenberg, C.S. Anderson, K. Becker, B.R. Bendok, M. Cushman, G.L. Fung, J.N. Goldstein, R.L. Macdonald, P.H. Mitchell, P.A. Scott, M.H. Selim and D. Woo, "Guidelines for the Management of Spontaneous Intracerebral Hemorrhage: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association",. 46 2032–2060, 2015.
• Kaggle, RSNA Intracranial Hemorrhage Detection, https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/data, 2021.
• M.I. Aguilar and T.G. Brott, "Update in intracerebral hemorrhage", The Neurohosp., 1, 148–159, 2011.
• A. Patel, F.H.B.M. Schreuder, C.J.M. Klijn, M. Prokop, B. van Ginneken, H.A. Marquering, Y.B.W.E.M. Roos, M.I. Baharoglu, F.J.A. Meijer and R. Manniesing, "Intracerebral haemorrhage segmentation in non-contrast CT. ", Sci. Rep. 9, 17858, 2019.
• J. Xu, R. Zhang, Z. Zhou, C. Wu, Q. Gong, H. Zhang, S. Wu, G. Wu, Y. Deng, C. Xia and J. Ma, "Deep network for the automatic segmentation and quantification of ıntracranial hemorrhage on CT", Front. Neurosci. 14, 1084, 2021.
• K. Hu, K. Chen, X. He, Y. Zhang, Z. Chen, X. Li and X. Gao, "Automatic segmentation of intracerebral hemorrhage in CT images using encoder–decoder convolutional neural network", Infor. Process. & Manag., 57, 102352, 2020.
• T. Falk, D. Mai, R. Bensch, Ö. Çiçek, A. Abdulkadir, Y. Marrakchi, A. Böhm, J. Deubner, Z. Jäckel, K. Seiwald, A. Dovzhenko, O. Tietz, C. Dal Bosco, S. Walsh, D. Saltukoglu, T.L. Tay, M. Prinz, K. Palme, M. Simons, I. Diester, T. Brox and O. Ronneberger, "U-Net: deep learning for cell counting, detection, and morphometry", Nat. Methods. 16, 67–70, 2019.
• A. Arab, B. Chinda, G. Medvedev, W. Siu, H. Guo, T. Gu, S. Moreno, G. Hamarneh, M. Ester and X. Song, "A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT", Sci. Rep. 10, 19389, 2020.
• S. Jadon, O.P. Leary, I. Pan, T.J. Harder, D.W. Wright, L.H. Merck and D. Merck, "A comparative study of 2D image segmentation algorithms for traumatic brain lesions using CT data from the ProTECTIII multicenter clinical trial", in: T.M. Deserno, P.-H. Chen (Eds.), Imaging Informatics Heal. Res. Appl., SPIE, p. 48, 2020.
• G. Cao, Y. Wang, X. Zhu, M. Li, X. Wang and Y. Chen, "Segmentation of İntracerebral Hemorrhage Based on İmproved U-Net", in: 2020 IEEE Conf. Telecommun. Opt. Comput. Sci., IEEE, pp. 183–185, 2020.
• J.L. Wang, H. Farooq, H. Zhuang and A.K. Ibrahim, "Segmentation of Intracranial Hemorrhage Using Semi-Supervised Multi-Task Attention-Based U-Net", Appl. Sci. 10, 2020.
• B.S. Maya and T. Asha, "Segmentation and classification of brain hemorrhage using U-net and CapsNet", J. Seybold Rep. 1, 18–25, 2020.
• M.D. Hssayeni, M.S. Croock, A. Al-Ani, H.F. Al-Khafaji, Z.A. Yahya and B. Ghoraani, "Intracranial Hemorrhage Segmentation Using Deep Convolutional Model, (n.d.).
• V. Abramova, A. Clèrigues, A. Quiles, D.G. Figueredo, Y. Silva, S. Pedraza, A. Oliver and X. Lladó, "Hemorrhagic stroke lesion segmentation using a 3D U-Net with squeeze-and-excitation blocks", Comput. Med. Imaging Graph. 90, 101908, 2021.
• Y. Liu, Q. Fang, A. Jiang, Q. Meng, G. Pang and X. Deng, "Texture analysis based on U-Net neural network for intracranial hemorrhage identification predicts early enlargement", Comput. Methods Programs Biomed. 206, 106140, 2021.
• C. Sai Manasa and V. Bhavana, "Deep Learning Algorithms to Detect and Localize Acute Intracranial Hemorrhages", in: Sabu M. Thampi, Sri Krishnan, Rajesh M. Hegde, Domenico Ciuonzo, Thomas Hanne, Jagadeesh Kannan R. (Eds.), Adv. Signal Process. Intell. Recognit. Syst., Springer, Singapore, 367–374, 2021.
• J. He, "Automated Detection of Intracranial Hemorrhage on Head Computed Tomography with Deep Learning", in: Proc. 2020 10th Int. Conf. Biomed. Eng. Technol., ACM, New York, NY, USA, 117–121, 2020.
• S. Castro, Juan Sebastian Chabert, C. Saavedra and R. Salas, "Convolutional neural networks for detection intracranial hemorrhage in CT images", in: CEUR Workshop Proc., 37–43, 2020.
• M. Burduja, R.T. Ionescu and N. Verga, "Accurate and efficient ıntracranial hemorrhage detection and subtype classification in 3d ct scans with convolutional and long short-term memory neural networks", Sensors. 20, 5611, 2020.
