Deteksi Region of Interest Tulang pada Citra B-mode secara Otomatis Menggunakan Region Proposal Networks
Abstract
Bone imaging using ultrasound is a safe technique since it does not involve ionizing radiation and non-invasive. However, bone detection and localization to find its region of interest (RoI) is a challenging task because b-mode ultrasound images are characterized by high level of noise and reverberation artifacts. The image quality is user-dependent and the boundary between tissues is blurry, which makes it challenging to interpret images. In this paper, the deep learning approach using Region Proposal Networks was implemented to detect bone’s RoI in b-mode images. The Faster Region-based Convolutional Neural Network model was fine-tuned to detect and determine the bone location in b-mode images automatically. To evaluate the results, in-vivo experiments were carried out using human arm specimens. A total of 1,066 b-mode bone images from six different subjects were used in the training phase and testing phase. The proposed method was successful in determining the bone RoI with the value of the mAP, the accuracy of detection, and the accuracy of localization of 0.87, 98.33%, and 95.99% respectively.
References
D.A. Dharmawan, "Deteksi Kanker Serviks Otomatis Berbasis Jaringan Saraf Tiruan LVQ dan DCT," J. Nas. Tek. Elektro dan Teknol. Inf. Vol. 3, No. 4, hal. 269–272, 2014.
N.P. Husain dan C. Fatichah, "Segmentasi Citra Sel Tunggal Smear Serviks Menggunakan Radiating Component Normalized Generalized GVFS," J. Nas. Tek. Elektro dan Teknol. Inf., Vol. 6, No. 1, hal. 107–114, 2017.
N. Syakrani, Y. Widhiyasana, dan A.A. Efendi, "Deteksi Tumor Hati dengan Graph Cut dan Taksiran Volume Tumornya," J. Nas. Tek. Elektro dan Teknol. Inf., Vol. 7, No. 1, hal. 35-43,2018.
O. Herliana, T.S. Widodo, dan I. Soesanti, "Klasifikasi Nomsupervised Citra Thermal Kanker Payudara Berbasis Fuzzy C-MEANS," J. Nas. Tek. Elektro dan Teknol. Inf., Vol. 1, No. 3, hal. 55-59, 2012.
T. Karlita, E.M. Yuniarno, I.K.E. Purnama, dan M.H. Purnomo, "Automatic Bone Outer Contour Extraction from B-Modes Ultrasound Images Based on Local Phase Symmetry and Quadratic Polynomial," Second Int. Work. Pattern Recognit. (IWPR 2017) 2017, pp. 165–170.
P.J.S. Gonçalves dan P. Torres, "Extracting Bone Contours in Ultrasound Images: Energetic Versus Probabilistic Methods," Rom. Rev. Precis. Mech. Opt. Mechatronics. Vol. 20, No. 37, hal. 105–110, 2010.
I. Hacihaliloglu, P. Guy, A.J. Hodgson, dan R. Abugharbieh, "Automatic Extraction of Bone Surfaces from 3D Ultrasound Images in Orthopaedic Trauma Cases," Int. J. Comput. Assist. Radiol. Surg., Vol. 10, hal. 1279–1287, 2015.
J. Kowal, C. Amstutz, F. Langlotz, H. Talib, dan M.G. Ballester, Automated Bone Contour Detection in Ultrasound B-Mode Images For Minimally Invasive Registration in Computer-Assisted Surgery – An In Vitro Evaluation," Int. J. Med. Robot. Comput. Assist. Surg. MRCAS, Vol. 3, No. 4, hal. 341–348, 2007.
R.W. Prager, R.N. Rohling, A.H. Gee, dan L. Berman, "Rapid Calibration for 3-D Freehand Ultrasound," Ultrasound Med. Biol. Vol. 24, No. 6, hal. 855–869, 1998.
A.K. Jain dan R.H. Taylor, "Understanding Bone Responses in B-Mode Ultrasound Images and Automatic Bone Surface Extraction Using a Bayesian Probabilistic Framework," Proc. SPIE, Med. Imaging 2004 Ultrason. Imaging Signal Process., 2004, Vol. 5373, hal. 131-142.
V. Chan dan A. Perlas, "Basics of Ultrasound Imaging," in Atlas Ultrasound-Guided Proced. Interv. Pain Manag., S.N. Narouze, Ed., Toronto, ON, Canada, Springer Science+Business Media, 2011, hal. 13–20.
K.E. Purnama, M.H.F. Wilkinson, A.G. Veldhuizen, P.M.A. Van Ooijen, J. Lubbers, J.G.M. Burgerhof, T.A. Sardjono, dan G.J. Verkerke, "A Framework for Human Spine Imaging Using a Freehand 3D Ultrasound System," Technol. Heal. Care., Vol. 18, No. 1, hal. 1–17. 2010.
