Automatic Image Registration Algorithm Research Articles

Image registration method using representative feature detection and iterative coherent spatial mapping for infrared medical images with flat regions

In the registration of medical images, nonrigid registration targets, images with large displacement caused by different postures of the human body, and frequent variations in image intensity due to physiological phenomena are substantial problems that make medical images less suitable for intensity-based image registration modes. These problems also greatly increase the difficulty and complexity of feature detection and matching for feature-based image registration modes. This research introduces an automatic image registration algorithm for infrared medical images that offers the following benefits: effective detection of feature points in flat regions (cold patterns) that appear due to changes in the human body’s thermal patterns, improved mismatch removal through coherent spatial mapping for improved feature point matching, and large-displacement optical flow for optimal transformation. This method was compared with various classical gold standard image registration methods to evaluate its performance. The models were compared for the three key steps of the registration process—feature detection, feature point matching, and image transformation—and the results are presented visually and quantitatively. The results demonstrate that the proposed method outperforms existing methods in all tasks, including in terms of the features detected, uniformity of feature points, matching accuracy, and control point sparsity, and achieves optimal image transformation. The performance of the proposed method with four common image types was also evaluated, and the results verify that the proposed method has a high degree of stability and can effectively register medical images under a variety of conditions.

PERFORMANCE ANALYSIS OF SELECTED FEATURE DESCRIPTORS USED FOR AUTOMATIC IMAGE REGISTRATION

Abstract. Automatic detection and extraction of corresponding features is very crucial in the development of an automatic image registration algorithm. Different feature descriptors have been developed and implemented in image registration and other disciplines. These descriptors affect the speed of feature extraction and the measure of extracted conjugate features, which affects the processing speed and overall accuracy of the registration scheme. This article is aimed at reviewing the performance of most-widely implemented feature descriptors in an automatic image registration scheme. Ten (10) descriptors were selected and analysed under seven (7) conditions viz: Invariance to rotation, scale and zoom, their robustness, repeatability, localization and efficiency using UAV acquired images. The analysis shows that though four (4) descriptors performed better than the other Six (6), no single feature descriptor can be affirmed to be the best, as different descriptors perform differently under different conditions. The Modified Harris and Stephen Corner Detector (MHCD) proved to be invariant to scale and zoom while it is excellent in robustness, repeatability, localization and efficiency, but it is variant to rotation. Also, the Scale Invariant feature Transform (SIFT), Speeded Up Robust Features (SURF) and the Maximally Stable Extremal Region (MSER) algorithms proved to be invariant to scale, zoom and rotation, and very good in terms of repeatability, localization and efficiency, though MSER proved to be not as robust as SIFT and SURF. The implication of the findings of this research is that the choice of feature descriptors must be informed by the imaging conditions of the image registration analysts.

Algorithm Selection in Multimodal Medical Image Registration

Over the past few decades, fast<strong>-</strong>growth has occurred in the area of medical image acquisition devices, and physicians now rely on the utilization of medical images for the diagnosis, treatment plans, and surgical guidance. Researchers have classified medical images according to two structures: anatomical and functional structures. Due to this classification, the data obtained from two or more images of the same object frequently provide complementary and more abundant information through a process known as multimodal medical model registration. Image registration is spatially mapping the coordinate system of the two images obtained from a different viewpoint and utilizing various sensors. Several automatic multimodal medical image registration algorithms have been introduced based on types of medical images and their applications to increase the reliability, robustness, and accuracy. Due to the diversity in imaging and the different demands<strong> </strong>for applications, there is no single registration algorithm that can do that. This paper introduces a novel method for developing a multimodal medical image registration system that can select the most accepted registration algorithm from a group of registration algorithms for a variety of input datasets. The method described here is based on a machine learning technique that selects the most promising candidate. Several experiments have been conducted, and the results reveal that the novel approach leads to considerably faster reliability, accuracy, and more robustness registration algorithm selection.

Geometry planning and image registration in magnetic particle imaging using bimodal fiducial markers.

