Computed tomography (CT) is the standard imaging modality for patient dose calculation for radiation therapy. framework. The synthetic CT is registered to the original CT using a deformable registration and the computed deformation is applied to the MRI. In contrast to most existing methods we do not need any manual intervention such as picking landmarks or regions of interests. The proposed method was validated on ten brain cancer patient cases showing 25% improvement in MI and correlation between MR and CT images after registration compared to state-of-the-art registration methods. × × × 1 with = and v= 1 … = 1 … and are the number of nonzero voxels in the subject and the atlas respectively. We combine the patch pairs as 2× 1 vectors and and two associated atlas patches qand qis assumed to arise from the Gaussian distribution is a covariance matrix associated with the atlas patches and ∈ Ψ where Ψ is the set of all pairs of atlas patch indices and ∈ (0 1 is a mixing coe cient for the subject patch to the atlas patch-pairs. In essence each subject patch follows an (using expectation-maximization (EM) to find the synthetic contrast patches uas the indicator function that pcomes from a GMM of the = {atlas pair Σ= 1 ?∈ {0 1 Then IB-MECA the probability of observing pcan be written as = p? ? (1 ? ≡ Σis often less IB-MECA robust to estimate. Instead we assume it to be separable and block diagonal = 1 … ∈ Ψ and the maximum likelihood estimators of Θ are found by maximizing Eqn. 3 using EM. The EM algorithm be outlined as E-step: to find new update Θ(iteration compute the expectation (is an indicator function it can be shown that being the posterior probability of poriginating from the Gaussian distribution of the atlas patches qand qand are the expressions defined in Eqn. 3 but with and denote the corresponding values with atlas patches belonging to the ?pair ? ∈ Ψ with with its expectation. The maximization is involved by the M-step of the log of the expectation w.r.t. the parameters given the current (0) = (1) = 1 ? is considered the synthetic is used as the voxel. The imaging model is valid for those atlas and subject patches that are close in intensity. Using a nonlocal type of criterion 18 for every subject patch xatlas patches such that they are the nearest neighbors of xsubject patch follows an = 40. 3 RESULTS We experimented on images from ten brain cancer patients with various sizes and shapes of tumors each having one MR SDC1 and CT acquisition. A di erent subject was chosen as the atlas for which the MRI was carefully registered to the CT using a commercial software.7 This registered MR-CT pair was used as the atlas a1 a2. For each of the ten subjects we registered the MRI to CT using b-spline SyN and registration7.8 We also generated the sCT image from the MRI (b1) registered (SyN) sCT to the original CT and applied the deformation to the MRI to get registered MRI. An example of the atlas a1 a2 subject MR b1 registration results from b-spline SyN sCT and the corresponding deformed MR images from their registrations are shown in Fig. 2. Figure 2 Top row shows a registered pair of MR-CT images used as atlas. Middle row shows the original subject CT image and the registered MRIs by b-spline7 and SyN.8 Bottom row shows the sCT SyN registered sCT and IB-MECA the corresponding deformed MR with the deformation … Fig. 3 top image shows absolute values of correlation and MI between CT and the registered MR brain volumes of ten subjects. The brain volumes are obtained from skull-stripping19 masks of the MR images. Both MI and correlations increase (p-value < 0 significantly.05) after registration via sCT indicating significant improvement in MR-CT registration of the brains. Another registration metric is the variability of CT IB-MECA intensities for di erent tissue classes. For each subject we segmented the registered MRI into three classes cerebro-spinal fluid (CSF) gray matter (GM) and white matter (WM) using an atlas based method.20 The mean and standard deviations of CT intensities for each of the classes are plotted in Fig. 3 bottom row. For every tissue standard deviations from sCT deformed MRI reduce (p-value < 0 significantly.05).