N. Baka, S. Leenstra, dan T. van Walsum, "Random Forest-Based Bone Segmentation in Ultrasound," Ultrasound Med. Biol., Vol. 43, No. 10, hal. 2426-2437, 2017.
N. Quader, A. Hodgson, dan R. Abugharbieh, Confidence Weighted Local Phase Features for Robust Bone Surface Segmentation in Ultrasound, Lect. Notes Comput. Sci. (Including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), Cham, Switzerland: Springer, 2014, Vol. 8680, hal. 76–83.
R. Jia, S.J. Mellon, S. Hansjee, A.P. Monk, D.W. Murray, dan J.A. Noble, "Automatic Bone Segmentation in Ultrasound Images Using Local Phase Features and Dynamic Programming," IEEE 13th Int. Symp. Biomed. Imaging, 2016, hal. 1005–1008.
F. Berton, F. Cheriet, M.C. Miron, dan C. Laporte, "Segmentation of The Spinous Process and Its Acoustic Shadow in Vertebral Ultrasound Images," Comput. Biol. Med. Vol. 72, hal. 201–211, 2016.
L. Lopez-Perez, J. Lemaitre, A. Alfiansyah, dan M.-E. Bellemare, "Bone Surface Reconstruction Using Localized Freehand Ultrasound Imaging," 30th Annual International IEEE EMBS Conference, 2008, hal. 2964-2967.
I. Hacihaliloglu, R. Abugharbieh, A.J. Hodgson, dan R.N. Rohling, "Bone Surface Localization in Ultrasound Using Image Phase-Based Features," Ultrasound Med. Biol., Vol. 35, No. 9, hal. 1475–1487, 2009.
D. Yang, S. Zhang, Z. Yan, C. Tan, K. Li, dan D. Metaxas, "Automated Anatomical Landmark Detection on Distal Femur Surface Using Convolutional Neural Network," Proc. - Int. Symp. Biomed. Imaging, 2015, hal. 17–21.
H. Ravishankar, S.M. Prabhu, V. Vaidya, dan N. Singhal, "Hybrid Approach for Automatic Segmentation of Fetal Abdomen from Ultrasound Images Using Deep Learning," Proc. - Int. Symp. Biomed. Imaging, 2016, hal. 779–782.
J.C. Nascimento dan G. Carneiro, Multi-Atlas Segmentation Using Manifold Learning with Deep Belief Networks, Proc. - Int. Symp. Biomed. Imaging, 2016, hal. 867–871.
G. Carneiro dan J.C. Nascimento, "Combining Multiple Dynamic Models and Deep Learning Architectures for Tracking the Left Ventricle Endocardium in Ultrasound Data," IEEE Trans. Pattern Anal. Mach. Intell., Vol. 35, No. 11, hal. 2592–2607, 2013.
Y. Gao, M.A. Maraci, dan J.A. Noble, "Describing Ultrasound Video Content Using Deep Convolutional Neural Networks," Proc. - Int. Symp. Biomed. Imaging, 2016, hal. 787–790.
P.M.B. Torres, J.M. Sanches, P.J.S. Goncalves, dan J.M.M. Martins, "3D Femur Reconstruction Using a Robotized Ultrasound Probe," Proc. IEEE RAS EMBS Int. Conf. Biomed. Robot. Biomechatronics., 2012, hal. 884–888.
S. Ren, K. He, R. Girshick, dan J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," ArXiv Prepr. ArXiv1506.01497., Vol. 74, hal. 1–14, 2015.
(2018) "Colaboratory - Frequently Asked Questions," [Online] https://research.google.com/colaboratory/faq.html, tanggal akses: 12-Nov-2018.
T. Carneiro, R. Victor, M. Da, T. Nepomuceno, G. Bian, dan V.H.C.D.E. Albuquerque, "Performance Analysis of Google Colaboratory as a Tool for Accelerating Deep Learning Applications," IEEE Access Trends, Perspect. Prospect. Mach. Learn. Appl. to Biomed. Syst. Internet Med. Things., Vol. 6, hal. 61677–61685, 2018.
H. Gao, (2017) "Faster R-CNN Explained - Medium," [Online] https://medium.com/@smallfishbigsea/faster-r-cnn-explained-864d4fb7e3f8, tanggal akses: 12-Nov-2018.
© Jurnal Nasional Teknik Elektro dan Teknologi Informasi, under the terms of the Creative Commons Attribution-ShareAlike 4.0 International License.
1.png)
