Image Registration Results Research Articles

A multi-institutional comparison of retrospective deformable dose accumulation for online adaptive magnetic resonance-guided radiotherapy

Background and PurposeApplication of different deformable dose accumulation (DDA) solutions makes institutional comparisons after online-adaptive magnetic resonance-guided radiotherapy (OA-MRgRT) challenging. The aim of this multi-institutional study was to analyze accuracy and agreement of DDA-implementations in OA-MRgRT. Material and MethodsOne gold standard (GS) case deformed with a biomechanical-model and five clinical cases consisting of prostate (2x), cervix, liver, and lymph node cancer, treated with OA-MRgRT, were analyzed. Six centers conducted DDA using institutional implementations. Deformable image registration (DIR) and DDA results were compared using the contour metrics Dice Similarity Coefficient (DSC), surface-DSC, Hausdorff-distance (HD95%), and accumulated dose-volume histograms (DVHs) analyzed via intraclass correlation coefficient (ICC) and clinical dosimetric criteria (CDC). ResultsFor the GS, median DDA errors ranged from 0.0 to 2.8 Gy across contours and implementations. DIR of clinical cases resulted in DSC > 0.8 for up to 81.3% of contours and a variability of surface-DSC values depending on the implementation. Maximum HD95%=73.3 mm was found for duodenum in the liver case. Although DVH ICC > 0.90 was found after DDA for all but two contours, relevant absolute CDC differences were observed in clinical cases: Prostate I/II showed maximum differences in bladder V28Gy (10.2/7.6%), while for cervix, liver, and lymph node the highest differences were found for rectum D2cm3 (2.8 Gy), duodenum Dmax (7.1 Gy), and rectum D0.5cm3 (4.6 Gy). ConclusionOverall, high agreement was found between the different DIR and DDA implementations. Case- and algorithm-dependent differences were observed, leading to potentially clinically relevant results. Larger studies are needed to define future DDA-guidelines.

Voxel-wise body composition analysis using image registration of a three-slice CT imaging protocol: methodology and proof-of-concept studies

BackgroundComputed tomography (CT) is an imaging modality commonly used for studies of internal body structures and very useful for detailed studies of body composition. The aim of this study was to develop and evaluate a fully automatic image registration framework for inter-subject CT slice registration. The aim was also to use the results, in a set of proof-of-concept studies, for voxel-wise statistical body composition analysis (Imiomics) of correlations between imaging and non-imaging data.MethodsThe current study utilized three single-slice CT images of the liver, abdomen, and thigh from two large cohort studies, SCAPIS and IGT. The image registration method developed and evaluated used both CT images together with image-derived tissue and organ segmentation masks. To evaluate the performance of the registration method, a set of baseline 3-single-slice CT images (from 2780 subjects including 8285 slices) from the SCAPIS and IGT cohorts were registered. Vector magnitude and intensity magnitude error indicating inverse consistency were used for evaluation. Image registration results were further used for voxel-wise analysis of associations between the CT images (as represented by tissue volume from Hounsfield unit and Jacobian determinant) and various explicit measurements of various tissues, fat depots, and organs collected in both cohort studies.ResultsOur findings demonstrated that the key organs and anatomical structures were registered appropriately. The evaluation parameters of inverse consistency, such as vector magnitude and intensity magnitude error, were on average less than 3 mm and 50 Hounsfield units. The registration followed by Imiomics analysis enabled the examination of associations between various explicit measurements (liver, spleen, abdominal muscle, visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), thigh SAT, intermuscular adipose tissue (IMAT), and thigh muscle) and the voxel-wise image information.ConclusionThe developed and evaluated framework allows accurate image registrations of the collected three single-slice CT images and enables detailed voxel-wise studies of associations between body composition and associated diseases and risk factors.

Anatomically constrained and attention-guided deep feature fusion for joint segmentation and deformable medical image registration.

