| Meersman D, Scheunders P, Van Dyck D. Detection of microcalcifications using neural networks. In: Proceedings of the 3rd International Workshop on Digital Mammography, 1996. p. 287--90. |
....various methodologies. The performance of the proposed Fig. 8. ROC curves obtained with the neural network and hybrid intelligent system using the MIAS database with complete dataset and cross validation. method is comparable with the reported results. For the Nijmegen dataset, Meersman et al. [28] using a neural network approach reported a sensitivity level 0.84 and two false positive clusters per image. Using an adaptive filtering method Gurgan et al. 18] achieved a sensitivity 1.0 with 2.3 false clusters per image. Yu [58] obtained a sensitivity 0.9 with 0.5 false clusters per image and ....
Meersman D, Scheunders P, Van Dyck D. Detection of microcalcifications using neural networks. In: Proceedings of the 3rd International Workshop on Digital Mammography, 1996. p. 287--90.
....of an experienced radiologist. 3.2.5 Neural networks : In an effort to detect individual microcalcifications, neural networks are most commonly applied to regions containing clusters of microcalcifications. A more effective technique has been developed by Meersman, Scheunders and Van Dyke [13] at the University of Antwerp in Belgium which uses neural networks to perform a pixel based classification. The networks used were three layered feedforward neural networks with varying dimensions. The neighbourhood of each pixel (9x9 or 15x15 pixels) within the image was given as input to the ....
Meersman D, Scheunders P et al "Detection of microcalcifications using neural networks", Third International Workshop on Digital Mammography, Chicago June 1996.
No context found.
D. Meersman, P. Scheunders and D. Van Dyck, "Detection of Microcalcifications using Neural Networks", Proc. of the 3 rd Int. Workshop on Digital Mammograph, Chicago, IL, 1996, pp.97-103.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC