D R Chen1, R F Chang, Y L Huang. 1. Department of General Surgery, China Medical College and Hospital, Taichung, Taiwan, Republic of China. dlchen88@ms13.hinet.net
Abstract
PURPOSE: To increase the capabilities of ultrasonographic (US) technology for the differential diagnosis of solid breast tumors by using a neural network. MATERIALS AND METHODS: One hundred forty US images of solid breast nodules were evaluated. When a sonogram was obtained, an analog video signal from the VCR output of the scanner was transmitted to a notebook computer. A frame grabber connected to the printer port of the computer was then used to digitize the data. The suspicious tumor region on the digitized US image was manually selected. The texture information of the subimage was extracted, and a neural network classifier with autocorrelation features was used to classify the tumor as benign or malignant. In this experiment, 140 pathologically proved tumors (52 malignant and 88 benign tumors) were sampled with k-fold cross-validation (k = 10) to evaluate the performance with receiver operating characteristic curves. RESULTS: The accuracy of neural networks for classifying malignancies was 95.0% (133 of 140 tumors), the sensitivity was 98% (51 of 52), the specificity was 93% (82 of 88), the positive predictive value was 89% (51 of 57), and the negative predictive value was 99% (82 of 83). CONCLUSION: This system differentiated solid breast nodules with relatively high accuracy and helped inexperienced operators to avoid misdiagnoses. Because the neural network is trainable, it could be optimized if a larger set of tumor images is supplied.
PURPOSE: To increase the capabilities of ultrasonographic (US) technology for the differential diagnosis of solid breast tumors by using a neural network. MATERIALS AND METHODS: One hundred forty US images of solid breast nodules were evaluated. When a sonogram was obtained, an analog video signal from the VCR output of the scanner was transmitted to a notebook computer. A frame grabber connected to the printer port of the computer was then used to digitize the data. The suspicious tumor region on the digitized US image was manually selected. The texture information of the subimage was extracted, and a neural network classifier with autocorrelation features was used to classify the tumor as benign or malignant. In this experiment, 140 pathologically proved tumors (52 malignant and 88 benign tumors) were sampled with k-fold cross-validation (k = 10) to evaluate the performance with receiver operating characteristic curves. RESULTS: The accuracy of neural networks for classifying malignancies was 95.0% (133 of 140 tumors), the sensitivity was 98% (51 of 52), the specificity was 93% (82 of 88), the positive predictive value was 89% (51 of 57), and the negative predictive value was 99% (82 of 83). CONCLUSION: This system differentiated solid breast nodules with relatively high accuracy and helped inexperienced operators to avoid misdiagnoses. Because the neural network is trainable, it could be optimized if a larger set of tumor images is supplied.
Authors: Berkman Sahiner; Heang-Ping Chan; Marilyn A Roubidoux; Lubomir M Hadjiiski; Mark A Helvie; Chintana Paramagul; Janet Bailey; Alexis V Nees; Caroline Blane Journal: Radiology Date: 2007-01-23 Impact factor: 11.105
Authors: Tim König; Johannes Steffen; Marko Rak; Grit Neumann; Ludwig von Rohden; Klaus D Tönnies Journal: Int J Comput Assist Radiol Surg Date: 2014-12-02 Impact factor: 2.924