Decoupled automated rotational and translational registration for functional MRI time series data: The dart registration algorithm

A rapid, in-plane image registration algorithm that accurately estimates and corrects for rotational and translational motion is described. This automated, one-pass method achieves its computational efficiency by decoupling the estimation of rotation and translation, allowing the application of rapid cross-correlation and cross-spectrum techniques for the determination of displacement parameters. k-space regridding and modulation techniques are used for image correction as alternatives to linear interpolation. The performance of this method was analyzed with simulations and echo-planar image data from both phantoms and human subjects. The processing time for image registration on a Hewlett-Packard 735/125 is 7.5 s for a 128 × 128 pixel image and 1.7 s for a 64 × 64 pixel image. Imaging phantom data demonstrate the accuracy of the method (mean rotational error, ?0.09°; standard deviation = 0.17°; range, ?0.44° to + 0.31°; mean translational error = ?0.035 pixels; standard deviation = 0.054 pixels; range, ?0.16 to + 0.06 pixels). Registered human functional imaging data demonstrate a significant reduction in motion artifacts such as linear trends in pixel time series and activation artifacts due to stimulus-correlated motion. The advantages of this technique are its noniterative one-pass nature, the reduction in image degradation as compared to previous methods, and the speed of computation.     

Method for spatially interleaving two images to halve EPI readout times: two reduced acquisitions interleaved (TRAIL).

A new MRI method is presented that can generate images using half the normal readout time or, more usefully, half the number of phase-encode steps, combining two readouts per excitation. However, the corresponding data are interleaved in image space-not in k-space, as in many other fast techniques. This gives a resilience to the phase-related artifacts that can occur in many other techniques due to subject motion. A modified stimulated-echo experiment is used to create two low-resolution images from a single sequence. The magnetization that contributes to these images is nonuniformly distributed within each pixel, forming two sinusoidal waves in quadrature, with an oscillation period of exactly two pixels. Since only half of each pixel contributes significant signal, the two images can be interleaved to create a full image with twice as many pixels and double the resolution. When the technique is used in the phase-encode direction, the effective imaging time is halved, though with two readouts per TR period. When two half-length echo-planar readouts are used, the method can also reduce blurring and distortion by halving the effective readout time for echo-planar imaging (EPI). For even further improvements, the technique can be combined with partial Fourier or parallel imaging.     

Transmission imaging for registration of ictal and interictal single-photon emission tomography, magnetic resonance imaging and electroencephalography

A method developed for registration of ictal and interictal single-photon emission tomography (SPET), magnetic resonance imaging (MRI) and electroencephalography (EEG) is described. For SPET studies, technetium-99m ethyl cysteinate dimer (ECD) was injected intravenously while the patient was monitored on video-EEG to document the ictal or interictal state. Imaging was performed using a triple-head gamma camera equipped with a transmission imaging device using a gadolinium-153 source. The images (128x128 pixels, voxel size 3.7x3.7x3.6 mm3) were reconstructed using an iterative algorithm and postfiltered with a Wiener filter. The gold-plated silver electrodes on the patient's scalp were utilized as markers for registration of the ictal and interictal SPET images, as these metallic markers were clearly seen on the transmission images. Fitting of the marker sets was based on a non-iterative least squares method. The interictal SPET image was subtracted from the ictal image after scaling. The T1-weighted MPRAGE MR images with voxel size of 1.0x1.0x1.0 mm3 were obtained with a 1.5-T scanner. For registration of MR and subtraction SPET images, the external marker set of the ictal SPET study was fitted to the surface of the head segmented from MR images. The SPET registration was tested with a phantom experiment. Registration of ictal and interictal SPET in five patient studies resulted in a 2-mm RMS residual of the marker sets. The estimated RMS error of registration in the final result combining locations of the electrodes, subtraction SPET and MR images was 3-5 mm. In conclusion, transmission imaging can be utilized for an accurate and easily implemented registration procedure for ictal and interictal SPET, MRI and EEG.     