The main objective of anatomically plausible results for deformable image registration is to improve model's registration accuracy by minimizing the difference between a pair of fixed and moving images. Since many anatomical features are closely related to each other, leveraging supervision from auxiliary tasks (such as supervised anatomical segmentation) has the potential to enhance the realism of the warped images after registration. In this work, we employ a Multi-Task Learning framework to formulate registration and segmentation as a joint issue, in which we utilize anatomical constraint from auxiliary supervised segmentation to enhance the realism of the predicted images. First, we propose a Cross-Task Attention Block to fuse the high-level feature from both the registration and segmentation network. With the help of initial anatomical segmentation, the registration network can benefit from learning the task-shared feature correlation and rapidly focusing on the parts that need deformation. On the other hand, the anatomical segmentation discrepancy from ground-truth fixed annotations and predicted segmentation maps of initial warped images are integrated into the loss function to guide the convergence of the registration network. Ideally, a good deformation field should be able to minimize the loss function of registration and segmentation. The voxel-wise anatomical constraint inferred from segmentation helps the registration network to reach a global optimum for both deformable and segmentation learning. Both networks can be employed independently during the testing phase, enabling only the registration output to be predicted when the segmentation labels are unavailable. Qualitative and quantitative results indicate that our proposed methodology significantly outperforms the previous state-of-the-art approaches on inter-patient brain MRI registration and pre- and intra-operative uterus MRI registration tasks within our specific experimental setup, which leads to state-of-the-art registration quality scores of 0.755 and 0.731 (i.e., by 0.8% and 0.5% increases) DSC for both tasks, respectively.

Deformable Image Registration Using Vision Transformers for Cardiac Motion Estimation from Cine Cardiac MRI Images.

Accurate cardiac motion estimation is a crucial step in assessing the kinematic and contractile properties of the cardiac chambers, thereby directly quantifying the regional cardiac function, which plays an important role in understanding myocardial diseases and planning their treatment. Since the cine cardiac magnetic resonance imaging (MRI) provides dynamic, high-resolution 3D images of the heart that depict cardiac motion throughout the cardiac cycle, cardiac motion can be estimated by finding the optical flow representation between the consecutive 3D volumes from a 4D cine cardiac MRI dataset, thereby formulating it as an image registration problem. Therefore, we propose a hybrid convolutional neural network (CNN) and Vision Transformer (ViT) architecture for deformable image registration of 3D cine cardiac MRI images for consistent cardiac motion estimation. We compare the image registration results of our proposed method with those of the VoxelMorph CNN model and conventional B-spline free form deformation (FFD) non-rigid image registration algorithm. We conduct all our experiments on the open-source Automated Cardiac Diagnosis Challenge (ACDC) dataset. Our experiments show that the deformable image registration results obtained using the proposed method outperform the CNN model and the traditional FFD image registration method.

Evaluation of optimal kilovoltage-cone beam technique on image quality, registration accuracy, time of imaging and relative dose for head radiotherapy: A phantom study

Background: Image-guided radiation therapy (IGRT) has improved geometry accuracy of patient positioning in radiotherapy. Nowadays, a kilo-voltage cone-beam computed tomography (kV-CBCT) is widely applied in IGRT to confirm daily patient positioning. However, one problem with IGRT is the increased dose to normal tissue outside the target area. Reduction in the dose of kV-CBCT verification imaging leads to deterioration in image quality and registration results. Objectives: The objective of this study was to evaluate an optimal kV cone-beam technique on image quality, registration accuracy, relative dose, and imaging time using different combinations of angular range, angular separation, and mA (in total eight protocols) in the head phantom. Materials and methods: Catphan® 503 phantom was used to evaluate image quality and a PIXY Anthropomorphic Training/Teaching Phantom was used to verify image registration accuracy. The kV-CBCT imaging was performed using an X-ray volumetric imaging (XVI) system mounted on an Elekta Versa HD linear accelerator. The absorbed dose of kV-CBCT imaging was determined using the head phantom with an ionization chamber. The eight protocols were analyzed for image quality, image registration, relative dose change, and imaging delivery time. Results: Image quality parameter results showed that maximum contrast to noise ratio (CNR) increased by approximately 65% at 100 kV, 20 mA, θ=360°, and Δθ=0.54°, while uniformity compared with 100 kV, 10 mA, θ=200°, Δθ=0.54° (default protocol) varied within 7%. The spatial resolution showed no change (0.167 cm), with geometric distortions of less than 0.2 mm. Image registration errors were within 0.2 cm, with the highest magnitude of error seen in the vertical direction. Imaging dose can be reduced using 100 kV, 10 mA, θ=200°, Δθ=1.09° about 50% and 100 kV, 10 mA, θ=360°, Δθ=1.09° about 15% with similar CNR of default protocol. Imaging time decreased by approximately 2-folds, while Δθ increased by 2-folds. Conclusion: The suggested protocols suitable for optimal kV-CBCT image quality, accuracy of image registration, decreased imaging time, and reduced image dose in the head region were 1) 10 mA, θ=200°, Δθ=1.09° 2) 10 mA, θ=360°, Δθ=1.09°, and 3) 20 mA, θ=360°, Δθ=1.09°. These protocols decreased imaging dose about 50%, 15%, and 2%, respectively. The developing kV-CBCT technique should be counterbalanced by careful consideration of imaging dose, image quality, imaging time, and accuracy of image registration.

Self-Supervised Rigid Registration for Multimodal Retinal Images.

The ability to accurately overlay one modality retinal image to another is critical in ophthalmology. Our previous framework achieved the state-of-the-art results for multimodal retinal image registration. However, it requires human-annotated labels due to the supervised approach of the previous work. In this paper, we propose a self-supervised multimodal retina registration method to alleviate the burdens of time and expense to prepare for training data, that is, aiming to automatically register multimodal retinal images without any human annotations. Specially, we focus on registering color fundus images with infrared reflectance and fluorescein angiography images, and compare registration results with several conventional and supervised and unsupervised deep learning methods. From the experimental results, the proposed self-supervised framework achieves a comparable accuracy comparing to the state-of-the-art supervised learning method in terms of registration accuracy and Dice coefficient.

An Effective Framework of Automated Visual Surface Defect Detection for Metal Parts

Developing effective automated visual surface defect detection of metal parts is a challenge due to highly reflective surfaces and miscellaneous defect patterns. In this work, an effective framework consisting of an image registration module and a defect detection module is proposed to cope with this challenge. First, an algorithm based on dual weighted principal component analysis is designed for the image registration module. In this algorithm, the visual feature and spatial feature are integrated into a unified principal component to obtain highly accurate image registration results. Then, a defect detection module is constructed using an image difference algorithm with prior constraints. In this module, two categories of constraint bounding boxes are designed to reduce the misdetection due to the susceptible edges and improve the accuracy of surface defect detection of metal parts. Three applications, including a magnetic resonance imaging dataset, a printed circuit board dataset, and an industrial case of the stainless-steel positioning pins are provided to illustrate the superiority of the proposed framework.

Effects of Exposure Parameters and Voxel Size for Cone-Beam Computed Tomography on the Image Matching Accuracy with an Optical Dental Scan Image: An In Vitro Study

This study is aimed at assessing the effects of exposure parameters and voxel size for cone-beam computed tomography (CBCT) on the image matching accuracy with an optical dental scan image. CBCT and optical scan images of a dry human mandible were obtained. Different CBCT settings were used: tube voltage, 60, 80, and 100 kVp; tube current, 6 and 8 mA; and voxel size, 100, 200, and 300 μm. Image matching between the CBCT and optical scan images was performed using implant planning software by dental professionals (n = 18). The image matching accuracy in each combination of CBCT settings was evaluated by assessing the linear discrepancy between the three-dimensionally reconstructed radiological image and the registered optical scan image using an image analysis software program. The Kruskal-Wallis test and a post hoc Mann–Whitney U test with Bonferroni correction were used to compare the accuracy of image registration between the groups (α = 0.05). Overall, the image matching accuracy was not significantly different between tube voltage and current settings; however, significantly higher image registration errors were found at the combination of 100 kVp tube voltage/8 mA tube current (F = 8.44, P < 0.001). Changes in voxel sizes did not significantly interfere with the image registration results. No interaction was found among voltage, current, and voxel size in terms of image registration accuracy (F = 2.022, P = 0.091). Different exposure parameter settings in tube voltage and tube current did not significantly influence the image matching accuracy between CBCT and optical dental scan images; however, a high radiation dose could be inappropriate. The image matching accuracy was not significantly affected by changing the voxel sizes of CBCT.

