Document Type : Full Research Paper

Authors

1 K. N. Toosi University of Technology

2 Tehran University of Medical Science

3 Tarbiat Modarres University

10.22041/ijbme.2005.13523

Abstract

This paper presents a CAD system for detection and diagnosis of microcalcification clusters in mammograms. The proposed algorithm is composed of three main stages. In the first stage, the image pixels are examined for corresponding to individual microcalcification objects. For this purpose, the wavelet transform of the image is computed. Then two wavelet coefficients as well as two statistical features are used with a neural network for a primary classification of the image pixels. In the second stage, some noisy pixels extracted by the first step are eliminated. Then 18 features defined for each microcalcification are used with a nonlinear classifier for accurate detection of microcalcifications. For training of this classifier we used 16 regions from a database containing 379 microcalcifications. Finally, in the third stage five features defined for each microcalcification cluster with a neural network are used to recognize malignant microcalcification clusters. For training of this network, 22 clusters including 8 malignant and 14 benign cases were used. The performance of the algorithm was evaluated using a separate image set composed of 22 clusters including 10 malignant and 12 benign cases. Using these tests images and the threshold value of 0.45, the sensitivity of the algorithm was 100% and its specificity was 91.6%.

Keywords

Main Subjects

[1]     Kocur C, Rogers S, Myers L, Burns T, Kabrisky M, Hoffmeister J, Bauer K, Steppe J; Using neural networks to select wavelet features for breast cancer diagnosis; IEEE Engineering in Medicine and Biology Magazine 1996; 15(3): 95-102.
[2]     Netsch T, Peitgen H; Scale-space signatures for the detection of clustered microcalcifications in digital mammograms; IEEE Transactions on Medical Imaging 1999, 18(9): 774-786.
[3]     Olsan SL; Breast calcifications analysis of imaging properties; Radiology 1988, 169(2): 329-333.
[4]     Betal D, Roberts N; Segmentation and numerical analysis of microcalcifications on mammograms using mathematical morphology; British Journal of Radiology 1997; 1: 903-917.
[5]     Qian W; Digital mammography: Hybrid four channel wavelet transform for microcalcifications segmentation; Academic Radiology 1998; 5: 354-364.
[6]     Gavrielides M, Lo J; Segmentation of suspicious clustered microcalcifications in mammograms; Medical Physics 2000; 27: 13-22.
[7]     Verma B, Zakos J; A computer-aided diagnosis system for digital mammograms based on fuzzy-neural and feature extraction techniques; IEEE Tran Information Technology in Biomed. 2001; 5(1): 46-54.
[8]     Desarmaud S, Sehad S, Strauss A; Artificial neural networks in mammography: Application to malignant and benign clustered microcalcifications classification; Third International Conference on Neural Networks and Their Applications, France,1997.
[9]     Aghdasi F, Ward RK, Morgan-Parkes J, Palcic B; Feature selection for classification of mammographic microcalcification clusters; Proc of the Annual Intl Conference of the IEEE Engineering in Medicine & Biology Society, San Diego 1993; 15: 58-59.
[10] Jiang Y, Nishikawa RM, Wolverton DE, Metz CE, Giger ML, Schmidt RA, Vyborny CJ, Doi K; Malignant and benign clustered microcalcifications: Automated feature analysis and classification; Radiology 1996; 198: 671-678.
[11] Jiang Y, Nishikawa RM, Wolverton DE, Giger ML, Doi K, Schmidt RA, Vyborny CJ; Computerized classification of malignant and benign clustered microcalcification in mammograms; Proc of the 19th Annual International Conf on Engineering in Medicine and Biology Society 1997; 1: 594-599.
[12] De Vito S, Vento M, Tortorella F; Automatic classification of clustered microcalcifications by a multiple expert system; Proc 10th Intl Conference on Image Analysis and Processing, Italy 1999; 27-29.
[13] Abdolmaleki P, Buadu LD, Nadermanesh H; Feature extraction and classification of breast cancer on dynamic magnetic resonance imaging using artificial neural network; Cancer Letters 2001; 171(2): 183-191.
[14] پورعبدالله نژاد س، سلطانیان زاده ح؛ استفاده از ویژگی های ظاهری میکروکلسیفیکاسیون ها به منظور طبقه بندی خوشه های میکروکلسیفیکاسیون در تصاویر ماموگرافی؛ کنفرانس مهندسی برق، اصفهان، 1379، 135-142.
[15] پورعبدالله نژاد س، سیدین س ع؛ آشکارسازی و تفکیک میکروکلسیفیکاسیون ها در ماموگرام با استفاده از بانک فیلتر وفقی و فیلترهای همومورفیک؛ هفتمین کنفرانس مهندسی برق، تهران 1378.
[16] سپهر ح، سیدین س ع؛ تشخیض میکروکلسیفیکاسیون های بدخیم با استفاده از بسته های ویولتی و ویژگی های بافت تصویر در ماموگرام های دیجیتال؛ پنجمین کنفرانس بین‌المللی انجمن کامپیوتر ایران، اسفند 1378.
[17] پورعبدالله نژاد س، سیدین س ع؛ روش هایی برای بهبود کیفیت تصاویر ماموگرافی در جهت تشخیص میکروکلسیفیکاسیون ها؛ هفتمین کنفرانس مهندسی برق، تهران 1378.
[18] Strickland RN, Hahn HI; Wavelet transform for detecting microcalcifications in mammograms; IEEE Trans Med Imag 1996; 15(4): 218-229.
[19] Mallat S; A Wavelet Tour of Signal Processing; Academic Press, 1998.
[20] ابریشمی مقدم ح؛ تبدیل موجک و کاربردهای آن در پردازش تصویر؛ انتشارات دانشکده برق، دانشگاه صنعتی خواجه نصیرالدین طوسی، 1380.
[21] ابریشمی مقدم ح، مصطفی ع، گیتی م؛ بهبود تصاویر ماموگرافی به کمک تبدیل موجک؛ اولین کنفرانس ماشین بینایی و پردازش تصویر، بیرجند 1379.
[22] ابریشمی مقدم ح، مصطفی ع، گیتی م؛ آشکارسازی میکروکلسیفیکاسیون ها در تصاویر ماموگرافی به کمک تبدیل موجک و شبکه های عصبی؛ نهمین کنفرانس مهندسی برق، تهران 1380.
[23] Hagan MT, Menhaj MB; Training feedforward networks with the marquardt algorithm; IEEE Trans Neural Networks 1994; 5(6): 989-993.
[24] شیخ حسنی ع؛ تشخیص خودکار تصاویر ماموگرافی با استفاده از تبدیل موجک و شبکه های عصبی؛ پایان نامه کارشناسی ارشد، دانشکده برق، دانشگاه صنعتی خواجه نصیرالدین طوسی، بهار 1381.
[25] Heath M, Bowyer K, Kopans D, Moore R, Kegelmeyer JP; The digital database for screening mammography; Proc of the 5th International Workshop on Digital Mammography, Medical Physics Publishing (Madison, WI), Toronto, Canada, June 2000.