基于体素灰度三维多模医学图像配准中相似性测度的选取

目的：在基于体素灰度医学图像配准领域，找出最适合于临床应用的多模医学图像配准相似性测度。方法：在极端的刚体配准条件下，检验出互相关系数，互信息和相关比相似性测度为适合的相似性测度。同时进一步解释了基于互信息相似性测度的医学图像配准易于陷入局部最优，而基于相关比相似性测度的方法易于保证配准得到全局最优，最后，利用加速的多分辨率配准方案和Powell‘s优化算法，对临床医学图像进行了基于相关比相似性测度的多模图像配准试验。结果：通过临床医学专家的判断，利用相关比相似性测度进行多模医学图像配准，安全能满足临床的要求，进行MR／CT，MR／PET三维多模医学图像配准时效果非常理想，结论：相比于其他相似性测度，互相关比相似性测度在基于体素灰度，三维多模医学图像配准领域，是一个更为适宜和准确的相似性测度。     

A contrast correction method for dental images based on histogram registration

Contrast correction is often required in digital subtraction radiography when comparing medical data acquired over different time periods owing to dissimilarities in the acquisition process. This paper focuses on dental radiographs and introduces a novel approach for correcting the contrast in dental image pairs.The proposed method modifies the subject images by applying typical registration techniques on their histograms. The proposed histogram registration method reshapes the histograms of the two subject images in such a way that these images are matched in terms of their contrast deviation. The method was extensively tested over 4 sets of dental images, consisting of 72 registered dental image pairs with unknown contrast differences as well as 20 dental pairs with known contrast differences. The proposed method was directly compared against the well-known histogram-based contrast correction method.The two methods were qualitatively and quantitatively evaluated for all 92 available dental image pairs. The two methods were compared in terms of the contrast root mean square difference between the reference image and the corrected image in each case. The obtained results were also verified statistically using appropriate t-tests in each set.The proposed method exhibited superior performance compared with the well-established method, in terms of the contrast root mean square difference between the reference and the corrected images. After suitable statistical analysis, it was deduced that the performance advantage of the proposed approach was statistically significant.     

Single-shot magnetic field mapping embedded in echo-planar time-course imaging.

A technique for acquiring magnetic field maps simultaneously with gradient-recalled echo-planar time-course data is described. This technique uses a trajectory in which the central part of k-space is collected twice. For a 64 x 64 image acquired with a 125-kHz bandwidth, a field map suitable for geometric correction can be collected simultaneously with the echo-planar time-course data in <70 ms. The field maps generated by this technique are registered with the magnitude images because they are calculated using the same data. They do not suffer from errors due to subject motion, or from different geometric distortions that can result from using different pulse sequences. In addition to correcting geometric distortions that resulted from dynamic magnetic field perturbations, this method was used to measure field shifts arising from respiration and jaw motion across five subjects. Values ranged from 0.035 to 0.165 parts per million (ppm).     

Transmission imaging for registration of ictal and interictal single-photon emission tomography, magnetic resonance imaging and electroencephalography

A method developed for registration of ictal and interictal single-photon emission tomography (SPET), magnetic resonance imaging (MRI) and electroencephalography (EEG) is described. For SPET studies, technetium-99m ethyl cysteinate dimer (ECD) was injected intravenously while the patient was monitored on video-EEG to document the ictal or interictal state. Imaging was performed using a triple-head gamma camera equipped with a transmission imaging device using a gadolinium-153 source. The images (128×128 pixels, voxel size 3.7×3.7×3.6 mm³) were reconstructed using an iterative algorithm and postfiltered with a Wiener filter. The gold-plated silver electrodes on the patient’s scalp were utilized as markers for registration of the ictal and interictal SPET images, as these metallic markers were clearly seen on the transmission images. Fitting of the marker sets was based on a non-iterative least squares method. The interictal SPET image was subtracted from the ictal image after scaling. The T1-weighted MPRAGE MR images with voxel size of 1.0×1.0×1.0 mm³ were obtained with a 1.5-T scanner. For registration of MR and subtraction SPET images, the external marker set of the ictal SPET study was fitted to the surface of the head segmented from MR images. The SPET registration was tested with a phantom experiment. Registration of ictal and interictal SPET in five patient studies resulted in a 2-mm RMS residual of the marker sets. The estimated RMS error of registration in the final result combining locations of the electrodes, subtraction SPET and MR images was 3–5 mm. In conclusion, transmission imaging can be utilized for an accurate and easily implemented registration procedure for ictal and interictal SPET, MRI and EEG. Received 20 September and in revised form 16 October 1999     

