Classification of mass and normal breast tissue: a convolution neural network classifier with spatial domain and texture images

• Published in: IEEE transactions on medical imaging Vol. 15; no. 5; pp. 598 - 610
• Main Authors: Sahiner, B., Heang-Ping Chan, Petrick, N., Datong Wei, Helvie, M.A., Adler, D.D., Goodsitt, M.M.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.10.1996; Institute of Electrical and Electronics Engineers
• Subjects: Backpropagation; Biological and medical sciences; Breast tissue; Cellular neural networks; Computer architecture; Convolution; Feature extraction; Genital system. Mammary gland; Investigative techniques, diagnostic techniques (general aspects); Kernel; Medical sciences; Neural networks; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Testing; Two dimensional displays; Human; Mammary gland diseases; Tissue; Texture analysis; Convolution; Spatial distribution; Classification; Female; Medical imagery; Mammary gland; Neural network; Malignant tumor
• ISSN: 0278-0062; 1558-254X
• DOI: 10.1109/42.538937

The authors investigated the classification of regions of interest (ROI's) on mammograms as either mass or normal tissue using a convolution neural network (CNN). A CNN is a backpropagation neural network with two-dimensional (2-D) weight kernels that operate on images. A generalized, fast and stable implementation of the CNN was developed. The input images to the CNN were obtained from the ROI's using two techniques. The first technique employed averaging and subsampling. The second technique employed texture feature extraction methods applied to small subregions inside the ROI. Features computed over different subregions were arranged as texture images, which were subsequently used as CNN inputs. The effects of CNN architecture and texture feature parameters on classification accuracy were studied. Receiver operating characteristic (ROC) methodology was used to evaluate the classification accuracy. A data set consisting of 168 ROIs containing biopsy-proven masses and 504 ROI's containing normal breast tissue was extracted from 168 mammograms by radiologists experienced in mammography. This data set was used for training and testing the CNN. With the best combination of CNN architecture and texture feature parameters, the area under the test ROC curve reached 0.87, which corresponded to a true-positive fraction of 90% at a false positive fraction of 31%. The authors' results demonstrate the feasibility of using a CNN for classification of masses and normal tissue on mammograms.