A Reconfigurable Memory based Fast VLSI Architecture for Histogram Computation

Histogram computation is the crucial task used in processing so many image guided applications like pattern recognition, image segmentation etc. Image registration is one of the fundamental techniques for pre-processing of the images. Registration is the process of overlaying multiple images to geometrically align them. In medical Image processing, the improper registration can have negative impact on the analysis of the image which influences the final diagnosis. The accurate result of image registration is obtained by matching of multimodal images. Mutual Information is one of the commonly used techniques to find the similarity measurement between multi-modal images. Measurement of similarity requires a computation of histogram of individual images and joint histogram between the images. The hardware implementation of histogram computation has advantages in terms of flexible design, low power consumption, high speed, less execution time than the software implementation. This paper proposed a parallel algorithm for histogram computation. A memory based pipeline architecture is designed for implementing the proposed algorithm. The hardware mapping of the algorithm on FPGA is proposed and simulating them using Xilinx software tools.

A fast multi grid algorithm for 2D diffeomorphic image registration model

In Han and Wang (2020), a 2D diffeomorphic image registration model is proposed to eliminate mesh folding. To solve the 2D diffeomorphic model, a diffeomorphic fractional-order image registration algorithm (DFIRA for short) is proposed in Han and Wang (2020). DFIRA achieves a satisfactory image registration result but it costs too much CPU time. To accelerate DFIRA, we propose a fast multi grid algorithm for 2D diffeomorphic image registration model in this paper. This algorithm achieves a satisfactory image registration result by using only one-tenth CPU time of DFIRA. At the same time, no mesh folding occurs in proposed algorithm. Furthermore, convergence analysis of the proposed algorithm is also presented. Moreover, numerical tests are also performed to show that the proposed algorithm is competitive compared with some other algorithms.

Electric Shovel Teeth Missing Detection Method Based on Deep Learning.

Electric shovels are widely used in the mining industry to dig ore, and the teeth in shovels' bucket can be lost due to the tremendous pressure exerted by ore materials during operation. When the teeth fall off and enter the crusher with other ore materials, serious damages to crusher gears and other equipment happen, which causes millions of economic loss, because it is made of high-manganese steel. Thus, it is urgent to develop an efficient and automatic algorithm for detecting broken teeth. However, existing methods for detecting broken teeth have little effect and most research studies depended on sensor skills, which will be disturbed by closed cavity in shovel and not stable in practice. In this paper, we present an intelligent computer vision system for monitoring teeth condition and detecting missing teeth. Since the pixel-level algorithm is carried out, the amount of calculation should be reduced to improve the superiority of the algorithm. To release computational pressure of subsequent work, salient detection based on deep learning is proposed for extracting the key frame images from video flow taken by the camera installed on the shovel including the teeth we intend to analyze. Additionally, in order to more efficiently monitor teeth condition and detect missing teeth, semantic segmentation based on deep learning is processed to get the relative position of the teeth in the image. Once semantic segmentation is done, floating images containing the shape of teeth are obtained. Then, to detect missing teeth effectively, image registration is proposed. Finally, the result of image registration shows whether teeth are missing or not, and the system will immediately alert staff to check the shovel when teeth fall off. Through sufficient experiments, statistical result had demonstrated superiority of our presented model that serves more promising prospect in mining industry.

