Interpolation has become a default operation in image processing and medical imaging and is one of the important factors in the\r\nsuccess of an intensity-based registration method. Interpolation is needed if the fractional unit ofmotion is not matched and located\r\non the high resolution (HR) grid. The purpose of this work is to present a systematic evaluation of eight standard interpolation\r\ntechniques (trilinear, nearest neighbor, cubic Lagrangian, quintic Lagrangian, hepatic Lagrangian, windowed Sinc, B-spline 3rd\r\norder, and B-spline 4th order) and to compare the effect of cost functions (least squares (LS), normalized mutual information\r\n(NMI), normalized cross correlation (NCC), and correlation ratio (CR)) for optimized automatic image registration (OAIR) on 3D\r\nspoiled gradient recalled (SPGR) magnetic resonance images (MRI) of the brain acquired using a 3T GE MR scanner. Subsampling\r\nwas performed in the axial, sagittal, and coronal directions to emulate three low resolution datasets. Afterwards, the low resolution\r\ndatasets were upsampled using different interpolation methods, and they were then compared to the high resolution data.Themean\r\nsquared error, peak signal to noise, joint entropy, and cost functions were computed for quantitative assessment of the method.\r\nMagnetic resonance image scans and joint histogram were used for qualitative assessment of the method.
Loading....