A new and fast image feature selection method for developing an optimal mammographic mass detection scheme.

PURPOSE Selecting optimal features from a large image feature pool remains a major challenge in developing computer-aided detection (CAD) schemes of medical images. The objective of this study is to investigate a new approach to significantly improve efficacy of image feature selection and classifier optimization in developing a CAD scheme of mammographic masses. METHODS An image dataset including 1600 regions of interest (ROIs) in which 800 are positive (depicting malignant masses) and 800 are negative (depicting CAD-generated false positive regions) was used in this study. After segmentation of each suspicious lesion by a multilayer topographic region growth algorithm, 271 features were computed in different feature categories including shape, texture, contrast, isodensity, spiculation, local topological features, as well as the features related to the presence and location of fat and calcifications. Besides computing features from the original images, the authors also computed new texture features from the dilated lesion segments. In order to select optimal features from this initial feature pool and build a highly performing classifier, the authors examined and compared four feature selection methods to optimize an artificial neural network (ANN) based classifier, namely: (1) Phased Searching with NEAT in a Time-Scaled Framework, (2) A sequential floating forward selection (SFFS) method, (3) A genetic algorithm (GA), and (4) A sequential forward selection (SFS) method. Performances of the four approaches were assessed using a tenfold cross validation method. RESULTS Among these four methods, SFFS has highest efficacy, which takes 3%-5% of computational time as compared to GA approach, and yields the highest performance level with the area under a receiver operating characteristic curve (AUC) = 0.864 ± 0.034. The results also demonstrated that except using GA, including the new texture features computed from the dilated mass segments improved the AUC results of the ANNs optimized using other three feature selection methods. In addition, among 271 features, the shape, local morphological features, fat and calcification based features were the most frequently selected features to build ANNs. CONCLUSIONS Although conventional GA is a powerful tool in optimizing classifiers used in CAD schemes of medical images, it is very computationally intensive. This study demonstrated that using a new SFFS based approach enabled to significantly improve efficacy of image feature selection for developing CAD schemes.

[1]  Ping Lu,et al.  Computerized detection of mammographic lesions: performance of artificial neural network with enhanced feature extraction , 1995, Medical Imaging.

[2]  Nico Karssemeijer,et al.  Detection of stellate distortions in mammograms , 1996, IEEE Trans. Medical Imaging.

[3]  Arnau Oliver,et al.  A review of automatic mass detection and segmentation in mammographic images , 2010, Medical Image Anal..

[4]  Y H Chang,et al.  Applying computer-assisted detection schemes to digitized mammograms after JPEG data compression: an assessment. , 2000, Academic radiology.

[5]  Martin D. Fox,et al.  Classifying mammographic lesions using computerized image analysis , 1993, IEEE Trans. Medical Imaging.

[6]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[7]  D. Wolverton,et al.  Performance parameters for screening and diagnostic mammography: specialist and general radiologists. , 2002, Radiology.

[8]  Yongyi Yang,et al.  Computer-Aided Detection and Diagnosis of Breast Cancer With Mammography: Recent Advances , 2009, IEEE Transactions on Information Technology in Biomedicine.

[9]  Berkman Sahiner,et al.  Classification of mass and normal breast tissue: a convolution neural network classifier with spatial domain and texture images , 1996, IEEE Trans. Medical Imaging.

[10]  C. Balleyguier,et al.  Computed-aided diagnosis (CAD) in the detection of breast cancer. , 2013, European journal of radiology.

[11]  P. Burman.  A comparative study of ordinary cross-validation, v-fold cross-validation and the repeated learning-testing methods , 1989 .

[12]  Bin Zheng,et al.  A new mass classification system derived from multiple features and a trained MLP model , 2014, Medical Imaging.

[13]  Constantine Kotropoulos,et al.  Information Loss of the Mahalanobis Distance in High Dimensions: Application to Feature Selection , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Matthew A. Kupinski,et al.  Feature selection and classifiers for the computerized detection of mass lesions in digital mammography , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[15]  Robert M. Nishikawa,et al.  Current status and future directions of computer-aided diagnosis in mammography , 2007, Comput. Medical Imaging Graph..

