Characterization of mammographic masses using a gradient-based segmentation algorithm and a neural classifier

Computerized methods have recently shown a great potential in providing radiologists with a second opinion about the visual diagnosis of the malignancy of mammographic masses. The computer-aided diagnosis (CAD) system we developed for the mass characterization is mainly based on a segmentation algorithm and on the neural classification of several features computed on the segmented mass. Mass-segmentation plays a key role in most computerized systems. Our technique is a gradient-based one, showing the main characteristic that no free parameters have been evaluated on the data set used in this analysis, thus it can directly be applied to data sets acquired in different conditions without any ad hoc modification. A data set of 226 masses (109 malignant and 117 benign) has been used in this study. The segmentation algorithm works with a comparable efficiency both on malignant and benign masses. Sixteen features based on shape, size and intensity of the segmented masses are extracted and analyzed by a multi-layered perceptron neural network trained with the error back-propagation algorithm. The capability of the system in discriminating malignant from benign masses has been evaluated in terms of the receiver-operating characteristic (ROC) analysis. A feature selection procedure has been carried out on the basis of the feature discriminating power and of the linear correlations interplaying among them. The comparison of the areas under the ROC curves obtained by varying the number of features to be classified has shown that 12 selected features out of the 16 computed ones are powerful enough to achieve the best classifier performances. The radiologist assigned the segmented masses to three different categories: correctly-, acceptably- and non-acceptably-segmented masses. We initially estimated the area under ROC curve only on the first category of segmented masses (the 88.5% of the data set), then extending the classification to the second subclass (reaching the 97.8% of the data set) and finally to the whole data set, obtaining A(z)=0.805+/-0.030, 0.787+/-0.024 and 0.780+/-0.023, respectively.

[1]  Sheng Liu,et al.  Multiresolution detection of spiculated lesions in digital mammograms , 2001, IEEE Trans. Image Process..

[2]  Michael A. Wirth,et al.  Segmentation of the breast region in mammograms using active contours , 2003, Visual Communications and Image Processing.

[3]  Taylor Murray,et al.  Cancer Statistics, 2001 , 2001, CA: a cancer journal for clinicians.

[4]  Germano C. Vasconcelos,et al.  Investigating feedforward neural networks with respect to the rejection of spurious patterns , 1995, Pattern Recognit. Lett..

[5]  N. Petrick,et al.  Computerized characterization of masses on mammograms: the rubber band straightening transform and texture analysis. , 1998, Medical physics.

[6]  J. Kelsey,et al.  Epidemiology of breast cancer. , 1990, Epidemiologic reviews.

[7]  R. Prevete,et al.  The MAGIC-5 Project: medical applications on a GRID infrastructure connection , 2004, IEEE Symposium Conference Record Nuclear Science 2004..

[8]  Terry E. Weymouth,et al.  Using Dynamic Programming For Minimizing The Energy Of Active Contours In The Presence Of Hard Constraints , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[9]  Adler Dd,et al.  Mammographic biopsy recommendations. , 1992 .

[10]  M. Stone.  Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[11]  K. Ebeling,et al.  [Epidemiology of breast cancer]. , 1988, Archiv fur Geschwulstforschung.

[12]  M L Giger,et al.  Computerized classification of benign and malignant masses on digitized mammograms: a study of robustness. , 2000, Academic radiology.

[13]  Lubomir M. Hadjiiski,et al.  Analysis of temporal changes of mammographic features: computer-aided classification of malignant and benign breast masses. , 2001, Medical physics.

[14]  Lubomir M. Hadjiiski,et al.  Improvement of mammographic mass characterization using spiculation meausures and morphological features. , 2001, Medical physics.

[15]  Rangaraj M. Rangayyan,et al.  Gradient and texture analysis for the classification of mammographic masses , 2000, IEEE Transactions on Medical Imaging.

[16]  Ruey-Feng Chang,et al.  Diagnosis of breast tumors with sonographic texture analysis using wavelet transform and neural networks. , 2002, Ultrasound in medicine & biology.

[17]  C. Metz.  ROC Methodology in Radiologic Imaging , 1986, Investigative radiology.

