1522-2586/asset/cover.gif?v=1&s=7744bbc5c5ee23bd9677f861606033d1d6742e48 "Journal of Magnetic Resonance Imaging")
Presented in part at Radiological Society of North America (RSNA), 2005, November 27–December 2, McCormick Place, Chicago, IL, USA.
1522-2586/asset/JMRI_left.gif?v=1&s=b7fad2e13b2fe41d8e616be0fde3492c7a0033f8)
1522-2586/asset/JMRI_right.gif?v=1&s=62cf6203f6392175649e5bce75bc388c750f03e8)
Original Research
Breast MRI lesion classification: Improved performance of human readers with a backpropagation neural network computer-aided diagnosis (CAD) system†
Article first published online: 11 DEC 2006
DOI: 10.1002/jmri.20794
Copyright © 2006 Wiley-Liss, Inc.
Additional Information
How to Cite
Meinel, L. A., Stolpen, A. H., Berbaum, K. S., Fajardo, L. L. and Reinhardt, J. M. (2007), Breast MRI lesion classification: Improved performance of human readers with a backpropagation neural network computer-aided diagnosis (CAD) system. Journal of Magnetic Resonance Imaging, 25: 89–95. doi: 10.1002/jmri.20794
- †
Publication History
- Issue published online: 21 DEC 2006
- Article first published online: 11 DEC 2006
- Manuscript Accepted: 31 AUG 2006
- Manuscript Received: 3 NOV 2005
Funded by
- Confirma, Inc
- Abstract
- Article
- References
- Cited By
Keywords:
- breast MRI;
- neural networks;
- pattern recognition;
- computer-aided diagnosis;
- shape and texture features;
- kinetics enhancement
Abstract
Purpose
To develop and test a computer-aided diagnosis (CAD) system to improve the performance of radiologists in classifying lesions on breast MRI (BMRI).
Materials and Methods
A CAD system was developed that uses a semiautomated segmentation method. After segmentation, 42 features based on lesion shape, texture, and enhancement kinetics were computed, and the 13 best features were selected and used as inputs to a backpropagation neural network (BNN). The BNN was trained and tested using the leave-one-out method on 80 BMRI lesions (37 benign, 43 malignant). Lesion histopathology was used as the reference standard. Five human readers classified the 80 lesions first without and then with CAD assistance. The performance of the computer classifier and the human readers was assessed using receiver operating characteristic curves; the performance of the human readers was also evaluated using multireader multicase (MRMC) analysis.
Results
The performance of the human readers significantly improved when aided by the CAD system (P < 0.05). MRMC analysis showed that human reader performance with and without CAD system assistance can be generalized to the population of cases (P < 0.001).
Conclusion
A CAD system based on lesion morphology and enhancement kinetics can improve the performance of human readers in classifying lesions on breast MRI. J. Magn. Reson. Imaging 2007. © 2006 Wiley-Liss, Inc.