[16]  Berkman Sahiner,et al.  Optimal neural network architecture selection: improvement in computerized detection of microcalcifications. , 2002, Academic radiology.

[17]  Risto Miikkulainen,et al.  Real-time neuroevolution in the NERO video game , 2005, IEEE Transactions on Evolutionary Computation.

[18]  Heng-Da Cheng,et al.  Approaches for automated detection and classification of masses in mammograms , 2006, Pattern Recognit..

[19]  L. Tabár,et al.  Beyond randomized controlled trials , 2001, Cancer.

[20]  Josef Kittler,et al.  Floating search methods in feature selection , 1994, Pattern Recognit. Lett..

[21]  Berkman Sahiner,et al.  Comparison of similarity measures for the task of template matching of masses on serial mammograms. , 2005, Medical physics.

[22]  Constantine Kotropoulos,et al.  Fast and accurate sequential floating forward feature selection with the Bayes classifier applied to speech emotion recognition , 2008, Signal Process..

[23]  Alexander Horsch,et al.  Needs assessment for next generation computer-aided mammography reference image databases and evaluation studies , 2011, International Journal of Computer Assisted Radiology and Surgery.

[24]  N Karssemeijer,et al.  Use of border information in the classification of mammographic masses , 2006, Physics in medicine and biology.

[25]  Bin Zheng,et al.  Optimization of breast mass classification using sequential forward floating selection (SFFS) and a support vector machine (SVM) model , 2014, International Journal of Computer Assisted Radiology and Surgery.

[26]  H. D. Cheng,et al.  Mass lesion detection with a fuzzy neural network , 2004, Pattern Recognit..

[27]  Hussein A. Abbass,et al.  An evolutionary artificial neural networks approach for breast cancer diagnosis , 2002, Artif. Intell. Medicine.

[28]  Rangaraj M. Rangayyan,et al.  Measures of acutance and shape for classification of breast tumors , 1997, IEEE Transactions on Medical Imaging.

[29]  B. Zheng,et al.  Assessment of performance improvement in content-based medical image retrieval schemes using fractal dimension. , 2009, Academic radiology.

[30]  Risto Miikkulainen,et al.  Competitive Coevolution through Evolutionary Complexification , 2011, J. Artif. Intell. Res..

[31]  H P Chan,et al.  Image feature selection by a genetic algorithm: application to classification of mass and normal breast tissue. , 1996, Medical physics.

[32]  D B Fogel,et al.  Evolving neural networks for detecting breast cancer. , 1995, Cancer letters.

[33]  Y H Chang,et al.  Feature selection for computerized mass detection in digitized mammograms by using a genetic algorithm. , 1999, Academic radiology.

[34]  Nico Karssemeijer,et al.  Single and multiscale detection of masses in digital mammograms , 1999, IEEE Transactions on Medical Imaging.

[35]  N. Petrick,et al.  Design of a high-sensitivity classifier based on a genetic algorithm: application to computer-aided diagnosis , 1998, Physics in medicine and biology.

[36]  Berkman Sahiner,et al.  An adaptive density-weighted contrast enhancement filter for mammographic breast mass detection , 1996, IEEE Trans. Medical Imaging.

[37]  Mia K Markey,et al.  Correspondence in texture features between two mammographic views. , 2005, Medical physics.

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

[39]  Berkman Sahiner,et al.  Computer-aided detection of breast masses on full field digital mammograms. , 2005, Medical physics.

[40]  Rangaraj M. Rangayyan,et al.  A review of computer-aided diagnosis of breast cancer: Toward the detection of subtle signs , 2007, J. Frankl. Inst..

[41]  Risto Miikkulainen,et al.  Evolving Neural Networks through Augmenting Topologies , 2002, Evolutionary Computation.

[42]  Mary M. Galloway,et al.  Texture analysis using gray level run lengths , 1974 .

[43]  D Brzakovic,et al.  An approach to automated detection of tumors in mammograms. , 1990, IEEE transactions on medical imaging.

