تشخیص، دسته‌بندی و اندازه‌گیری اتوماتیک ندول‌های ریوی با استفاده از دسته‌بند ترکیبی در تصاویر سی تی اسکن

نوع مقاله : علمی-پژوهشی

نویسندگان

1 دانشکده مهندسی - دانشگاه شهید چمران اهواز

2 دانشکده پیراپزشکی - دانشگاه علوم پزشکی جندی شاپور

چکیده

سرطان ریه یکی از سخت‌ترین و خطرناک‌ترین سرطان‌ها به شمار می‌رود که می‌تواند در مراحل اولیه، به صورت یک جسم کوچک با قطری کمتر از سه سانتی متر، بنام ندول، مشاهده شود. این ندول‌ها به دو دسته‌ی خوش‌خیم و بدخیم یا سرطانی تقسیم‌بندی می‌شوند. در این مقاله، یک سیستم تشخیصی جهت شناسایی و دسته‌بندی ندول‌های ریوی پیشنهاد می‌شود، که در فاز اول ریه‌ها از تصویر سی‌تی‌اسکن طی عملیات ناحیه‌بندی کانتور فعال جدا می‌شوند. سپس براساس ویژگی‌های سیفت (SIFT)، دسته‌بند بگینگ پیشنهادی تصاویر ریه را به دو دسته سالم و بیمار دسته‌بندی می‌کند. در فاز دوم، براساس یک ناحیه‌بندی گراف کات تمام خودکار، ندول‌ها از تصویر ریه استخراج شده و قطر آن‌ها اندازه‌گیری می‌شود. در پایان، ندول‌ها براساس اندازه و ویژگی‌های بافتی تصویر (هارالیک) به دو دسته خوش‌خیم و بدخیم طبقه‌بندی می‌شوند. جهت ارزیابی عملکرد روش پیشنهادی، از تصاویر مجموعه داده LIDC استفاده گردید و کارایی آن در شناسایی ندولها و در مقایسه با روش‌های دیگر با معیار دقت 97% و از نظر طبقه‌بندی ندول‌ها به خوش‌خیم و بدخیم با دقت 96% قابل رقابت است. 

کلیدواژه‌ها


عنوان مقاله [English]

Automatic Detection, Classification and Measurement of Lung Nodules using Combined Classifiers in CT Scan Images

نویسندگان [English]

  • M. Naderan 1
  • A. Jamshidnejad 2
  • N. Mirderikvand 1
1 Department of Computer Engineering, Shahid Chamran University of Ahvaz, Ahvaz, Iran
2 Department of Medical Informatics, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
چکیده [English]

Lung cancer is one of the hardest and most dangerous types of known cancer in the world which can be detected in its beginning stages as a small mass of tissue, less than 3 cm in diameter, called a nodule. These nodules are classified to two classes of benign or malignant. In this paper, a detection system for detection and classification of lung nodules is proposed which in the first phase, lungs are separated from the CT scan images according to the active contour segmentation method. Next, based on the SIFT features the proposed Bagging classifier, classifies the lung images into two classes of patient and healthy. In the second phase, according to a fully automatic Graph-Cut segmentation method the nodules are extracted from patient images and their diameters are measured. Finally, nodules are classified to two classes of benign and malignant based on their size and texture Haralick features. To evaluate the proposed method, images of the LIDC database are used and its performance in detection of nodules compared to other methods has an accuracy of 97% and in classification of nodules to benign and malignant an accuracy of 96% is reached.  