Magnetic particle imaging (MPI) is a quantitative imaging modality that allows the distribution of superparamagnetic nanoparticles to be visualized. Compared to other imaging techniques like x-ray radiography, computed tomography (CT), and magnetic resonance imaging (MRI), MPI only provides a signal from the administered tracer, but no additional morphological information, which complicates geometry planning and the interpretation of MP images. The purpose of the authors' study was to develop bimodal fiducial markers that can be visualized by MPI and MRI in order to create MP-MR fusion images. A certain arrangement of three bimodal fiducial markers was developed and used in a combined MRI/MPI phantom and also during in vivo experiments in order to investigate its suitability for geometry planning and image fusion. An algorithm for automated marker extraction in both MR and MP images and rigid registration was established. The developed bimodal fiducial markers can be visualized by MRI and MPI and allow for geometry planning as well as automated registration and fusion of MR-MP images. To date, exact positioning of the object to be imaged within the field of view (FOV) and the assignment of reconstructed MPI signals to corresponding morphological regions has been difficult. The developed bimodal fiducial markers and the automated image registration algorithm help to overcome these difficulties.

Abstract B2-51: SpEeCH: Quantifying Spatial Expression of Clonal Heterogeneity in breast cancer

Abstract In routine breast cancer clinical practice, multiple serial sections are stained for guiding therapy. These include the most widely used hematoxylin & eosin (H&E) as well as established immunohistochemical biomarkers such as estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and increasingly so, Ki67, collectively known as the IHC4. Analysis of these sections involves visual assessment which is labor intensive and inefficient. More importantly, it ignores the wealth of spatial information in histological samples that can be used to better understand clonal heterogeneity, which may potentially be responsible for mixed response to treatment and drug resistance. Material and Methods: We have developed an automated image registration algorithm that aligns H&E and IHC4 serial section images from the same tumor by employing a combination of rigid and non-rigid registration. This automated image registration algorithm subsequently enables us to develop SpEeCH (spatial expression of clonal heterogeneity), a novel method to identify spatial clusters with differential expression. At the core of SpEeCH is a clustering method that exploits spatial cellular characteristics of tumor regions to identify spatial clonal clusters. Gaussian mixture Model (GMM) is used in combination with the Bayesian information criterion and resampling to identify the optimal number of stable clusters. As GMM ignores local spatial dependency which is an important tumor characteristic, we are also comparing GMM with conditional random fields that are undirected graphical models to account for local dependencies instead of full joint distribution. To discover spatial clusters, we applied SpEeCH to a set of 13 high-grade surgically resected breast tumors with serial sections of H&E and IHC4. Tumor regions were defined as 5 μm x 5 μm squares, resulting in >10,000 regions for clustering in each tumor. Clinical implications of the spatial clusters were verified in an independent dataset (METABRIC) that consists of gene expression whole-tumor profiles and disease-specific survival data of 2,000 breast cancer tumors. Results: By employing SpEeCH, we identified five stable spatial clusters based on histological section of H&E and IHC4 in the discovery set, with significantly different disease-specific survival validated in METABRIC gene expression data (p=0, log-rank test). This analysis revealed a group of poor prognosis ER-/HER2+/Ki67+ patients (p=0, hazard ratio = 4.4, 95% confidence interval = 3.0-6.3, log-rank test compared to a good prognosis group ER+/HER2-/Ki67-). Marked differences in prognosis among these groups highlight the significance of identified spatial clusters. For example, one of the tumors in the discovery set contains both ER-/HER2+/Ki67+ and ER+/HER2-/Ki67- clones. Therefore, such clonal heterogeneity, which is ignored during diagnosis, may help explain treatment resistance. To this end, our observation confirmed the clinical relevance of spatial clusters by leveraging a large-scale dataset under the assumption that such data represents signal from major clones. We propose that the number of clonal clusters and their spatial relationship within a tumor can be used as measures of intra-tumor heterogeneity. Their efficacy is currently being evaluated on independent clinical datasets. Conclusion: Large numbers of histology samples are available from clinical diagnostics and research laboratories, yet there is a lack of tools that facilitate objective analysis of tumor heterogeneity in these samples. Compared to high-cost multimarker immunofluorescence approaches, our automated image and statistical analysis pipeline enables efficient and effective analysis of available histology samples, which is critical for understanding cancers as highly heterogeneous diseases. Citation Format: Andreas Heindl, Adnan Mujahid Khan, Yinyin Yuan. SpEeCH: Quantifying Spatial Expression of Clonal Heterogeneity in breast cancer. [abstract]. In: Proceedings of the AACR Special Conference on Computational and Systems Biology of Cancer; Feb 8-11 2015; San Francisco, CA. Philadelphia (PA): AACR; Cancer Res 2015;75(22 Suppl 2):Abstract nr B2-51.

