Mammogram Segmentation by Contour Searching and Mass Lesions Classification with Neural Network

Abstract

The mammography is the most effective procedure for an early diagnosis of the breast cancer. In this paper, an algorithm for detecting masses in mammographic images will be presented. The database consists of 3762 digital images acquired in several hospitals belonging to the MAGIC-5 collaboration (Medical Applications on a Grid Infrastructure Connection). A reduction of the whole image’s area under investigation is achieved through a segmentation process, by means of a ROI Hunter algorithm, without loss of meaningful information. In the following classification step, feature extraction plays a fundamental role: some features give geometrical information, other ones provide shape parameters. Once the features are computed for each ROI, they are used as inputs to a supervised neural network with momentum. The output neuron provides the probability that the ROI is pathological or not. Results are provided in terms of ROC and FROC curves: the area under the ROC curve was found to be AZ = 0.862 ± 0.007, and we get a 2.8 FP/Image at a sensitivity of 82%. This software is included in the CAD station actually working in the hospitals belonging to the MAGIC-5 Collaboration. Asterisk indicates corresponding author. D. Cascio, *F.Fauci, R.Magro and G. Raso are with Dip. di Fisica e Tecnologie Relative, Università di Palermo, Palermo, Italy and INFN, Sezione di Catania, Catania, Italy. S. Bagnasco is with INFN, Sezione di Torino, Torino, Italy. S. C. Cheran is with INFN, Sezione di Torino, Torino, Italy and Dip. di Informatica, Università di Torino, Torino, Italy. R. Bellotti, F. De Carlo and S. Tangaro are with Dip. di Fisica, Università di Bari, Bari, Italy and INFN, Sezione di Bari, Bari, Italy. G. De Nunzio is with Dip. di Scienza dei Materiali, Università di Lecce, Lecce, Italy and INFN, Sezione di Lecce, Lecce, Italy. M. Quarta is with Dip. di Matematica, Università di Lecce, Lecce, Italy and INFN, Sezione di Lecce, Lecce, Italy. M. E. Fantacci and A. Retico are with Dip. di Fisica, Università di Pisa, Pisa, Italy and INFN, Sezione di Pisa, Pisa, Italy. G. Forni is with Dip. di Scienze Fisiche, Università di Napoli, Napoli, Italy and INFN, Sezione di Napoli, Napoli, Italy. A. Lauria is with Dip. di Matematica e Fisica, Università di Sassari, Sassari, Italy and INFN, Sezione di Napoli, Napoli, Italy. G. L. Masala and P. Oliva are with Dip. di Matematica e Fisica, Università di Sassari, Sassari, Italy and INFN, Sezione di Cagliari, Cagliari, Italy. E. Lopez Torres is with CEADEN, Havana, Cuba.

11 Figures and Tables

0200400'06'07'08'09'10'11'12'13'14'15'16'17
Citations per Year

1,605 Citations

Semantic Scholar estimates that this publication has 1,605 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Cascio2006MammogramSB, title={Mammogram Segmentation by Contour Searching and Mass Lesions Classification with Neural Network}, author={Donato Cascio and Francesco Fauci and Rosario Magro and Giuseppe Raso and Robero Bellotti and Francesco De Carlo and Sonia Tangaro and Giovanni De Nunzio and Maurizio Quarta and Giulia Forni and Adele Lauria and Maria Evelina Fantacci and Alessandra Retico and Giovanni Luca Masala and Piernicola Oliva and Stefano Bagnasco and Sorin Cristian Cheran and Ernesto Lopez Torres}, year={2006} }