کلیدواژه‌ها [English]

  • Lung nodules
  • CT images
  • Graph-cut segmentation method
  • SIFT features
  • Haralick features
  • Combined classifiers
[1]      R. L. Siegel, K. D. Miller and A. Jemal, “Cancer Statistics,” CA Cancer J Clin., vol. 65, pp. 5-29, 2015.
[2]      C. N. Carter, CAD System for Lung Nodule Analysis, Master’s Thesis, University of Louisville, 2012.
[3]      N. M. Noor, J. C. M. Than, O. M. Rijal, R. M. Kassim, A. Yunus, A. A. Zeki, M. Anzidei, L. Saba and J. S. Suri, “Automatic Lung Segmentation Using Control Feedback System,” Journal of Medical Systems, vol. 39, pp. 22-40, 2015.
[4]      S. Biradar and K. Agalatakatti, “Lung Cancer Identification Using CT Images,” International Journal Of Engineering And Computer Science, vol. 4, pp. 13022-13025, 2015.
[5]      F. Zhang, Y. Song, W. Cai, M.-Z. Lee, Y. Zhou, H. Huang, S. Shan, M. J. Fulham and D. D. Feng, “Lung Nodule Classification With Multilevel Patch-Based Context Analysis,” IEEE Transactions on Biomdecal Engineering, vol. 61, 2014.
[6]      N. Camarlinghi, “Automatic detection of lung nodules incomputed tomography images: training and validation ofalgorithms using public research databases,” European Physical Journal Plus, vol. 128, no. 9, 2013.
[7]      R. Kakinuma, K. Ashizawa, T. Kobayashi, A. Fukushima, H. Hayashi, T. Kondo, M. Machida, M. Matsusako, K. Minami, K. Oikado, M. Okuda, S. Takamatsu, M. Sugawara, S. Gomi, Y. Muramatsu and K. Hanai, “Comparison of sensitivity of lung nodule detection between radiologists and technologists on low-dose CT lung cancer screening images,” Brtitain Journal on Radiology, vol. 85, 2012.
[8]      S. Namin Taghavi, H. Abrishami Moghaddam, R. Jafari, M. Esmaeil-Zadeh and M. Gity, “Automated detection and classification of pulmonary nodules in 3D thoracic CT images,” 2010 IEEE International Conference on Systems Man and Cybernetics (SMC), Istanbul, Turkey, 2010.
[9]      K. Suzuki, F. Li, S. Sone and K. Doi, “Computer-aided diagnostic scheme for distinction between benign and malignant nodules in thoracic low-dose CT by use of massive training artificial neural network,” IEEE Transactions on Medical Imaging,vol. 24, 2005.
[10]      J. Kim, j. Fisher, A. Yezzi, M. Cetin and A. Willsky, “A nonparametric statistical method for image segmentation using information theory and curve evolution,” IEEE Transactions on Image Processing,vol. 14, pp. 1486 - 1502, 2005.
[11]      T. McInerney and D. Terzopoulos, “Deformable Models in Medical Image Analysis: A Survey,” Medical Image Analysis,vol. 1, pp. 91–108, 1996.
[12]      R. Bellotti, D. Carlo, G. Gargano, S. Tangaro, D. Cascio, E. Catanzariti, P. Cerello5, S. Cheran, P. Delogu, D. Mitri, C. Fulcher, D. Grosso, A. Retico, S. Squarcia, E. Tommasi and G. Bruno, “A CAD system for nodule detection in low-dose lung CTs based on region growing and a new active contour model,” Medical Physics, vol. 34, 2007.
[13]      M. Keshani, Z. Azimifar, R. Boostani and A. Shakibafar, “Lung nodule segmentation using active contour modeling,” Machine Vision and Image Processing, Isfahan, 2010.
[14]      Y. Y. Boykov and M.-P. Jolly, “Interactive Graph Cuts for Optimal Boundary & Region Segmentation of Objects in N-D Images,” International Conference on Computer Vision, Vancouver, Canada, 2001.
[15]      K. Nag and N. R. Pal, “A Multiobjective Genetic Programming-Based Ensemble for Simultaneous Feature Selection
and Classification,” IEEE Transactions on Cybernetics, vol. 46, no. 2, pp. 499-510, 2016.
