Separation of malignant and benign masses using maximum-likelihood modeling and neural networks

This study attempted to accurately segment the masses and distinguish malignant from benign tumors. The masses were segmented using a technique that combines pixel aggregation with maximum likelihood analysis. We found that the segmentation method can delineate the tumor body as well as tumor peripheral regions covering typical mass boundaries and some spiculation patterns. We have developed a Multiple Circular Path Convolution Neural Network (MCPCNN) to analyze a set of mass intensity, shape, and texture features for determination of the tumors as malignant or benign. The features were also fed into a conventional neural network for comparison. We also used values obtained from the maximum likelihood values as inputs into a conventional backpropagation neural network. We have tested these methods on 51 mammograms using a grouped Jackknife experiment incorporated with the ROC method. Tumor sizes ranged from 6mm to 3cm. The conventional neural network whose inputs were image features achieved an Az of 0.66. However the MCPCNN achieved an Az value of 0.71. The conventional neural network whose inputs were maximum likelihood values achieved an Az value of 0.84. In addition, the maximum likelihood segmentation method can identify the mass body and boundary regions, which is essential to the analysis of mammographic masses.

[1]  P. G. Tahoces,et al.  Computer-aided diagnosis: automatic detection of malignant masses in digitized mammograms. , 1998, Medical physics.

[2]  R A Clark,et al.  False-positive reduction in CAD mass detection using a competitive classification strategy. , 2001, Medical physics.

[3]  Richard H. Moore,et al.  Current Status of the Digital Database for Screening Mammography , 1998, Digital Mammography / IWDM.

[4]  S Shapiro,et al.  Screening: Assessment of current studies , 1994, Cancer.

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

[6]  K. J. Ray Liu,et al.  Computerized radiographic mass detection. I. Lesion site selection by morphological enhancement and contextual segmentation , 2001, IEEE Transactions on Medical Imaging.

[7]  L. Tabár,et al.  Update of the Swedish two-county program of mammographic screening for breast cancer. , 1992, Radiologic clinics of North America.

[8]  E A Sickles,et al.  Initial versus subsequent screening mammography: comparison of findings and their prognostic significance. , 1995, AJR. American journal of roentgenology.

[9]  D. Kopans.  The positive predictive value of mammography. , 1992, AJR. American journal of roentgenology.

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

[11]  Huai Li,et al.  A multiple circular path convolution neural network system for detection of mammographic masses , 2002, IEEE Transactions on Medical Imaging.

[12]  N. Obuchowski,et al.  Quantitative classification of breast tumors in digitized mammograms. , 1996, Medical physics.

[13]  S. Wall,et al.  Breast cancer screening with mammography: overview of Swedish randomised trials , 1993, The Lancet.

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

[15]  Rafael C. González,et al.  Local Determination of a Moving Contrast Edge , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.