Segmentation of Mammary Lesions in Ultrasound Images Applying Mask R-CNN
Resumo
Breast cancer is the most frequent malignant tumor in women and one of the most common in the world. One of the most important issues in this condition is early detection. Computer-aided diagnostic (CAD) systems are objects of research, aiming to provide a second opinion to the health professional. A fundamental aspect within the CAD system is the segmentation of the lesion, allowing an adequate extraction of the lesion characteristics. The use of a computerized segmentation method helps eliminate human variability and, consequently, improve the performance of the lesion classifier. Convolutional Neural Networks (CNNs) are being used in segmentation problems, such as various types of medical imaging, people and road signs detection, for example. So inspired by these promissing results, the present work has as main objective to analyze and implement the Mask R-CNN as a tool of segmentation of mammary lesions in images obtained by ultrasound, to propose an efficient method of segmentation, aiding in the classification process of the CAD systems.
Referências
RA. Smith, KS. Andrews, D. Brooks, SA. Fedewa, D. Manassaram-Baptiste, D. Saslow, OW. Brawley and RC. Wender, Cancer screening in the United States, 2017: a review of current American Cancer Society guidelines and current issues in cancer screening. CA: a cancer journal for clinicians, 2017.
M. Giger Computer-aided diagnosis of breast lesions in medical images. Computing in Science & Engineering, 2000.
L. Tabr, B. Vitak, T. Chen, A. Yen, A. Cohen, T. Tot, S. Chiu, S. Chen, J. Fann, J. Rosell, H. Fohlin, R. Smith, S. Duffy, Swedish Two-County Trial: Impact of Mammographic Screening on Breast Cancer Mortality during 3 Decades. Radiology:Imaging Cancer, 2011, https://doi.org/10.1148/radiol.11110469
K. Kazemi, A. Adibi, M. Golshahi and M. Sirus, Breast cancer screening: Evidence of the effect of adjunct ultrasound screening in women with unilateral mammography-negative dense breasts. J Res Med Sci. 2015 Mar; 20(3): 228 232.
G. Albonico, Seleo de Atributos e Classificao Automtica de Leses Mamrias em Imagens de Ultrassom. Dissertation (Master degree) - State University of Western Paran, Brazil, 2017
W. Flores, Desarrollo de una Metodologa Computacional para la Clasificacin de Lesiones de Mama en Imgenes Ultrasnicas. Thesis (Doctorate) - Center for Research and Advanced Studies, National Polytechnic Institute. Mexico, 2009
W. Flores, W. Pereira and A. Catelli, Improving classification performance of breast lesions on ultrasonography. Pattern Recognition, Elsevier, 2015.
J. Rogowska, Overview and fundamentals of medical image segmentation. Handbook of Medical Imaging Processing and Analysis, Academic Press, San Diego, 2000
M. Calas, R. Almeida, B. Gutfilen and W. Pereira, Interobserver concordance in the BI-RADS classification of breast ultrasound exams. Clinics, SciELO Brasil, 2012.
RM. Menchon and JL. Sancho, Fully automatic segmentation of ultrasound common carotid artery images based on machine learning. Neurocomputing, 2015.
M. Xian, Y. Zhang and H. Cheng, Fully automatic segmentation of breast ultrasound images based on breast characteristics in space and frequency domains. Pattern Recognition, Elsevier, 2015.
Y. Li, Segmentation of Medical Ultrasound Images Using Convolutional Neural Networks with Noisy Activating Functions You. Stanford Edu, 2016
A. Dutta, A. Gupta and A. Zissermann. VGG Image Annotator (VIA), 2016. http://www.robots.ox.ac.uk/ vgg/software/via/
S. Ren, K. He, R. Girshick, and J. Sun, Faster R-CNN: To- wards real-time object detection with region proposal networks. NIPS, 2015.
K. He, R. Gkioxari, G. Dollar and P. Girshick, Mask R-CNN. Proceedings of the IEEE International Conference on Computer Vision, 2017. https://doi.org/10.1109/ICCV.2017.322
A. Rahman and Y. Wang, Advances in Visual Computing, 2013. https://doi.org/10.1007/978-3-642-41939-3