SU‐E‐J‐29: Automatic Image Registration Performance of Three IGRT Systems for Prostate Radiotherapy

Purpose:To compare the performance of an automatic image registration algorithm on image sets collected on three commercial image guidance systems, and explore its relationship with imaging parameters such as dose and sharpness.Methods:Images of a CIRS Virtually Human Male Pelvis phantom (VHMP) were collected on the CBCT systems of Varian TrueBeam/OBI and Elekta Synergy/XVI linear accelerators, across a range of mAs settings; and MVCT on a Tomotherapy Hi‐ART accelerator with a range of pitch. Using the 6D correlation ratio algorithm of XVI, each image was registered to a mask of the prostate volume with a 5 mm expansion. Registrations were repeated 100 times, with random initial offsets introduced to simulate daily matching. Residual registration errors were calculated by correcting for the initial phantom set‐up error. Automatic registration was also repeated after reconstructing images with different sharpness filters.Results:All three systems showed good registration performance, with residual translations <0.5mm (1σ) for typical clinical dose and reconstruction settings. Residual rotational error had larger range, with 0.8°, 1.2° and 1.9° for 1σ in XVI, OBI and Tomotherapy respectively. The registration accuracy of XVI images showed a strong dependence on imaging dose, particularly below 4mGy. No evidence of reduced performance was observed at the lowest dose settings for OBI and Tomotherapy, but these were above 4mGy. Registration failures (maximum target registration error > 3.6 mm on the surface of a 30mm sphere) occurred in 5% to 10% of registrations. Changing the sharpness of image reconstruction had no significant effect on registration performance.Conclusions:Using the present automatic image registration algorithm, all IGRT systems tested provided satisfactory registrations for clinical use, within a normal range of acquisition settings.

An Automatic Robust Image Registration Algorithm for Aerial Mapping

Aerial mapping is attracting more attention due to the development in unmanned aerial vehicles (UAVs) and their availability and also vast applications that require a wide aerial photograph of a region in a specific time. The cross-modality as well as translation, rotation, scale change and illumination are the main challenges in aerial image registration. This paper concentrates on an algorithm for aerial image registration to overcome the aforementioned issues. The proposed method is able to sample automatically and align the sensed images to form the final map. The results are compared with satellite images that shows a reasonable performance with geometrically correct registration.

Preprocedural magnetic resonance imaging for image-guided catheter ablation of scar-related ventricular tachycardia.

To present and validate a highly automated MRI analysis workflow for image-guided catheter ablation of scar-related ventricular tachycardia (VT) ablation procedures. A cohort of 15 post-infarction patients underwent MRI prior to VT ablation. The MRI study included a black-blood turbo spin echo sequence for visualizing the aortic root and ostium of the left main (LM) coronary artery, and a 3D late gadolinium enhanced sequence for visualizing the LV anatomy and myocardial scar substrate. Semi-automated segmentation of the LV, aortic root and ostium of LM was performed, followed by fully automated segmentation of myocardial scar. All segmented structures were aligned using an automated image registration algorithm to remove inter-scan displacement. MRI was integrated at the beginning of the procedure after mapping a single LM point. The integration performance was compared to that of the traditional iterative closest point (ICP) method. The proposed method required a single LM mapping point only, compared to 255 ± 43 points with the ICP method. The single-point method achieved a mean point-to-surface distance of 4.9 ± 1.5 mm on the LV surface and 5.1 ± 1.7 mm on the aorta surface (ICP: 3.7 ± 0.8 and 9.2 ± 7.2 mm, P < 0.05). The Cohen's kappa coefficient between the MRI-defined and EAM-defined scar was 0.36 ± 0.16 for the presented method, significantly higher than that of ICP method (0.23 ± 0.21, P = 0.03), indicating more accurate scar substrate localization during integration. This study demonstrated the feasibility of preprocedural MRI integration into the VT ablation procedure, with highly automated image analysis workflow and minimal mapping effort.