Computed Tomography/Magnetic Resonance Imaging (CT/MRI) Image Registration and Fusion Assessment for Accurate Glioblastoma Radiotherapy Treatment Planning

Background: In this study, computed tomography/magnetic resonance imaging (CT/MRI) image registration and fusion in the 3D conformal radiotherapy treatment planning of Glioblastoma brain tumor was investigated. Good CT/MRI image registration and fusion made a great impact on dose calculation and treatment planning accuracy. Indeed, the uncertainly associated with the registration and fusion methods must be well verified and communicated. Unfortunately, there is no standard procedure or mathematical formalism to perform this verification due to noise, distortion, and complicated anatomical situations. Objectives: This study aimed at assessing the effective contribution of MRI in Glioma radiotherapy treatment by improving the localization of target volumes and organs at risk (OARs). It is also a question to provide clinicians with some suitable metrics to evaluate the CT/MRI image registration and fusion results. Methods: Quantitative image registration and fusion evaluation were used in this study to compare Eclipse TPS tools and Elastix CT/MRI image registration fusion. Thus, Dice score coefficient (DSC), Jaccard similarity coefficient (JSC), and Hausdorff distance (HD) were found to be suitable metrics for the evaluation and comparison of the image registration and fusion methods of Eclipse TPS and Elastix. Results: The programmed tumor’s volumes (PTV) delineated on CT slices were approximately 1.38 times smaller than those delineated on CT/MRI fused images. Large differences were observed for the edema and the brainstem. It was also found that MRI considerably optimized the dose to be delivered to the optic nerve and brainstem. Conclusions: Image registration and fusion is a fundamental step for suitable and efficient Glioma treatment planning in 3D conformal radiotherapy that ensure accurate dose delivery and unnecessary OAR irradiation. MRI can provide accurate localization of targeted volumes leading to better irradiation control of Glioma tumor.

Improvement Prediction on Contour Deformation Accuracy Using Deformable Image Registration Results Compared to Rigid Image Registration Results

The aim of this study was to analyze improvement prediction on contour deformation accuracy using deformable image registration (DIR) results compared to rigid image registration (RIR) results. Radiotherapy plans for 31 cases (seven head and neck cases, 10 chest cases, six abdomen cases and eight female pelvis cases) from the privately open database for DIR were used. These cases used at least two radiotherapy plans, and registration was performed using two plans, not only for one case but also for different cases. The DIR and RIR were performed using the DIR software MIM Maestro (MIM software Inc., Cleveland, USA). The registration results for the following organs were analyzed: eye balls, optic nerves, brain stem, spinal cord and right and left parotid glands for head and neck; right and left lungs for chest; liver and right and left kidneys for abdomen; and rectum and bladder for pelvis. Dice similarity coefficient (DSC) for the organs was calculated from the results of RIR and DIR. The improvement in the DSC was observed. DIR improved the DSC values by more than 0.2 for simple shapes, well-defined boundaries and large volumes such as eye balls, brain stem, lungs and liver. The minimum DSC for these organs was approximately 0.7. The improvement in DSC for the organs eye balls, brain stem, lungs and liver had ceiling values 0.95, 0.90, 1.0 and 1.0, respectively. DSC for the spinal cord, parotid gland, bladder and kidney also improved by DIR compared to RIR; however, DIR could not improve the DSC value for rectum compared to RIR because of a large difference in the position, shape and size due to stool and gas.

A Novel Coarse-to-Fine Scheme for Remote Sensing Image Registration Based on SIFT and Phase Correlation

Automatic image registration has been wildly used in remote sensing applications. However, the feature-based registration method is sometimes inaccurate and unstable for images with large scale difference, grayscale and texture differences. In this manuscript, a coarse-to-fine registration scheme is proposed, which combines the advantage of feature-based registration and phase correlation-based registration. The scheme consists of four steps. First, feature-based registration method is adopted for coarse registration. A geometrical outlier removal method is applied to improve the accuracy of coarse registration, which uses geometric similarities of inliers. Then, the sensed image is modified through the coarse registration result under affine deformation model. After that, the modified sensed image is registered to the reference image by extended phase correlation. Lastly, the final registration results are calculated by the fusion of the coarse registration and the fine registration. High universality of feature-based registration and high accuracy of extended phase correlation-based registration are both preserved in the proposed method. Experimental results of several different remote sensing images, which come from several published image registration papers, demonstrate the high robustness and accuracy of the proposed method. The evaluation contains root mean square error (RMSE), Laplace mean square error (LMSE) and red–green image registration results.