Processing strategies for time-course data sets in functional mri of the human brain

Image processing strategies for functional magnetic resonance imaging (FMRI) data sets acquired using a gradient-recalled echo-planar imaging sequence are considered. The analysis is carried out using the mathematics of vector spaces. Data sets consisting of N sequential images of the same slice of brain tissue are analyzed in the time-domain and also, after Fourier transformation, in the frequency domain. A technique for thresholding is introduced that uses the shape of the response in a pixel compared with the shape of a reference waveform as the decision criterion. A method is presented to eliminate drifts in data that arise from subject movement. The methods are applied to experimental FMRI data from the motor—cortex and compared with more conventional image—subtraction methods. Several finger motion paradigms are considered in the context of the various image processing strategies. The most effective method for image processing involves thresholding by shape as characterized by the correlation coefficient of the data with respect to a reference waveform followed by formation of a cross-correlation image. Emphasis is placed not only on image formation, but also on the use of signal processing techniques to characterize the temporal response of the brain to the paradigm.     

Registration and display of the combined bone scan and radiograph in the diagnosis and management of wrist injuries

A system has been developed for the registration and combined display of the X-ray image and isotope bone scan. This system has been evaluated by prospectively studying 23 patients who were referred for suspected injury of the wrist. The registration system has an inherent precision for registration of about 1 mm. When patient positioning errors are included, this increases to about 4 mm. Two observers evaluated the sets of images prior to registration and after registration and combined display. They judged that in 16 out of 17 cases judged abnormal (observer 1) and 12 out of 18 cases judged abnormal (observer 2), the registration technique improved localization of a lesion. One case was rejected as registration was not possible due to incorrect positioning of the hand.     

Image-based EPI ghost correction using an algorithm based on projection onto convex sets (POCS).

This work describes the use of a method, based on the projection onto convex sets (POCS) algorithm, for reduction of the N/2 ghost in echo-planar imaging (EPI). In this method, ghosts outside the parent image are set to zero and a model k-space is obtained from the Fourier transform (FT) of the resulting image. The zeroth- and first-order phase corrections for each line of the original k-space are estimated by comparison with the corresponding line in the model k-space. To overcome problems of phase wrapping, the first-order phase corrections for the lines of the original k-space are estimated by registration with the corresponding lines in the model k-space. It is shown that applying these corrections will result in a reduction of the ghost, and that iterating the process will result in a convergence towards an image in which the ghost is minimized. The method is tested on spin-echo EPI data. The results show that the method is robust and remarkably effective, reducing the N/2 ghost to a level nearly comparable to that achieved with reference scans.     

Assessment of a technique for 2D-3D registration of cerebral intra-arterial angiography

This study assesses the ability of a computer algorithm to perform automated 2D-3D registrations of digitally subtracted cerebral angiograms. The technique was tested on clinical studies of five patients with intracranial aneurysms. The automated procedure was compared against a gold standard manual registration, and achieved a mean registration accuracy of 1.3 mm (SD 0.6 mm). Two registration strategies were tested using coarse (128 x 128 pixel) or fine (256 x 256 pixel) images. The mean registration errors proved similar but registration of the lower resolution images was 3 times quicker (mean registration times 33 s, SD 13 s for low and 150 s SD 48 s for high resolution images). The automated techniques were considerably faster than manual registrations but achieved similar accuracy. The technique has several potential uses but is particularly applicable to endovascular treatment techniques.     

