Abstract. Magnetic resonance-guided high intensity focused ultrasound treatment of the liver is a promising noninvasive technique for ablation of liver lesions. For the technique to be used in clinical practice, however, the issue of liver motion needs to be addressed. A subject-specific four-dimensional liver motion model is presented that is created based on registration of dynamically acquired magnetic resonance data. This model can be used for predicting the tumor motion trajectory for treatment planning and to indicate the tumor position for treatment guidance. The performance of the model was evaluated on a dynamic scan series that was not used to build the model. The method achieved an average Dice coefficient of 0.93 between the predicted and actual liver profiles and an average vessel misalignment of 3.0 mm. The model performed robustly, with a small variation in the results per subject. The results demonstrate the potential of the model to be used for MRI-guided treatment of liver lesions. Furthermore, the model can possibly be applied in other image-guided therapies, for instance radiotherapy of the liver.

Abstract. A liver motion model based on registration of dynamic MRI data, as previously proposed by the authors, was extended with temporal prediction and respiratory signal data. The potential improvements of these extensions with respect to the original model were investigated. Additional evaluations were performed to investigate the limitations of the model regarding temporal prediction and extreme breathing motion.Data were acquired of four volunteers, with breathing instructions and a respiratory belt. The model was built from these data using spatial prediction only and using temporal forward prediction of 300 ms to 1200 ms.From temporal prediction of 0 ms to 1200 ms ahead, the Dice coefficient of liver overlap decreased with 0.85%, the median liver surface distance increased with 20.6% and the vessel misalignment increased with 20%. The mean vessel misalignment was 2.9 mm for the original method, 3.42 mm for spatial prediction with a respiratory signal and 4.01 mm for prediction of 1200 ms ahead with a respiratory signal.Although the extension of the model to temporal prediction yields a decreased prediction accuracy, the results are still acceptable. The use of the breathing signal as input to the model is feasible. Sudden changes in the breathing pattern can yield large errors. However, these errors only persist during a short time interval, after which they can be corrected automatically. Therefore, this model could be useful in a clinical setting.

MR-HIFU is a new non-invasive treatment modality that can be used for palliation in patients with painful bone metastases. Since treatment strategies are mainly focused on the ablation of periosteal nerves, information on the presence and geometry of cortical bone influences the treatment strategy, both in determining the acoustic power and in avoiding safety issues related to far-field heating. Although MRI is available for imaging during treatment, CT is best used for examining the cortical bone. We present a registration method for registering CT and MR images of patients with bone metastases prior to therapy. CT and MRI data were obtained from nine patients with metastatic bone lesions at varying locations. A two-step registration approach was used, performing simultaneous rigid registration of all available MR images in the first step and an affine and deformable registration with an additional bone metric in the second step. The performance was evaluated using landmark annotation by clinical observers. An average registration error of 4.5 mm was obtained, which was comparable to the slice thickness of the data. The performance of the registration algorithm was satisfactory, even with differences in MRI acquisition parameters and for various anatomical sites. The obtained CT overlay is useful for treatment planning, as it allows an assessment of the integrity of the cortical bone. CT-MR registration is therefore recommended for HIFU treatment planning of patients with bone metastases.