A mosaic method for multi-temporal data registration by using convolutional neural networks for forestry remote sensing applications

Image registration is one of the most important processes for the generation of remote sensing image mosaics. This paper focuses on the special problems related to remote sensing data registration, and multi-temporal data mosaic applications in the domain of forestry. It proposes an image registration method based on hierarchical convolutional features, and applies it to improve the efficiency of large scale forestry image mosaic generation. This method uses a deep learning architecture to adaptively obtain image features from deep convolutional neural networks. The features derived from different images at different depth are sent to a correlation filter to compute the similarity between them; then the locations of the feature points are computed precisely. Based on this method, we study forestry image registration and the mosaic framework. We apply our approach to remote sensing images under different weather and seasonal conditions, and compare the results with those generated using the traditional SIFT image mosaic method. The experimental result shows that our method can detect and match the image feature points with significant spectral difference, and effectively extract feature points to generate accurate image registration and mosaic results. This demonstrates the effectiveness and robustness of the proposed approach.

Multi-modal image analysis for semi-automatic segmentation of the total liver and liver arterial perfusion territories for radioembolization

PurposeWe have developed a multi-modal imaging approach for SIRT, combining 99mTc-MAA SPECT/CT and/or 90Y PET, 18F-FDG PET/CT, and contrast-enhanced CBCT for voxel-based dosimetry, as a tool for treatment planning and verification. For radiation dose prediction calculations, a segmentation of the total liver volume and of the liver perfusion territories is required.MethodIn this paper, we proposed a procedure for multi-modal image analysis to assist SIRT treatment planning. The pre-treatment 18F-FDG PET/CT, 99mTc-MAA SPECT/CT, and contrast-enhanced CBCT images were registered to a common space using an initial rigid, followed by a deformable registration. The registration was scored by an expert using Likert scores. The total liver was segmented semi-automatically based on the PET/CT and SPECT/CT images, and the liver perfusion territories were determined based on the CBCT images. The segmentations of the liver and liver lobes were compared to the manual segmentations by an expert on a CT image.ResultOur methodology showed that multi-modal image analysis can be used for determination of the liver and perfusion territories using CBCT in SIRT using all pre-treatment studies. The results for image registration showed acceptable alignment with limited impact on dosimetry.The image registration performs well according to the expert reviewer (scored as perfect or with little misalignment in 94% of the cases). The semi-automatic liver segmentation agreed well with manual liver segmentation (dice coefficient of 0.92 and an average Hausdorff distance of 3.04 mm). The results showed disagreement between lobe segmentation using CBCT images compared to lobe segmentation based on CT images (average Hausdorff distance of 14.18 mm), with a high impact on the dosimetry (differences up to 9 Gy for right and 21 Gy for the left liver lobe).ConclusionThis methodology can be used for pre-treatment dosimetry and for SIRT planning including the determination of the activity that should be administered to achieve the therapeutical goal. The inclusion of perfusion CBCT enables perfusion-based definition of the liver lobes, which was shown to be markedly different from the anatomical definition in some of the patients.

A prospective gating method to acquire a diverse set of free-breathing CT images for model-based 4DCT