[18]  Joseph M. Reinhardt,et al.  Classification of mammographic masses: comparison between Backpropagation Neural Network (BNN) and human readers , 2003, SPIE Medical Imaging.

[19]  Tryphon Lambrou,et al.  Statistical Classification of Digital Mammograms Using Features from the Spatial and Wavelet Domains , 2002 .

[20]  Berkman Sahiner,et al.  Dual system approach to computer-aided detection of breast masses on mammograms. , 2006, Medical physics.

[21]  Maryellen L. Giger,et al.  Automated seeded lesion segmentation on digital mammograms , 1998, IEEE Transactions on Medical Imaging.

[22]  N. Karssemeijer,et al.  A new 2D segmentation method based on dynamic programming applied to computer aided detection in mammography. , 2004, Medical physics.

[23]  Matthew T. Freedman,et al.  Separation of malignant and benign masses using maximum-likelihood modeling and neural networks , 2002, SPIE Medical Imaging.

[24]  Thomas G. Dietterich.  Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms , 1998, Neural Computation.

[25]  P. Langenberg,et al.  Breast Imaging Reporting and Data System: inter- and intraobserver variability in feature analysis and final assessment. , 2000, AJR. American journal of roentgenology.

[26]  M. Giger,et al.  Automated computerized classification of malignant and benign masses on digitized mammograms. , 1998, Academic radiology.

[27]  Evangelos Dermatas,et al.  Neural classification of abnormal tissue in digital mammography using statistical features of the texture , 1999, ICECS'99. Proceedings of ICECS '99. 6th IEEE International Conference on Electronics, Circuits and Systems (Cat. No.99EX357).

[28]  Berkman Sahiner,et al.  Classification of malignant and benign masses based on hybrid ART2LDA approach , 1999, IEEE Transactions on Medical Imaging.

[29]  Wei Qian,et al.  Image feature extraction for mass detection in digital mammography: Influence of wavelet analysis , 1999 .

[30]  L. Bruce,et al.  Classifying mammographic mass shapes using the wavelet transform modulus-maxima method , 1999, IEEE Transactions on Medical Imaging.

[31]  Berkman Sahiner,et al.  Computer-aided characterization of mammographic masses: accuracy of mass segmentation and its effects on characterization , 2001, IEEE Transactions on Medical Imaging.

[32]  T. Key,et al.  Epidemiology of breast cancer. , 2001, The Lancet. Oncology.

[33]  Francesco Fauci,et al.  Search of microcalcification clusters with the CALMA CAD station , 2002, SPIE Medical Imaging.

[34]  Taylor Murray,et al.  Cancer statistics, 1999 , 1999, CA: a cancer journal for clinicians.

[35]  F Levi,et al.  Trends in cancer mortality in the European Union and accession countries, 1980-2000. , 2004, Annals of oncology : official journal of the European Society for Medical Oncology.

[36]  N. Petrick,et al.  Classification of mass and normal breast tissue on digital mammograms: multiresolution texture analysis. , 1995, Medical physics.

[37]  I. M. Ariel,et al.  Breast cancer: Diagnosis and treatment , 1987 .

[38]  Meng Joo Er,et al.  Classification of mammographic masses using generalized dynamic fuzzy neural networks , 2003, SPIE Medical Imaging.

[39]  A. Lauria,et al.  The CALMA system: an artificial neural network method for detecting masses and microcalcifications in digitized mammograms , 2003, physics/0307099.

[40]  Mario Vento,et al.  To reject or not to reject: that is the question-an answer in case of neural classifiers , 2000, IEEE Trans. Syst. Man Cybern. Part C.

[41]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[42]  Matthew T. Freedman,et al.  Separation of malignant and benign masses using image and segmentation features , 2003, SPIE Medical Imaging.

[43]  J F Hinkamp.  The role of mammography in the diagnosis of breast cancer: survey in a community hospital. , 1971, American journal of obstetrics and gynecology.

[44]  M. Giger,et al.  Analysis of spiculation in the computerized classification of mammographic masses. , 1995, Medical physics.