Medical Imaging : MRI

Medical Imaging Portfolios


Method for Data-Consistency Preparation and Improved Image Reconstruction

Recognizing the link between data consistency and artifact mitigation, UW–Madison professor Guang-Hong Chen has developed an improved processing method that is applicable across modalities, including CT imaging, PET, SPECT and MRI.

The new method provides a practical means to define a data inconsistency metric (DIM) that can be used to locally characterize the inconsistency level of each acquired datum or a view of acquired data. The DIM can be used in a data classification technique to select an optimal data set with a minimal data inconsistency level to reconstruct images with minimal artifact contamination.

The acquired datasets can be classified into one or more subsets based upon the value of DIM. After data classification, a reconstruction technique, such as the SMART-RECON algorithm pioneered by Prof. Chen, can be used to reconstruct these sub-images. Each sub-image is consistent with the subset of the projection view angles for a given range of DIM values. The result is substantially improved images.

Fat- and Iron-Corrected T1 Mapping for Diagnosing Liver Disease

UW–Madison researchers have developed a confounder-corrected MR method for evaluating liver fat content. In the new method, multiple datasets with different Tl weighting are generated using variable repetition time (TR; sequential or interleaved) and/or multiple flip angles. By acquiring multiple datasets with differential Tl weighting, as well as multiple echoes, Tl maps can be generated that are both fat- and iron-corrected.

This approach can be performed as a single acquisition and therefore the underlying source data are all inherently co-registered with one another. That is, simultaneously produced and co-registered estimations of Tl (water and fat signals), R2* and proton density fat fraction (PDFF) are now clinically available for the first time. This allows for the ready creation of fat- and iron-corrected Tl and R2* maps.

The method yields other clinically valuable information, such as estimating tissue fat or tissue water concentration in the form of PDFF or proton density water fraction (PDWF). PDFF is a well validated biomarker of liver fat content and PDWF is a biomarker of breast density, which is known to confer increased risk of future breast cancer.

Long-Lived Gadolinium-Based Agents for Tumor Imaging and Therapy

UW–Madison researchers have synthesized the first long-lived tumor-specific contrast agents for general broad spectrum tumor imaging and characterization. The new, gadolinium (Gd)-labeled analogs utilize an alkylphosphocholine carrier backbone. Their formulation properties render them suitable for injection while retaining tumor selectivity.

High Accuracy Angle Measuring Device for Industrial, Medical, Scientific or Recreational Use

A UW-Stout researcher has developed a high-accuracy angle measurement system that addresses the problems inherent to commercially available systems. In this novel device, a high accuracy rotary optical encoder is controlled by a microprocessor. The encoder consists of rotating optical disks and sensors that are precisely formed and placed to read angles with 0.001 arc second sensitivity (average) and better than ±0.1 arc second accuracy (single readings), which is comparable to the accuracy of the high-end commercial encoders currently on the market. This accuracy is maintained with strategies that combat the mechanical sources of error that are known disadvantages of other devices. The system can also be adjusted to compensate for any asymmetrical shifts that may occur. Mechanical sources of error and noise are further minimized by precision placement of disks and sensors, as well as low-friction reference points that keep components centered and level during rotation. In addition, multiple sensor heads eliminate major readout errors and remove the need for recalibration. All of these features and benefits are contained within a design that is both compact and portable. Beyond high accuracy and portability, the cost of this new angle measurement system is substantially lower than a high-end commercial system because it is easily constructed from readily available industrial grade components, bringing the production cost to roughly $2,000. Strikingly, this cost is comparable to the advertised price of other rotary position encoders that are less than one tenth as accurate. Its high accuracy, low cost, and portability make this new angle measurement system a strong option for use in virtually any of the current applications for absolute rotary encoders.

Robust Chemical Shift MRI Using Magnetization Transfer

A UW–Madison researcher has developed a method that significantly reduces fat-water separation errors using a fat-insensitive field map for calibration. The field map is generated by exploiting the magnetization transfer effect and its lack of influence on fat.

The new method acquires a static magnetic field map (B0) before application of the IDEAL algorithm using a fast prescan with a special radiofrequency pulse and post processing, which reduces separation errors without prolonged or intensive computation.

Rapid MRI Gradient Calibration Using Single-Point Imaging

UW–Madison researchers have developed a dynamic SPI-based method for MRI systems that allows simple, rapid and robust measurement of k-space trajectory.

To enable gradient measurement, they utilized the variable field-of-view (FOV) property of dynamic SPI, which is dependent on gradient shape. In the process, one-dimensional (1-D) dynamic SPI data are acquired from a targeted gradient axis, and then relative FOV scaling factors between 1-D images or k-spaces at varying encoding times are found. These relative scaling factors are the relative k-space position that can be used for image reconstruction.

The gradient measurement technique also can be used to estimate the gradient impulse response function for reproducible gradient estimation as a linear time invariant system.