Improving accuracy and efficiency of mutual information for multi-modal retinal image registration using adaptive probability density estimation

Mutual information (MI) is a popular similarity measure for performing image registration between different modalities. MI makes a statistical comparison between two images by computing the entropy from the probability distribution of the data. Therefore, to obtain an accurate registration it is important to have an accurate estimation of the true underlying probability distribution. Within the statistics literature, many methods have been proposed for finding the ‘optimal’ probability density, with the aim of improving the estimation by means of optimal histogram bin size selection. This provokes the common question of how many bins should actually be used when constructing a histogram. There is no definitive answer to this. This question itself has received little attention in the MI literature, and yet this issue is critical to the effectiveness of the algorithm. The purpose of this paper is to highlight this fundamental element of the MI algorithm. We present a comprehensive study that introduces methods from statistics literature and incorporates these for image registration. We demonstrate this work for registration of multi-modal retinal images: colour fundus photographs and scanning laser ophthalmoscope images. The registration of these modalities offers significant enhancement to early glaucoma detection, however traditional registration techniques fail to perform sufficiently well. We find that adaptive probability density estimation heavily impacts on registration accuracy and runtime, improving over traditional binning techniques.     

Correction of motion artifacts from cardiac cine magnetic resonance images

RATIONALE AND OBJECTIVES: An image registration method was developed to automatically correct motion artifacts, mostly from breathing, from cardiac cine magnetic resonance (MR) images. MATERIALS AND METHODS: The location of each slice in an image stack was optimized by maximizing a similarity measure of the slice with another image slice stack. The optimization was performed iteratively and both image stacks were corrected simultaneously. Two procedures to optimize the similarity were tested: standard gradient optimization and stochastic optimization in which one slice is chosen randomly from the image stacks and its location is optimized. In this work, cine short- and long-axis images were used. In addition to visual inspection results from real data, the performance of the algorithm was evaluated quantitatively by simulating the movements in four real MR data sets. The mean error and standard deviation were defined for 50 simulated movements as each slice was randomly displaced. The error rate, defined as the percentage of non-satisfactory registration results, was evaluated. The paired t-test was used to evaluate the statistical difference between the tested optimization methods. RESULTS: The algorithm developed was successfully applied to correct motion artifacts from real and simulated data. The results, where typical motion artifacts were simulated, indicated an error rate of about 3%. Subvoxel registration accuracy was also achieved. When different optimization methods were compared, the registration accuracy of the stochastic approach proved to be superior to the standard gradient technique (P < 10(-9)). CONCLUSIONS: The novel method was capable of robustly and accurately correcting motion artifacts from cardiac cine MR images.     

Chemical-shift imaging: a hybrid approach

The hybrid technique of projection-reconstruction echo-planar (PREP) imaging for obtaining chemical-shift images is demonstrated experimentally using a fluorine sample. The technique which is a variation on echo-planar imaging (EPI) relies on a multipass procedure. It is nevertheless quite efficient and consequently chemical-shift images may be produced in a few minutes. The method produces images in 64 chemical-shift regions, each region mapped spatially by 64 X 64 pixels. The imaging time was just over 4 min. These 64 chemical-shifted images can be straightforwardly added together to form an undistorted image of the complete object. In addition the chemical-shift spectrum can be extracted and the various chemical-shift images can be unambiguously assigned to the spectral peaks.     

Registration and display of the combined bone scan and radiograph in the diagnosis and management of wrist injuries

A system has been developed for the registration and combined display of the X-ray image and isotope bone scan. This system has been evaluated by prospectively studying 23 patients who were referred for suspected injury of the wrist. The registration system has an inherent precision for registration of about 1 mm. When patient positioning errors are included, this increases to about 4 mm. Two observers evaluated the sets of images prior to registration and after registration and combined display. They judged that in 16 out of 17 cases judged abnormal (observer 1) and 12 out of 18 cases judged abnormal (observer 2), the registration technique improved localization of a lesion. One case was rejected as registration was not possible due to incorrect positioning of the hand. Offprint requests to: D.J. Hawkes     

Spatial and temporal registration of cardiac SPECT and MR images: methods and evaluation

Image registration may assist in the integration of information from multiple sources by allowing direct point-for-point comparisons of studies. To determine the usefulness of such a technique, a method for the spatial and temporal registration of four-dimensional single photon emission computed tomographic (SPECT) and magnetic resonance (MR) cardiac images was developed. Automatically detected left ventricular endocardial surfaces were used to determine the best transform between the two sets of surface points, and that transform was applied to the original SPECT image. A fused image created from the MR and the transformed SPECT images combined the information in both. The authors tested the method with seven patient studies. Registration reduced the distance between the MR and SPECT left ventricular endocardial surfaces by 30%, to an average of 2.7 mm. The authors found that, by using the fused images, perfusion abnormalities could be easily localized and correlated to high-resolution endocardial wall motion and systolic wall thickening.     

