Deep Learning Based Computer Aided Diagnostic System for Brain Tumor Detection

Year 2022, Volume: 10 Issue: 4, 1748 - 1762, 25.10.2022

Tarıkcan Doğanay , Oktay Yıldız

Abstract

Brain MRI segmentation is important in many clinical applications. Brain analysis is based on various approaches, findings, correct segmentation of anatomical regions. Brain MRI quantitative analysis has been used commonly for characterization of brain disorders such as epilepsy, schizophrenia, Alzheimer's disease, cancer, and infectious degenerative diseases. In this study, an end-to-end deep learning segmentation method called Multi-Scale Multi-Level Network MM-Network has been developed. The main idea is to combine the global contextual features of multiple spatial scales at the convolutional network level in UNet. In addition, the extended convolution module is used, which allows the receptive field to expand at different rates depending on the size of the feature maps throughout the networks. The MM-Network model apply in this study was used for tumor segmentation in brain MR images, and successful results were obtained

Keywords

Deep learning, Computer aided diagnosis system, Segmentation

Beyin Tümör Tespiti İçin Derin Öğrenme Temelli Bilgisayar Destekli Tanı Sistemi

Abstract

Beyin MR segmentasyonu klinik uygulamalarda önem arz etmektedir. Beyin analizi çeşitli yaklaşımlarla bulgular ve anatomik bölgelerin doğru segmentasyonuna dayanır. Beyin MRI analizi, epilepsi, şizofreni, alzheimer, kanser ve bulaşıcı dejeneratif hastalıklar gibi beyin bozukluklarının tedavisi için yaygın bir şekilde kullanılmaktadır. Hasta MRI görüntülerinin doktorlar tarafından manuel segmentasyonu görüntülerin dilim dilim ana hatlarının çıkarılmasını gerektirir. Ancak manuel segmentasyonun uzman görüşü ve teknolojik kısıtları nedeniyle bazı dezavantajları vardır. Bununla birlikte görüntü yorumlama son derece zaman alan bir işlemdir. Tecrübeye bağlı olarak hata yapma oranı da yüksektir. Bu çalışmada, beyin MR görüntülerinden otomatik tümör tespiti için uçtan uca Çok Ölçekli Çok Düzeyli Ağ (Multi-Scale Multi-Level Network MM-Network) modeli önerilmiştir. Gerçekleştirilen çalışmada, UNet'teki evrişimli ağ seviyesinde çoklu uzamsal ölçeklerin küresel bağlamsal özelliklerini birleştirerek, ağlar boyunca özellik haritalarının boyutuna bağlı olarak alıcı alanın farklı oranlarda genişlemesini sağlayan genişletilmiş evrişim modülünden yararlanılmıştır. Yapılan deneysel çalışmalarda önerilen model ile yüksek doğrulukta tümör tespiti sağlanmıştır.