The Automatic Image Registration Algorithm Based on Bag of Words and Features Point

According to actual requirements and situation, a novel automatic image registration algorithm was presented based on bag of words and features point. Firstly, pyramid delaminating was used to preprocess the pre-registration images. Then, image normalized could be obtained. Secondly, the dense SIFT features points could be extracted. The paper could get the feature eigenvectors descriptions of image by k-means clustering algorithm. Thirdly, the feature eigenvectors descriptions of image were train and the reference and pre-registration images were classification through Support Vector Machine (SVM). Finally, image registration could implement by features matching points. The results of experiment with the with office room and building test images shown the proposed algorithm was effective and could obtain the better registration image.

SU‐E‐I‐94: Automated Image Quality Assessment of Radiographic Systems Using An Anthropomorphic Phantom

Purpose:In a large, academic medical center, consistent radiographic imaging performance is difficult to routinely monitor and maintain, especially for a fleet consisting of multiple vendors, models, software versions, and numerous imaging protocols. Thus, an automated image quality control methodology has been implemented using routine image quality assessment with a physical, stylized anthropomorphic chest phantom.Methods:The “Duke” Phantom (Digital Phantom 07‐646, Supertech, Elkhart, IN) was imaged twice on each of 13 radiographic units from a variety of vendors at 13 primary care clinics. The first acquisition used the clinical PA chest protocol to acquire the post‐processed “FOR PRESENTATION” image. The second image was acquired without an antiscatter grid followed by collection of the “FOR PROCESSING” image. Manual CNR measurements were made from the largest and thickest contrast‐detail inserts in the lung, heart, and abdominal regions of the phantom in each image. An automated image registration algorithm was used to estimate the CNR of the same insert using similar ROIs. Automated measurements were then compared to the manual measurements.Results:Automatic and manual CNR measurements obtained from “FOR PRESENTATION” images had average percent differences of 0.42%±5.18%, −3.44%±4.85%, and 1.04%±3.15% in the lung, heart, and abdominal regions, respectively; measurements obtained from “FOR PROCESSING” images had average percent differences of ‐0.63%±6.66%, −0.97%±3.92%, and −0.53%±4.18%, respectively. The maximum absolute difference in CNR was 15.78%, 10.89%, and 8.73% in the respective regions. In addition to CNR assessment of the largest and thickest contrast‐detail inserts, the automated method also provided CNR estimates for all 75 contrast‐detail inserts in each phantom image.Conclusion:Automated analysis of a radiographic phantom has been shown to be a fast, robust, and objective means for assessing radiographic image quality. The method reduces the burden of manual measurements and provides a means by which to monitor and compare radiographic system performance.

Robust initialisation for single-plane 3D CT to 2D fluoroscopy image registration

Gradient descent-based automatic image registration algorithms typically fail when the initial misalignment between objects is large. This is a major limitation for routine clinical applications. The registration task is even more difficult for multi-modal images because of the nonlinear relationship between the pixel intensities in the images to be aligned. In this paper, we present a fast and accurate multi-modal image registration algorithm which successfully registers three-dimensional (3D) computed tomography to two-dimensional single-plane fluoroscopy data for large initial displacements between the images. Our experimental results show that the proposed approach can increase the range of initial displacements up to ± 20 mm for all translations and up to ± 20° for rotation while maintaining high precision and small bias in all six 3D rigid body transform parameters.

Automatic ultrasound-MRI registration for neurosurgery using the 2D and 3D LC(2) Metric.

To enable image guided neurosurgery, the alignment of pre-interventional magnetic resonance imaging (MRI) and intra-operative ultrasound (US) is commonly required. We present two automatic image registration algorithms using the similarity measure Linear Correlation of Linear Combination (LC(2)) to align either freehand US slices or US volumes with MRI images. Both approaches allow an automatic and robust registration, while the three dimensional method yields a significantly improved percentage of optimally aligned registrations for randomly chosen clinically relevant initializations. This study presents a detailed description of the methodology and an extensive evaluation showing an accuracy of 2.51mm, precision of 0.85mm and capture range of 15mm (>95% convergence) using 14 clinical neurosurgical cases.