[44]  David Gur,et al.  A method to improve visual similarity of breast masses for an interactive computer-aided diagnosis environment. , 2005, Medical physics.

[45]  H P Chan,et al.  Selection of an optimal neural network architecture for computer-aided detection of microcalcifications--comparison of automated optimization techniques. , 2001, Medical physics.

[46]  D. Saslow,et al.  Cancer screening in the United States, 2011 , 2011, CA: a cancer journal for clinicians.

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

[48]  David Gur,et al.  Computer-aided detection schemes: the effect of limiting the number of cued regions in each case. , 2004, AJR. American journal of roentgenology.

[49]  Nicholas J. Radcliffe,et al.  Genetic set recombination and its application to neural network topology optimisation , 1993, Neural Computing & Applications.

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

[51]  H P Chan,et al.  Combined adaptive enhancement and region-growing segmentation of breast masses on digitized mammograms. , 1999, Medical physics.

[52]  Rangaraj M. Rangayyan,et al.  Detection of breast masses in mammograms by density slicing and texture flow-field analysis , 2001, IEEE Transactions on Medical Imaging.

[53]  James F. Greenleaf,et al.  Use of gray value distribution of run lengths for texture analysis , 1990, Pattern Recognit. Lett..

[54]  H P Chan,et al.  Automated detection of breast masses on mammograms using adaptive contrast enhancement and texture classification. , 1996, Medical physics.

[55]  Rangaraj M. Rangayyan,et al.  DETECTION AND CLASSIFICATION OF MAMMOGRAPHIC CALCIFICATIONS , 1993 .

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

[57]  E C Wasson,et al.  A step toward computer-assisted mammography using evolutionary programming and neural networks. , 1997, Cancer letters.

[58]  Y H Chang,et al.  Performance gain in computer-assisted detection schemes by averaging scores generated from artificial neural networks with adaptive filtering. , 2001, Medical physics.

[59]  Bin Zheng,et al.  Computerized prediction of risk for developing breast cancer based on bilateral mammographic breast tissue asymmetry. , 2011, Medical engineering & physics.

[60]  Jan Cornelis,et al.  Phased searching with NEAT in a Time-Scaled Framework: Experiments on a computer-aided detection system for lung nodules , 2013, Artif. Intell. Medicine.

[61]  David Gur,et al.  Computer-aided detection; the effect of training databases on detection of subtle breast masses. , 2010, Academic radiology.

[62]  Y H Chang,et al.  Computerized detection of masses in digitized mammograms using single-image segmentation and a multilayer topographic feature analysis. , 1995, Academic radiology.

[63]  David Gur,et al.  Performance change of mammographic CAD schemes optimized with most-recent and prior image databases. , 2003, Academic radiology.

[64]  Evangelos Dermatas,et al.  Fast detection of masses in computer-aided mammography , 2000, IEEE Signal Process. Mag..

[65]  Warren B. Powell,et al.  Approximate Dynamic Programming - Solving the Curses of Dimensionality , 2007 .

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

[67]  Luisa P. Wallace,et al.  Multiview-based computer-aided detection scheme for breast masses. , 2006, Medical physics.

[68]  David Gur,et al.  Application of a Bayesian belief network in a computer-assisted diagnosis scheme for mass detection , 1999, Medical Imaging.

[69]  Rudy Setiono,et al.  Extracting rules from pruned networks for breast cancer diagnosis , 1996, Artif. Intell. Medicine.

[70]  Peter J. Angeline,et al.  An evolutionary algorithm that constructs recurrent neural networks , 1994, IEEE Trans. Neural Networks.

[71]  N. Karssemeijer,et al.  An automatic method to discriminate malignant masses from normal tissue in digital mammograms1 , 2000, Physics in medicine and biology.

[72]  David Gur,et al.  Computer-aided detection performance in mammographic examination of masses: assessment. , 2004, Radiology.

[73]  Xiaoou Tang,et al.  Texture information in run-length matrices , 1998, IEEE Trans. Image Process..