• S. Ghosh and K.C. Santosh, "Tumor Segmentation in Brain MRI: U-Nets versus Feature Pyramid Network", IEEE 34th Int. Symp. Comput. Med. Syst., IEEE, 2021.
• Z. Sobhaninia, A. Emami, N. Karimi and S. Samavi, "Localization of Fetal Head in Ultrasound Images by Multiscale View and Deep Neural Networks", 25th Int. Comput. Conf. Comput. Soc. Iran, IEEE, 2020.
• D. Fan, C. Zhang, B. Lv, L. Wang, G. Wang, M. Wang, C. Lv and G. Xie, "Positive-Aware Lesion Detection Network with Cross-scale Feature Pyramid for OCT Images", 2020.
• X. Dai, Y. Lei, T. Wang, A.H. Dhabaan, M. McDonald, J.J. Beitler, W.J. Curran, J. Zhou, T. Liu and X. Yang, "Head-and-neck organs-at-risk auto-delineation using dual pyramid networks for CBCT-guided adaptive radiotherapy", Phys. Med. Biol. 66, 2021.
• J. Lo, S. Nithiyanantham, J. Cardinell, D. Young, S. Cho, A. Kirubarajan, M.W. Wagner, R. Azma, S. Miller, M. Seed, B. Ertl-Wagner and D. Sussman, "Cross Attention Squeeze Excitation Network (CASE-Net) for Whole Body Fetal MRI Segmentation", Sensors. 21, 4490, 2021.
• J. Singh, A. Tripathy, P. Garg and A. Kumar, "Lung tuberculosis detection using anti-aliased convolutional networks", Procedia Comput. Sci. 173, 281–290, 2020.
• J. Long, E. Shelhamer and T. Darrell, "Fully convolutional networks for semantic segmentation, IEEE Conf. Comput. Vis. Pattern Recognit., IEEE, 2015.
• O. Ronneberger, P. Fischer and T. Brox, "U-Net: Convolutional Networks for Biomedical Image Segmentation", N. Navab, J. Hornegger, W.M. Wells, A.F. Frangi (Eds.), Med. Image Comput. Comput. Interv. – MICCAI 2015, Lecture Notes in Computer Science, vol 9351.Springer, Munich, Germany, 234–241, 2015.
• L. Jiao and J. Zhao, "A survey on the new generation of deep learning in ımage processing", IEEE Access. 7, 2019.
• L. Li, M. Wei, B. Liu, K. Atchaneeyasakul, F. Zhou, Z. Pan, S.A. Kumar, J.Y. Zhang, Y. Pu, D.S. Liebeskind and F. Scalzo, "Deep learning for hemorrhagic lesion detection and segmentation on brain CT Images", IEEE J. Biomed. Heal. Informatics. 25, 2021.
• G. Cao, Y. Wang, X. Zhu, M. Li, X. Wang and Y. Chen, "Segmentation of intracerebral hemorrhage based on improved U-Net, IEEE Conf. Telecommun. Opt. Comput. Sci., IEEE, pp. 183–185, 2020.
• M.G. Oghli, A. Shabanzadeh, S. Moradi, N. Sirjani, R. Gerami, P. Ghaderi, M. Sanei Taheri, I. Shiri, H. Arabi and H. Zaidi, "Automatic fetal biometry prediction using a novel deep convolutional network architecture", Phys. Medica. 88, 127–137, 2021.
• A. Chaurasia and E. Culurciello, "LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentation", 2017.
• Z. Sobhaninia, S. Rezaei, N. Karimi, A. Emami and S. Samavi, "Brain Tumor Segmentation by Cascaded Deep Neural Networks Using Multiple Image Scales", 28th Iran. Conf. Electr. Eng., IEEE, 2020. S.S. Seferbekov, V.I. Iglovikov, A. V. Buslaev and A.A. Shvets, "Feature Pyramid Network for Multi-Class Land Segmentation", http://arxiv.org/abs/1806.03510, 2018.
• E.H. Adelson, C.H. Anderson, J.R. Bergen, P.J. Burt and J.M. Ogden, "Pyramid methods in image processing", RCA Eng. 29, 33–41. http://persci.mit.edu/pub_pdfs/RCA84.pdf, 1984.
• T.-Y. Lin, P. Dollar, R. Girshick, K. He, B. Hariharan and S. Belongie, "Feature Pyramid Networks for Object Detection", IEEE Conf. Comput. Vis. Pattern Recognit., IEEE, pp. 936–944, 2017.
• M. Hu, Y. Li, L. Fang and S. Wang, "A2-FPN: Attention Aggregation Based Feature Pyramid Network for Instance Segmentation", in: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., pp. 15343–15352, 2021.
• K. He, X. Zhang, S. Ren and J. Sun, "Deep Residual Learning for Image Recognition", arXiv:1512.03385v1 (n.d.) 1–12. http://image-net.org/challenges/LSVRC/2015/, 2021.
• M. Gao, J. Chen, H. Mu, D. Qi, "A transfer residual neural network based on resnet‐34 for detection of wood knot defects", Fores., 12, 1–16, 2021.
• M. Tan and Q. V. Le, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks", arXiv:1905.11946v5, http://arxiv.org/abs/1905.11946, 2019.
• T.Y. Lin, P. Goyal, R. Girshick, K. He and P. Dollar, "Focal loss for dense object detection", IEEE Trans. Pattern Anal. Mach. Intell. 42, 318–327, 2020.