[16]      J. Mukherjee, A.K.Choudhury, A. J. Mukherjee, A. Chakrabarti, S. H. Shaikh and M. Kar, “Automatic Detection and Classification of Solitary Pulmonary Nodules from Lung CT Images,” Fourth International Conference of Emerging Applications of Information Technology (EAIT), Kolkata, India, 2014.
[17]      G. Wei, H. Ma, W. Qian, F. Han, H. Jiang, S. Qi, M. Qiu, “Lung nodule classification using local kernel regression models with out-of-sample extension,” Biomedical Signal Processing and Control, vol. 40, pp. 1-9, 2018.
[18]      J. Yuan, X. Liu, F. Hou, H. Qin, A. Hao, “Hybrid-feature-guided lung nodule type classification on CT images,” Computers & Graphics, vol. 70, pp. 288-299, 2018.
[19]      C. Bhuoaneswari, p. Aruna and D. Hoganathan, “Classification of Lung Diseases by image processing Techniques using computed Tomography Images,” International Journal of Advanced Computer Research, vol. 4, no. 1, issue 14, pp. 87-93, March 2014.
[20]      K. Yokota, R. Tachibana, Y. Hirano and S. Kido, “Automatic Detection of GGO Regions on CT Images in LIDC Dataset Based on Statistical Features,” Joint 7th International Conference on Sotf Computing and Intelligent Systems (SCIS) and 15th International Symposium on Advanced Intelligent Systems (ISIS), Japan, 2014.
[21]      W. J. Choi and T. S. Choi, “Automated pulmonary nodule detection based on three-dimensional shape-based feature descriptor,” Computer Methods and Programs in Biomedicine, vol. 113, no. 1, pp. 37-54, 2014.
[22]      A. M. Santosa, A. O. d. C. Filhoa, A. C. C. d. Paivaa, Nunesb, R. Acatauassú, Gattassc and Marcelo, “Automatic detection of small lung nodules in 3D CT data using Gaussian mixture models, Tsallis entropy and SVM,” Engineering Applications of Artificial Intelligence,vol. 36, pp. 27-39, 2014.
[23]      N. Sharma and Sheenam, “Size Estimation of Lung Cancer Using Image Segmentation and Back Propagation,” International Journal for Research in Technological Studies, vol. 1, no. 9,  pp. 14-17, 2014.
[24]      S. S. Makhanov, “Active contours in medical image processing, Theory and applications,” 5th International Conference on Knowledge and Smart Technology (KST), Thailand, 2013.
[25]      C. Xu and J. L. Prince, “Generalized gradient vector ßow external forces for active contours,” Signal Processing,vol. 71, pp. 131-139, 1998.
[26]      D. Lowe, “Distinctive image features from scale-invariant keypoints,” International Journal of Computer Vision, vol. 60, pp. 91-110, 2004.
[27]      ELCAP Public Lung Image Database, 2016. [Online]. Available: http://www.via.cornell.edu/lungdb.html.
[28]      S. G. Armato and et. al, “The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): A completed reference database of lung nodules on CT scans,” Medical Physics,vol. 38, pp. 915-931, 2011.
[29]      H. Guanghui, L. Xiabi, H. Feifei, S. N. Tenaya, Z. Yanfeng, Z. Xinming and Z. Chunwu, “The LISS—A Public Database of Common Imaging Signs of Lung Diseases for Computer-Aided Detection and Diagnosis Research and Medical Education,” IEEE Transactions on Biomedical Engineering,vol. 62, 2015.
[30]      I. R. S. Rafael, P. C. Cortez, E. C. Neto, J. M. Soares, V. H. C. de Albuquerque and J. M. Tavares, “Automatic 3D pulmonary nodule detection in CTimages: A survey,” Computer methods and programs in biomedicine, vol. 124, pp. 91-107, 2015.
[31]      A. El-Baz, A. Soliman, P. McClure, G. Gimel'farb, M. Abo El-Ghar and R. Falk, “Early assessment of malignant lung nodules based on the spatial analysis of detected lung nodules,” 9th IEEE International Symposium on Biomedical Imaging (ISBI), Barcelona, Espain, 2012.