Keywords

Derin öğrenme, Bilgisayar destekli tanı sistemi, Segmentasyon

References

• [1] H. Mohsen, E.S. El-Dahshan, E.S. El-Horbarty and A.B M.Salem, “Brain tumor type classification based on support vector machine in magnetic resonance images,” Annals of Dunarea de Jos University of Galati, vol. Fascicle II, Year IX, 2017.
• [2] H. Chen, Q. Dou, L. Yu, J. Qin and P.A Heng, “VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images,” Neuroimage, vol.170, pp.446-455, 2018.
• [3] S.S. Yadav and S.M. Jadhav, “Deep convolutional neural network based medical image classification for disease diagnosis,” Journal of Big Data, vol.6, no.1, pp.113, 2019.
• [4] M. A. Al-masni, M. A. Al-antari, J.M Park, G. Gi, T.Y Kim, P.Rivera, E. Valarezo, M.T. Choi, S.M Han and T.S. Kim, “Simultaneous detection and classification of breast masses in digital mammograms via a deep learning YOLO-based CAD system,” Computer Methods and Programs in Biomedicine, vol.157, pp.85-94, 2018.
• [5] Zhao, Hengshuang, Jianping Shi, Xiaojuan Qi, Xiaogang Wang and Jiaya Jia. “Pyramid Scene Parsing Network.” IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp.6230-6239.
• [6] M.A. Al-masni and D.H. Kim, “CMM-Net: Contextual multi-scale multi-level network for efficient biomedical image segmentation,” Scientific Reports, vol.11, no.1, pp. 10191, 2021.
• [7] O. Yıldız, "Melanoma detection from dermoscopy images with deep learning methods: A comprehensive study," Journal of the Faculty of Engineering and Architecture of Gazi University , vol.34, no.4, pp.2241-2260, 2019.
• [8] H. Bingol ve B. Alatas , "Classification of brain tumor ımages using deep learning methods," Turkish Journal of Science and Technology, c. 16, sayı. 1, ss. 137-143, 2021.
• [9] I. Rizwan I Haque and J. Neubert, “Deep learning approaches to biomedical image segmentation,” Informatics in Medicine Unlocked, vol.18, pp. 100297, 2020.
• [10] F. Bulut, İ. Kılıç ve F. İnce, “Beyin tümörü tespitinde görüntü bölütleme yöntemlerine ait başarımların karşılaştırılması ve analizi,” Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi, c. 20, ss. 173-186, 2018.
• [11] R. Kijowski, F.Liu, F. Caliva and V.Pedoia, "Deep learning for lesion detection, Progression, and Prediction of Musculoskeletal Disease," J Magn Reson Imaging, vol. 52, no. 6, pp.1607-1619, 2020.
• [12] Y. Hou, Z. Liu, T. Zhang and Y. Li, "C-UNet: Complement UNet for remote sensing road extraction," Sensors (Basel, Switzerland), vol. 21, no. 6, pp. 2153, 2021.
• [13] Y.Fisher and K.Vladlen, "Multi-Scale context aggregation by dilated convolutions," CoRR, vol. 1511.07122, 2016.
• [14] M. A. Al-masni, M. A. Al-antari, M.T. Choi, S.M. Han and T.S. Kim, " Skin lesion segmentation in dermoscopy images via deep full resolution convolutional networks," Comput Methods Programs Biomed. vol.162, pp. 221-231, 2018.
• [15] S. Bakas, M. Reyes, A. Jakab, S. Bauer, M. Rempfler, A. Crimi, "Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge," 2018, arXiv:1811.02629.
• [16] Chen, L.Chieh and Zhu, Yukun and Papandreou, George and Schroff, Florian and Adam, Hartwig. “Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation.” 2018, arXiv 1802.02611.
• [17] Isensee, F., Kickingereder, P., Wick, W., Bendszus, M. ve Maier-Hein, ”No new-net.”, International MICCAI Brainlesion Workshop, 2018, pp. 234–244.
• [18] N. M. Aboelenein, P. Songhao, A. Koubaa, A. Noor and A. Afifi, "HTTU-Net: hybrid two track U-Net for automatic brain tumor segmentation," IEEE Access, 2020, pp. 101406-101415.
• [19] J. Zhang, Z. Jiang, J. Dong, Y. Hou and B. Liu, "Attention Gate ResU-Net for Automatic MRI Brain Tumor Segmentation," in IEEE Access, vol. 8, pp. 58533-58545, 2020.
• [20] Gu, Zaiwang and Cheng, Jun and Fu, Huazhu and Zhou, Kang and Hao, "CE-Net: Context Encoder Network for 2D Medical Image Segmentation," in IEEE Transactions on Medical Imaging, vol. 38, no. 10, pp. 2281-2292, 2019.
• [21] R. Olaf and Fischer, P. Brox, Thomas,” U-Net: Convolutional Networks for Biomedical Image Segmentation,” in International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI),2015, pp.234-241.
• [22] Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh and J. Liang. “UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation.” IEEE Transactions on Medical Imaging, 2020, pp.1856-1867.