Improved Phantom for Quantitative Diffusion MRI

UW–Madison researchers have developed a q-dMRI phantom with advantageous properties, including single-peak MR spectrum and Gaussian diffusion propagation. By varying the combined concentration of solvent (e.g., acetone) and solute (e.g., deuterium oxide or diacetyl), the diffusivity of the solution can be controlled to fall within a range of values found in a variety of biological tissues in different physiological conditions and environments.

Under temperature-controlled conditions (for example, submerging the phantom in an ice-water bath) the phantom can reproducibly exhibit ADC values that cover the entire physiological range. Furthermore, different types of paramagnetic salts may be added into the mixture to control T1 and T2 relaxation of the phantom.

Improved Phantom for Iron and Fat Quantification MRI

UW–Madison researchers have designed a phantom that accurately reflects in vivo MRI signal behavior in the presence of both fat and iron. The key innovation is that the new phantom is constructed using a lipid emulsion substrate with superparamagnetic iron oxide (SPIO) particles that are proportionately larger than the fat particles, such that the field from those particles encompasses the entirety of the fat signals.

Real-Time MRI Guides Surgical Intervention and Limits Human Error

UW–Madison researchers have developed a new method and accompanying software for using MRI to control and guide the placement of interventional devices during surgery. This method provides the clinician with rapid feedback that enables intuitive, real-time device manipulation.

A pivoting guide is arranged around a subject’s anatomy. Then rapidly acquired radial image data is used to measure two or more marker positions along the guide and calculate the desired trajectory for the interventional device. The device may be placed using a fully automated system, or an audio/visual signal may help the clinician adjust the guide and correctly place the device. In addition, image data may be used to measure two or more marker positions along the interventional device and calculate its location/orientation.

Controlling Motion Effects in MRI

UW–Madison researchers have developed a method for overcoming motion effects in MRI images. The new method makes dynamic contrast enhanced imaging less susceptible to a patient’s respiratory movement.

In essence, a sliding slice acquisition strategy is used to sample k-space in a pseudorandom manner relative to the trajectories extending between the center and peripheral areas of k-space. A two-dimensional (2-D) slice may be slid from one position to another faster than the patient is breathing/moving. This allows motion artifacts to be reflected as geometric distortions that do not detract from the clinical utility of the images.

Faster, Distortion-Free MRI Near Metallic Implants

Improving upon their earlier work, UW–Madison researchers have developed a method to accelerate MRI scans performed near metal. The new method can work with existing techniques such as MAVRIC.

The new method efficiently measures coil sensitivities across a broad off-resonance spectrum, enabling the use of externally calibrated PMRI techniques. The method saves significant time by eliminating the need to obtain fully sampled calibration regions for all of the acquisitions at different resonance frequency offsets.

Time-Resolved 3-D Angiography Captures Blood Flow, Vessel Dynamics

UW–Madison researchers have developed a system and method for integrating 4-D DSA with physiological information (blood flow, velocity) derived via MRI or ultrasound. More specifically, the image processing system receives angiographic data and flow data to generate a combined data set. The resulting images display time-resolved, color-coded flow information.

The new process can be referred to as 7-D DSA.

Faster, Higher Quality Medical Imaging

UW–Madison researchers have developed a reconstruction technique that uses a non-patient-specific signal model (e.g., a physical or physiological model) to improve image quality without compromising accuracy.

While other methods make use of such analytical models in the post-processing stage, the new technique utilizes the model earlier in the process, yielding clinically useful images from highly undersampled data. The reconstruction process is designed to accommodate deviations from the model when appropriate.

Semi-Automated Segmentation Improves Knee MRI

UW–Madison researchers have developed a semi-automated segmentation technique for 3-D MRI. The technique is particularly suitable for articular joint tissues, like knee cartilage, that are tough to delineate.

With the new technique a user selects a few specific points in a medical image. The points generate image intensity profiles along linear projections, which can be used to determine the boundaries of the target tissue. The volume contours of segmented tissues may then be constructed by processing a series of image slices.

Clearer MRI Near Metallic Implants

UW–Madison researchers have developed a new technique for faster, fully phase-encoded 3-D MRI that enables distortion-free imaging near metallic implants.

In the technique, multiple spectral bands associated with different resonance frequency offsets are simultaneously excited using a multiband excitation scheme. The MR signals generated in response to this excitation then are spatially encoded using phase-encoding along three dimensions. In other words, no frequency-encoding gradients are used.

The new technique can be referred to as multiband, fully phase-encoded (MB-FPE) imaging.

Robust Magnetic Field Map Estimation Improves MRI Fat-Water Separation

UW–Madison researchers have developed a method to improve the robustness of chemical species separation in MRI. Their approach uses an object-based initial estimate of the B0 field map.

More specifically, an MRI system scans a subject to acquire k-space data at different echo times and subsequently reconstructs images. The pixel values of these images are used to estimate a distribution of magnetic susceptibility values found in the subject. A magnetic field inhomogeneity map is estimated from the magnetic susceptibility distribution, and chemical species separation (e.g., fat-water separation) then can be performed.

The new approach is intended to improve the robustness of existing techniques for chemical shift encoded chemical species separations.