Breathing motion modeling requires observation of tissues at sufficiently distinct respiratory states for proper 4D characterization. This work proposes a method to improve sampling of the breathing cycle with limited imaging dose.We designed and tested a prospective free-breathing acquisition protocol with a simulation using datasets from five patients imaged with a model-based 4DCT technique. Each dataset contained 25 free-breathing fast helical CT scans with simultaneous breathing surrogate measurements. Tissue displacements were measured using deformable image registration. A correspondence model related tissue displacement to the surrogate. Model residual was computed by comparing predicted displacements to image registration results. To determine a stopping criteria for the prospective protocol, i.e. when the breathing cycle had been sufficiently sampled, subsets of N scans where 5 ⩽ N ⩽ 9 were used to fit reduced models for each patient. A previously published metric was employed to describe the phase coverage, or ‘spread’, of the respiratory trajectories of each subset. Minimum phase coverage necessary to achieve mean model residual within 0.5 mm of the full 25-scan model was determined and used as the stopping criteria. Using the patient breathing traces, a prospective acquisition protocol was simulated.In all patients, phase coverage greater than the threshold necessary for model accuracy within 0.5 mm of the 25 scan model was achieved in six or fewer scans. The prospectively selected respiratory trajectories ranked in the (97.5 ± 4.2)th percentile among subsets of the originally sampled scans on average. Simulation results suggest that the proposed prospective method provides an effective means to sample the breathing cycle with limited free-breathing scans. One application of the method is to reduce the imaging dose of a previously published model-based 4DCT protocol to 25% of its original value while achieving mean model residual within 0.5 mm.

γTools: A modular multifunction phantom for quality assurance in GammaKnife treatments

PurposeWe present the γTools, a new phantom designed to assess geometric and dosimetric accuracy in Gamma Knife treatments, together with first tests and results of applications. MethodsThe phantom is composed of two modules: the imaging module, a regular grid of 1660 control points to evaluate image distortions and image registration result and the dosimetry module for delivered dose distribution measurements. The phantom is accompanied by a MatLab routine for image distortions quantification. Dose measurement are performed with Gafchromic films fixed between two inserts and placed in various positions and orientations inside the dosimetry module thus covering a volume comparable to the full volume of a head. ResultsTests performed to assess the accuracy and precision of the imaging module demonstrated sub-millimetric values. As an example of possible applications, the phantom was employed to measure image distortions of two MRI scanners and to perform dosimetric studies of single shots delivered to homogeneous and heterogeneous materials. Due to the phantom material, the measured absolute dose do not correspond to the planned dose; doses comparisons are thus carried out between normalized dose distributions. Finally, an end-to-end test was carried out in the treatment of a neuroma-like target which resulted in a 100% gamma passing rate (2% local, 2 mm) and a distance between the real target perimeter and the prescription isodose centroids of about 1 mm. ConclusionsThe tests demonstrate that the proposed phantom is suitable to assess both the geometrical and relative dosimetric accuracy of Gamma Knife radiosurgery treatments.

Rat Brain Registration Using Improved Speeded Up Robust Features

This study proposes an approach that combines edge feature extraction (EFE) with speeded up robust features (SURF) to register rat brain slices. When extracting feature points, using the traditional scale-invariant feature transform (SIFT) approach has several shortcomings, including blurry edge information and the inability to accurately extract smooth edge targets. Therefore, a feature point description method that comprises EFE and SURF is proposed to match local feature points, thereby effectively capturing edge feature points and avoiding mistaking grains and noise in images for feature points. In addition, a wavelet feature description approach is used to resolve incorrect matching caused by noise. Feature point matching for local regions is devised to solve mismatch errors that occur in highly similar regions in an image. Compared to a state-of-the-art SIFT method, the proposed method generates better results for rat brain image registration in terms of subjective visual presentation and registration quality criteria.

Fast Geodesic Regression for Population-Based Image Analysis.

Geodesic regression on images enables studies of brain development and degeneration, disease progression, and tumor growth. The high-dimensional nature of image data presents significant computational challenges for the current regression approaches and prohibits large scale studies. In this paper, we present a fast geodesic regression method that dramatically decreases the computational cost of the inference procedure while maintaining prediction accuracy. We employ an efficient low dimensional representation of diffeomorphic transformations derived from the image data and characterize the regressed trajectory in the space of diffeomorphisms by its initial conditions, i.e., an initial image template and an initial velocity field computed as a weighted average of pairwise diffeomorphic image registration results. This construction is achieved by using a first-order approximation of pairwise distances between images. We demonstrate the efficiency of our model on a set of 3D brain MRI scans from the OASIS dataset and show that it is dramatically faster than the state-of-the-art regression methods while producing equally good regression results on the large subject cohort.