Local Time : 11:22 CET

C-3151 - Machine learning for challenging tumour detection and classification in breast cancer

I. Alvarez Illan1, A. Meyer-Baese1, J. Perez Matos1, M. B. I. Lobbes2, K. Pinker3; 1 Tallahassee/US 2 Maastricht/NL 3 Vienna/AT Type: Scientific Exhibit
Area of Interests: Breast, Computer applications
Imaging Techniques: Neural networks, MR
Procedures: CAD, Computer Applications-Detection, diagnosis
Add poster to my schedule In your schedule (remove)

Direct link to access the poster (click here)

Aims and objectives: Accurate methods for breast cancer diagnosis are of utmost importance for early detection, adequate treatment and positive patient outcomes. Since the introduction of dynamic-contrast-enhanced magnetic resonance imaging (DCE-MRI), a set of computer aided diagnosis  (CAD) systems have been devel[...]

Methods and materials: The DCE-MRI protocol contains a set of temporal images for each subject. The dataset is preprocessed by an affine co-registration of the temporal images using SPM. The non breast tissues such as the as the thoracic cavity and chest wall muscle are removed . A signal to noise ratio is improved by a 2[...]

Results: An example of on how the projection works is displayed in Figures 1-4 . Figure 1 shows two relevant mixing matrix coefficients obtained from the set of training images, which take the form of enhancement curves: curve 1 has a ’typical’ malign behaviour, while curve 2 is a normal enhancement. The rem[...]

Conclusion: Contrary to other ICA-based analysis, this work proposes to obtain the source enhancement curves in a inter-patient level usinginformation of multiple patient. The extracted features involve no a priori knowledge of the disease, and thus selection bias in contrast to commonly used features such[...]

Personal information: Ignacio Alvarez Illan, Marie-Curie Fellowship Researcher University of Granada (Spain), Department of Signal Theory and Communications Florida State University, Department of Scientific Computing Anke Meyer-Baese, PhD, Professor at Scientific Computing Florida State University, Department of [...]

References: [1] Agliozzo, S., De Luca, M., Bracco, C., Vignati, A., Giannini, V., Martincich, L., Carbonaro, L. A., Bert, A., Sardanelli, F., and Regge, D., “Computer-aided diagnosis for dynamic contrast-enhanced breast MRI of mass-like lesions using a multiparametric model combining a selection of morphologica[...]

This website uses cookies. Learn more