Joseph K. Maier

The technical characteristics of matched filtering in digital subtraction angiography

The technical characteristics of a new digital fluorographic image processing method called matched filtering are presented. This technique, a type of extensive temporal integration, takes a weighted sum of images acquired during passage of a contrast bolus through some area of interest. The weight of each image is governed by the magnitude of the contrast bolus in that image. An essential requirement of the matched filter is that its integral be zero. It is shown for equal exposure rates and typical bolus characteristics that matched filtering provides a factor of two higher signal-to-noise ratio (SNR) than conventional methods for bolus transit times of 10 s or higher. Equilvalently, matched filtering can yield images with quality comparable to conventional digital subtraction angiography (DSA) at a factor of four less patient exposure. The SNR obtained with matched filtering is shown to be within 30% of an ideal bound. Comparisons of matched filtering to standard recursive methods and simple integration are made. Experimental canine studies are presented which compare matched filtering with conventional DSA.

Some useful concepts of matched filtering in intravenous digital subtraction angiography

Using computer calculations and assumed contrast bolus curves, several aspects of the application of temporal integration methods and matched filtering to intravenous digital subtraction angiography (IV-DSA) were studied. The topics included the improvement in signal-to-noise ratio (SNR) of the final image provided by simple integration, a comparison of the SNR performance of matched filtering and extensive integration, the degradation of SNR caused by the motion of noniodinated objects and the sensitivity of SNR to variations in DSA bolus dynamics from patient to patient. Additionally the dependence of matched filter SNR on exposure position and duration was both estimated and demonstrated with clinical DSA images. The results indicate that a substantial improvement in SNR can be obtained with only moderate integration increasing to a two X improvement for longer durations. Integration methods are able to withstand moderate durations (2 seconds) of motion and still provide image quality superior to more conventional DSA results.

WE‐G‐217A‐06: Spatial Accuracy Quantification of an MR System

PURPOSE To develop a phantom and measurement protocol for quantifying spatial accuracy of an MR imaging system over its entire imaging volume. METHODS The measurement protocol is comprised of a phantom, a set of MR sequence parameters for imaging the phantom, and analysis software for calculating spatial errors in the acquired phantom images. The phantom covers the entire imaging volume of the scanner above the patient table. It consists of layers of tooling foam which does not produce any detectable signal on conventional MR images, embedded with a matrix of oil capsules to serve as markers. To account for possible spatial errors in the construction of the phantom, the phantom was imaged with CT to create a gold standard data set. On MR scanners, the phantom is acquired with a 3D FGRE sequence that covers an extended FOV of 61.44 mm and with bandwidth = ±62.5 kHz. Error measurements are performed by detecting markers in the image sets and identifying them based on their known locations on the phantom. The spatial error of a marker is defined as the difference between its locations on the MR and CT image sets. RESULTS The phantom was constructed and the measurement protocol was executed on two different MR scanners. Some markers were located in areas of severe field inhomogeneity or gradient nonlinearity, and could not be adequately detected for analysis. Maximum errors over concentric spherical regions were observed by plotting the error of each marker as a function of their distance from isocenter. CONCLUSION The proposed phantom and protocol can be an effective tool for verifying the spatial accuracy of an MR system, which in turn can improve the accuracy and confidence of MR guided therapies. Data from this protocol may also be used in the development of advanced distortion correction algorithms. Employed by General Electric Healthcare.

WE-G-217A-07: Increased MR Spatial Accuracy with Improved Gradient Nonlinearity and Magnet Inhomogeneity Correction

PURPOSE To develop improved distortion correction of MR images based on higher degree spherical harmonic models of the gradient system and the main magnetic field. METHODS The induced field gradient along all three axes can be modeled by first order spherical harmonics. These models provide a true encoding of the physical location of a spin to the frequency at which it is detected. Currently on many commercial systems, only the lower 5 degrees of the model are used for gradient nonlinearity correction. While this provides sufficient accuracy for diagnostic imaging, the gradient nonlinearity correction was extended to include all first order harmonics up to the 9th degree to improve the spatial accuracy of the images. Using zeroth degree spherical harmonics up to the 20th order, a model of the main magnetic field was also incorporated into the correction algorithm. Shifts caused by field inhomogeneity were calculated using knowledge of the receiver bandwidth, frequency encode direction, and the magnetic field at any given point. These corrections were applied to images of a 50 cm diameter phantom, acquired with an extended FOV 3D FGRE sequence. Any improvements in spatial accuracy were measured in the resulting images. RESULTS Visual improvements in spatial accuracy were observed with both corrections. With standard gradient nonlinearity correction, edges of the phantom were distorted in a wave-like fashion. With accurate models, almost all of the errors at the edges of the phantom were corrected when both gradient and field homogeneity corrections were applied. CONCLUSION With accurate models of the gradient and magnetic field, the two greatest system-induced spatial errors can be corrected. This improved spatial accuracy enables the use of widebore MR scanners for therapy planning on large FOV images and guidance of percutaneous devices. Further applications include extended FOV imaging for combined PET-MR systems. All authors are employed by General Electric Healthcare.

Echo Planar Imaging For Diffusion / Perfusion

Conventional MRI sequences acquiring only one line of spatial frequency data per RF excitation or repetition interval (TR) can be additionally fitted with diffusionlperfusion sensitizing gradient pulses which alter pixel intensities. Ideally, pixel values from images created at different gradient levels can be used to calculate local diffusionlperfusion maps of tissue. However, conventional imaging methods, with relatively long scan times, often allow bulk motion, flow and system instability artifacts (at or below levels tolerable for normal imaging) to corrupt the diffusionlperfusion sensitized data. This imposes constraints on patient motion which are often uncomfortable, difficult or impossible to achieve. Methods such as Echo Planar Imaging (EPI) that acquire reconstructable spatial frequency data within one or a few RF excitations are becoming methods of choice for motion artifact reduced imaging. An overview of the differences between conventional imaging and EPI will be discussed, emphasizing areas relating to pulse sequences, magnetlgradient hardware requirements and reconstruction corrections.