Reconstruction of three-dimensional breast-tumor model using multispectral gradient vector flow snake method

AbstractIn this study, we have proposed a three-dimensional (3D) model reconstruction system for breast tumors. The proposed system can establish an accurate 3D model of tumors, which will serve as a diagnostic reference for physicians and also address the shortcomings of the traditional breast needle localization method and other localization methods reported in previous studies. This developed system uses multispectral breast magnetic resonance images as input and detects the contour of the tumor in different sections using an active contour method — multispectral gradient vector flow snake (MGVFS) method. Thus, the system constructs a 3D model of only the tumor is contained in a breast surface model and excludes other tissues. Since the accuracy of the reconstructed 3D model depends on the accuracy of the tumor contour detection, for confirming the results obtained with the MGVFS method, we conducted experiments to evaluate its accuracy in contour detection, and compared the results with those traditional contour detection methods. Our results demonstrate that the MGVFS method has the highest accuracy in contour detection, with a correct contour detection rate as high as 99.79%