Automatic image registration performance for two different CBCT systems; variation with imaging dose

The performance of an automatic image registration algorithm was compared on image sets collected with two commercial CBCT systems, and the relationship with imaging dose was explored. CBCT images of a CIRS Virtually Human Male Pelvis phantom (VHMP) were collected on Varian TrueBeam/OBI and Elekta Synergy/XVI linear accelerators, across a range of mAs settings. Each CBCT image was registered 100 times, with random initial offsets introduced. Image registration was performed using the grey value correlation ratio algorithm in the Elekta XVI software, to a mask of the prostate volume with 5 mm expansion. Residual registration errors were calculated after correcting for the initial introduced phantom set-up error. Registration performance with the OBI images was similar to that of XVI. There was a clear dependence on imaging dose for the XVI images with residual errors increasing below 4mGy. It was not possible to acquire images with doses lower than ~5mGy with the OBI system and no evidence of reduced performance was observed at this dose. Registration failures (maximum target registration error > 3.6 mm on the surface of a 30mm sphere) occurred in 5% to 9% of registrations except for the lowest dose XVI scan (31%). The uncertainty in automatic image registration with both OBI and XVI images was found to be adequate for clinical use within a normal range of acquisition settings.

Accurate 3D shape measurement of multiple separate objects with stereo vision

3D shape measurement has emerged as a very useful tool in numerous fields because of its wide and ever-increasing applications. In this paper, we present a passive, fast and accurate 3D shape measurement technique using stereo vision approach. The technique first employs a scale-invariant feature transform algorithm to detect point matches at a number of discrete locations despite the discontinuities in the images. Then an automated image registration algorithm is applied to find full-field point matches with subpixel accuracy. After that, the 3D shapes of the objects can be reconstructed according to the obtained point matching and the camera information. The proposed technique is capable of performing a full-field 3D shape measurement with high accuracy even in the presence of discontinuities and multiple separate regions. The validity is verified by experiments.

An Effective Non-rigid Registration Approach for Ultrasound Image Based On “Demons” Algorithm

Medical image registration is an important component of computer-aided diagnosis system in diagnostics, therapy planning, and guidance of surgery. Because of its low signal/noise ratio (SNR), ultrasound (US) image registration is a difficult task. In this paper, a fully automatic non-rigid image registration algorithm based on demons algorithm is proposed for registration of ultrasound images. In the proposed method, an "inertia force" derived from the local motion trend of pixels in a Moore neighborhood system is produced and integrated into optical flow equation to estimate the demons force, which is helpful to handle the speckle noise and preserve the geometric continuity of US images. In the experiment, a series of US images and several similarity measure metrics are utilized for evaluating the performance. The experimental results demonstrate that the proposed method can register ultrasound images efficiently, robust to noise, quickly and automatically.

Dynamic lung modeling and tumor tracking using deformable image registration and geometric smoothing.

A greyscale-based fully automatic deformable image registration algorithm, based on an optical flow method together with geometric smoothing, is developed for dynamic lung modeling and tumor tracking. In our computational processing pipeline, the input data is a set of 4D CT images with 10 phases. The triangle mesh of the lung model is directly extracted from the more stable exhale phase (Phase 5). In addition, we represent the lung surface model in 3D volumetric format by applying a signed distance function and then generate tetrahedral meshes. Our registration algorithm works for both triangle and tetrahedral meshes. In CT images, the intensity value reflects the local tissue density. For each grid point, we calculate the displacement from the static image (phase 5) to match with the moving image (other phases) by using merely intensity values of the CT images. The optical flow computation is followed by a regularization of the deformation field using geometric smoothing. Lung volume change and the maximum lung tissue movement are used to evaluate the accuracy of the application. Our testing results suggest that the application of deformable registration algorithm is an effective way for delineating and tracking tumor motion in image-guided radiotherapy.

