نوع مقاله : مقاله کامل پژوهشی

نویسندگان

1 دانشجوی کارشناسی ارشد، دانشکده مهندسی برق، پزشکی و مکاترونیک، واحد قزوین، دانشگاه آزاد اسلامی، قزوین، ایران

2 دانشیار مهندسی پزشکی، پژوهشکده برق و فناوری اطلاعات، سازمان پژوهش های علمی و صنعتی ایران، تهران، ایران

چکیده

تشخیص مرز ضایعات، اولین گام در فرآیند تشخیص هوشمند ضایعات در تصاویر درموسکوپی است و به همین دلیل می‌تواند تأثیر مستقیم بر دقت و صحت مراحل بعدی بگذارد. متأسفانه، استخراج مرز ضایعات با محدودیت­هایی از قبیل وجود مرز­های نامنظم، کنتراست ضعیف در برخی نواحی و وجود آرتیفکت مواجه‌ است. هدف از این مقاله، ارائة نسخه­ای بهبود­یافته از تکنیک بهینه­سازی تابع انرژی برای تفکیک ضایعات از پوست در فرآیند پردازش تصاویر درموسکوپی است. این نسخه، از ایده­ای برمبنای مفهوم راستاهای شعاعی در روند تکامل کانتور استفاده می‌کند، که کاهش حساسیت تخمین مرز ضایعات به محدودیت‌های ذکرشده را به دنبال دارد. عملکرد روش ارائه‌شده روی مجموعه دادگانی از تصاویر درموسکوپی گرفته‌شده از ضایعات مختلف با اندازه و مرزهای متفاوت، آزموده شده و نتایج حاصل از این روش با استفاده از معیارهای استاندارد با نتایج چند روش رقیب مقایسه می­شود. افزایش نرخ تشخیص درست به میزان 6.17 درصد به موازات کاهش فاصلة هامود تا حدود 3/2 درصد توسط در روش پیشنهادی نسبت به نزدیک‌ترین رقیبش، نشان‌دهندة بهبود فرآیند تشخیص مرز ضایعه درتصاویر درموسکوپی نسبت به روش‌های موجود است.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

A New Method for Segmentation of Skin Lesions in Dermoscopy Images: Energy Optimization Based on the Radial Directions

نویسندگان [English]

  • Marjan Iranianpour Haghighi 1
  • Seyyed Vahab Shojaeddini 2

1 MSC Student in Electrical Engineering, Faculty of Electrical, Biomedical and Mechatronics Engineering, Qazvin Branch, Islamic Azad University, Qazvin, Iran

2 Associate Professor of Biomedical Engineering, Department of Electrical Engineering and Information Technology, Iranian Research Organization for Science and Technology, Tehran, Iran

چکیده [English]

Detecting lesion borders is the first step for intelligent lesion identification in dermoscopy, therefore it may influence the accuracy and validity of the next steps of this process. Unfortunately, extracting borders is hampered by some challenges such as losses associated with irregular borders, poor contrast, and artifacts encountered in some area. In this paper, the improved version of energy function optimization technique is introduced in order to separate the skin and lesions in the processing of dermoscopy images. This technique is based on the concept of radial directions in the contour development process, which reduces the sensitivity of estimating the boundaries of lesions to the above constraints. The performance of the proposed method is evaluated on a dataset of dermoscopy images which are captured from various lesions with different sizes and boundaries. The obtained results of the proposed method are compared with some other state-of-the-art lesion detection methods by using standard parameters. Increased True Detection Rate by 6.17% in parallel with decrease in Hammoud Distance by 2.3%, both compared to the best among alternative methods shows the effectiveness of the proposed scheme in detecting lesion borders of dermoscopy images.

کلیدواژه‌ها [English]