Elastic registration of 3D cartilage surfaces from MR image data for detecting local changes in cartilage thickness.

The objective of this work was to develop and validate a computational method for the registration (matching) of 3D cartilage plates from MR image data sets. The technique tracks local cartilage thickness changes over time. A 3D elastic registration technique was applied that identifies corresponding points of the bone-cartilage interface in MR data sets of 3D-reconstructed cartilage plates. In a first rigid preregistration step, the surfaces are aligned, using the principal axes decomposition to correct for different joint positions and orientations in the MR scanner. In a second step, the surfaces are deformed elastically, based on geometric surface features, until they are sufficiently similar to identify corresponding surface points. The method was validated against artificially corrupted cartilage surfaces and MR data obtained from in vivo and in vitro compression experiments. The in vivo reproducibility was tested on patellar data sets of volunteers, with repositioning of the joint in between replicate acquisitions.     

Registration of three-dimensional MR and PET images of the human brain without markers

Axial and sagittal magnetic resonance (MR) sections and contiguous sections of axial positron emission tomographic (PET) images obtained with fludeoxyglucose F-18 were used to evaluate a new method of registering three-dimensional images of the brain. The users specified the interhemispheric fissure plane in three dimensions for both the MR and PET data sets by specifying its endpoints within several axial sections. A transformation matrix aligning the interhemispheric fissure plane in MR and PET space was calculated and used to create one resectioned PET image on the resectioned PET image, and the user specified the remaining translations and rotation by moving the overlaid outline of the MR image. MR and PET data sets in four subjects were registered. The three-dimensional error on average was less than 3.8 mm and never exceeded 7.5 mm. Less than 1 hour per patient was required for registration. The method is accurate unless the interhemispheric fissure deviates significantly from a planar configuration. It does not need thin or contiguous MR sections and provides an estimate of the total registration error for every case.     

Computer-aided diagnosis in the detection of temporal changes in sequential chest radiographs

To aid radiologists in the diagnosis of screening chest radiographs, a temporal subtraction technique using digital image processing was developed. The accurate image registration of two sequential images enables us to detect even subtle changes in the "difference image" between them. In this report, a new method based on matching "lung markings" is introduced. Twenty-nine pairs of sequential posteroanterior chest radiographs with and without temporal changes were selected from cases examined with the computed radiography system. Image registration was employed, with the local matching of "lung markings" in previous and current radiographs. Observer performance tests were carried out by eight radiologists, with and without the "difference image." Observer performance tests with the temporal subtraction image showed that six of eight observers diagnosed them with higher sensitivity (mean, 43.9% vs. 55.3%) and a comparable false positive response. Mean area under the AFROC (alternative free-response receiver operating characteristics) curve also improved from 0.596 to 0.647, a statistically significant difference. The subtraction image using this registration technique improved diagnostic accuracy for subtle temporal changes.     

Simultaneous multislice acquisition with arterial-flow tagging (SMART) using echo planar imaging (EPI)

Arterial spin tagging techniques have been used to image tissue perfusion in MR without contrast injection or ionizing radiation. Currently, spin tagging studies are performed primarily using single-slice imaging sequences, which are time consuming. This note reports a multislice echo-planar arterial spin tagging technique (Simultaneous Multislice Acquisition with aRterial-flow Tagging, or “SMART”). Multiband RF encoding (Hadamard) is used to provide simultaneous multislice acquisition capability for spin tagging techniques (such as echo planar imaging signal targeting with alternating radio frequency and flow-sensitive alternative inversion recovery). The method is illustrated with a two-slice pulse sequence that was implemented using the FAIR technique to generate two perfusion weighted images simultaneously. Compared with single-slice sequences, this two-slice sequence provided similar image quality, signal-to-noise ratio, and twice the spatial coverage compared with the single-slice technique within the same scan time.