Evaluation of automated image registration algorithm for image-guided radiotherapy (IGRT)

The performance of an image registration (IR) software was evaluated for automatically detecting known errors simulated through the movement of ExactCouch using an onboard imager. Twenty-seven set-up errors (11 translations, 10 rotations, 6 translation and rotation) were simulated by introducing offset up to ± 15 mm in three principal axes and 0° to ± 1° in yaw. For every simulated error, orthogonal kV radiograph and cone beam CT were acquired in half-fan (CBCT_HF) and full-fan (CBCT_FF) mode. The orthogonal radiographs and CBCTs were automatically co-registered to reference digitally reconstructed radiographs (DRRs) and planning CT using 2D-2D and 3D-3D matching software based on mutual information transformation. A total of 79 image sets (ten pairs of kV X-rays and 69 session of CBCT) were analyzed to determine the (a) reproducibility of IR outcome and (b) residual error, defined as the deviation between the known and IR software detected displacement in translation and rotation. The reproducibility of automatic IR of planning CT and repeat CBCTs taken with and without kilovoltage detector and kilovoltage X-ray source arm movement was excellent with mean SD of 0.1 mm in the translation and 0.0° in rotation. The average residual errors in translation and rotation were within ± 0.5 mm and ± 0.2°, ± 0.9 mm and ± 0.3°, and ± 0.4 mm and ± 0.2° for setup simulated only in translation, rotation, and both translation and rotation. The mean (SD) 3D vector was largest when only translational error was simulated and was 1.7 (1.1) mm for 2D-2D match of reference DRR with radiograph, 1.4 (0.6) and 1.3 (0.5) mm for 3D-3D match of reference CT and CBCT with full fan and half fan, respectively. In conclusion, the image-guided radiation therapy (IGRT) system is accurate within 1.8 mm and 0.4° and reproducible under control condition. Inherent error from any IGRT process should be taken into account while setting clinical IGRT protocol.

Preserving the accuracy and resolution of the sodium bioscale from quantitative sodium MRI during intrasubject alignment across longitudinal studies

Emerging applications of sodium bioscales derived from quantitative sodium magnetic resonance imaging assess temporal changes in regional sodium concentration over intervals that vary from hours (monitoring tissue viability in stroke) to weeks (monitoring brain tumor treatment during radiation therapy) or even years (monitoring progression of neurodegenerative disease). Accurate interpretation of such quantitative data requires precise registration between magnetic resonance imaging sessions to avoid session-to-session changes in partial volume effects between normal tissue (∼38 mM sodium concentration), lesions (variable sodium concentration), and cerebrospinal fluid (∼144 mM sodium concentration). The existing Automated Image Registration algorithm is shown to be suitable for rapid, accurate, and precise determination of the transform that aligns sodium magnetic resonance images. Implementation of this transform during image reconstruction from the k-space data is shown to produce smaller errors than conventional image-domain interpolation. Experimental results at 9.4 T and 3.0 T demonstrating this registration approach to sodium data illustrate preservation of quantification accuracy during alignment of sodium magnetic resonance images acquired from the same subject during different imaging sessions.

Data Fusion of Radar Image and ECDIS Based on Matching of Harris Feature Points

This paper analyzes automatic image registration algorithm, and image features of radar and electronic chart, then, image registration algorithm based on Harris feature point are given, in which the methods are specially described. Wavelet transformation method on image feature should be considered. Moreover, simulate results of image fusion.

Video-Endoscopic Image Registration for Cartography Reconstruction of Hollow Organs

In examinations of hollow organs, video-endoscopic systems are used. These systems allow to analyze layers recovering the internal organ cavities (internal surfaces). In this kind of analysis, an expert must review the acquired image sequence during an endoscopic analysis, either to confirm the diagnosis issue of endoscope, or to evaluate the pathology’s evolution. For these reasons, it is difficult to find the surface of interest in a video-endoscopic image sequence (where each image shows a litt le area of examined tissue surface) due to: (1) the big number of images into the sequence, (2) the sequential nature of fi le (forcing doctor to review all sequence to find the interest region), and (3) frequently the interest region that could be physically contiguous is shown in frames far-away from each other. In this work, a methodology for automatic cartography reconstruction of the interest region is proposed. This methodology provides a panoramic view and eliminates redundant information. Furthermore, an automatic image registration algorithm for cartography construction is presented. Finally, an evaluation protocol and the obtained results are also presented.