  • Border Detection
  • Energy Function Optimization Algorithm
  • Radial Directions
  • Dermoscopy
  • Computer Aided Diagnosis
[1]     Ferlay, J., Eva Steliarova-Foucher, Joannie Lortet-Tieulent, Sonia Rosso, J. W. W. Coebergh, H. Comber, David   Forman, and Freddie Bray. " Reprintof: Cancer incidence and mortality patterns in Europe: estimates for 40 countries in 2012." European journal of cancer 51, no. 9 (2015): 1201-1202.
[2]     Binder, Michael, Margot Schwarz, Alexander Winkler, Andreas Steiner, Alexandra Kaider, Klaus Wolff, and Hubert Pehamberger. "Epiluminescence microscopy: a useful tool for the diagnosis of pigmented skin lesions for formally trained dermatologists." Archives of Dermatology 131, no. 3 (1995): 286-291.
[3]     Mayer, Justine. "Systematic review of the diagnostic accuracy of dermatoscopy in detecting malignant melanoma." The Medical Journal of Australia 167, no. 4 (1997): 206-210.
[4]     Silveira, Margarida, Jacinto C. Nascimento, Jorge S. Marques, André RS Marçal, Teresa Mendonça, Syogo Yamauchi, Junji Maeda, and Jorge Rozeira. "Comparison of segmentation methods for melanoma diagnosis in dermoscopy images." IEEE Journal of Selected Topics in Signal Processing3, no. 1 (2009): 35-45.
[5]     Ganster, Harald, P. Pinz, Reinhard Rohrer, Ernst Wildling, Michael Binder, and Harald Kittler. "Automated melanoma recognition." IEEE transactions on medical imaging 20, no. 3 (2001): 233-239.
[6]     Premaladha, J., M. Lakshmi Priya, S. Sujitha,   and K. S. Ravichandran. "Normalised Otsu's Segmentation Algorithm for Melanoma Diagnosis." Indian Journal of Science and Technology 8, no. 22 (2015): 1-6.         
[7]     Hongmin, Zhong Yang, Xinhua Cao, Weiming Xia, and Xiaoyin Xu. "A new iterative triclass thresholding technique in image segmentation." IEEE Transactions on Image Processing 23, no. 3 (2014): 1038-1046.
[8]     Otsu, Nobuyuki. "A threshold selection method from gray-level histograms." Automatica 11, no. 285-296 (1975): 23-27.
[9]     Pennisi, Andrea, Domenico D. Bloisi, Daniele Nardi, Anna Rita Giampetruzzi, Chiara Mondino, and Antonio Facchiano. "Skin lesion image segmentation using Delaunay Triangulation for melanoma detection." Computerized Medical Imaging and Graphics 52 (2016): 89-103.
[10] Erkol, Bulent, Randy H. Moss, R. Joe Stanley, William V. Stoecker, and Erik Hvatum. "Automatic lesion boundary detection in dermoscopy images using gradient vector flow snakes." Skin Research and Technology 11, no. 1 (2005): 17-26.
[11] Zhou, Huiyu, Xuelong Li, Gerald Schaefer, M. Emre Celebi, and Paul Miller. "Mean shift based gradient vector flow for image segmentation." Computer Vision and Image Understanding 117, no. 9 (2013): 1004-1016.
[12] Nock, Richard, and Frank Nielsen. "Statistical region merging." IEEE Transactions on pattern analysis and machine intelligence 26, no. 11 (2004): 1452-1458.
[13] Maeda, J., A. Kawano, S. Yamauchi, Y. Suzuki, A. R. S. Marçal, and T. Mendonça. "Perceptual image segmentation using fuzzy-based hierarchical algorithm and its application to dermoscopy images." In Soft Computing in Industrial Applications, 2008. SMCia'08. IEEE Conference on, pp. 66-71. IEEE, 2008.
[14] Abbas, Qaisar, Irene Fondón, and Muhammad Rashid. "Unsupervised skin lesions border detection via two-dimensional image analysis." Computer methods and programs in biomedicine 104, no. 3 (2011): e1-e15.
[15] Osher, Stanley, and Ronald Fedkiw. Level set methods and dynamic implicit surfaces. Vol. 153. Springer Science & Business Media, 2006.
[16] Chan, Tony F., and Luminita A. Vese. "Active contours without edges." IEEE Transactions on image processing 10, no. 2 (2001): 266-277.
[17]  Lie, Johan, Marius Lysaker, and Xue-Cheng Tai. "A binary level set model and some applications to Mumford-Shah image segmentation." IEEE Transactions on Image Processing 15, no. 5 (2006): 1171-1181.
[18] Glaister, Jeffrey, Alexander Wong, and David A. Clausi. "Segmentation of skin lesions from digital images using joint statistical texture distinctiveness." IEEE transactions on biomedical engineering 61, no. 4 (2014): 1220-1230.
[19] Razi J. Al-azawi, Abbas Abdulazez Abdulhameed, Hussein Majeed Ahmed, "A Robustness Segmentation Approach for Skin Cancer Image Detection Based on an Adaptive Automatic Thresholding Technique", American Journal of Intelligent Systems, Vol. 7 No. 4, (2017), pp. 107-112.
[20] Flores, Eliezer, and Jacob Scharcanski. "Segmentation of melanocytic skin lesions using feature learning and dictionaries." Expert Systems with Applications 56 (2016): 300-309.