Single MRI Scan Acquires Multiple Sets of Inversion Recovery Data

UW–Madison researchers have developed a method that expedites inversion recovery by acquiring data after each IR radiofrequency pulse. In this way, both single IR and DIR data can be obtained in a single, condensed scan.

In the method, each IR pulse is followed by an excitation pulse and data acquisition. Any suitable data acquisition scheme can be employed, such as VIPR (vastly undersampled isotropic projection reconstruction). Multiple images of the subject are reconstructed from this data. Data after the first image can produce a traditional T1-weighted image, while data after the second inversion produces a traditional DIR image.

Evaluating Systemic Cancers

UW–Madison researchers have developed a technique for extending molecular and functional imaging (e.g., PET, fMRI) assessment of the total disease and disease heterogeneity to a variety of different cancers, including systemic types throughout the body.

The method uses a combination of anatomical and functional masking to isolate multiple dispersed lesions from surrounding tissue. In this way, automatic identification tools can target likely tissue on a case-by-case basis, as guided by information about the type of cancer and imaging materials.

First, a patient is administered an imaging agent that identifies tumor tissue. After scanning, a program helps identify and measure the progression of multiple tumor locations based on how and where the agent is taken up. A color-coded output shows measurements at different locations.

Combining VIPR with Inversion Recovery for Improved MRI

UW–Madison researchers have now developed an algorithm combining VIPR with IR. This combination, along with a unique projection ordering, results in a large set of 3-D, high spatial resolution images with multiple different image contrasts.

VIPR-IR works by allowing more flexibility in creating segments of repetition times (TRs) that sweep a range of inversion times. The user may select a desired number of consecutive projections to be combined for better image quality. The number of these consecutive projections can be chosen after data acquisition. In other words, data is combined retrospectively.

Rather than trying to predict which imaging parameters will produce the best contrast for a particular patient, the method acquires multiple images across a range of contrast settings. The process does not take longer than a typical scan.

High-Resolution R2 Mapping with Chemical Species Separation

UW–Madison researchers have developed a method for producing a quantitative map of R2* while separating signal contributions from two or more chemical species, like fat and water.

The method works by producing quantitative R2* maps, quantitative fat fraction maps and separate R2*-corrected water and fat images. A low-resolution field map and a common water-fat phase are used to demodulate the effects of these parameters from the acquired data while separating the water and fat signals.

In this way, water, fat and R2* can be estimated simultaneously. A full resolution R2* map is reconstructed in addition to water, fat and fat fraction images that are corrected for the effects of R2*.

Point Sets for Higher-Quality MRI

A UW–Madison researcher has developed a new approach for generating uniformly distributed and antipodally symmetric point sets on a sphere. The point sets are useful for defining certain MRI acquisition parameters, specifically, diffusion-weighting directions and 3-D radial k-space trajectories. The point sets are efficiently computed using constrained centroidal Voronoi tessellation.

Confidence Maps for MRI Parametric Mapping

UW–Madison researchers have developed a method for reducing parametric mapping errors using ‘confidence maps’ that identify problematic areas.

In this approach, an MRI system acquires k-space data from a field of view and reconstructs an image series. The k-space data also is used to compute a confidence map depicting regions in the field of view affected by error sources. A parametric map is produced using the MRI image series but checked against the confidence map. Error-prone areas are avoided, removed or flagged so as not to contaminate the parametric map.

Medical Imaging with Better Temporal Fidelity Can Streamline Stroke Care

UW–Madison researchers have developed a method that increases temporal fidelity, sampling density and/or reduces noise of image frames obtained with a system such as CT, MRI or X-ray c-arm. After the images are acquired, a window function is selected and temporally deconvolves the image frames using a minimization technique. A temporal sampling density also may be selected and used in the temporal deconvolution.

Probing Disease Chemistry with Joint Spatial and Spectral Imaging

UW–Madison researchers have developed a method for simultaneously generating spectral and spatial images of a subject using an MRI system.

A subject receives a dose of hyperpolarized imaging compound. MR image data is acquired from the subject according to a k-space sampling trajectory that spatially oversamples to encode both spatial and spectral frequency information at the oversampled points. The MR image data then can be reconstructed into the different image types using a model-based reconstruction technique and prior knowledge of the chemical species associated with the compound.

Wirelessly Tracking Interventional Medical Device with MRI System

UW–Madison researchers have developed a method for driving and amplifying MR signals using an imaging coil that is coupled to a medical device and inserted into a subject. The position-tracking coil uses the MR system’s external radiofrequency pulses as a power source.

The coil switches between two duties: receiving and storing energy, and acquiring imaging data. Its signals can be wirelessly transmitted to the external MRI system using an amplifier and booster stages.

Detecting Iron Overload with MRI

UW–Madison researchers have developed a method for measuring iron and other substances in tissue using an MRI system, based on estimating tissue magnetic susceptibility.

The method acquires chemical-shift-encoded, water-fat separated data from a scanned region of interest. From this data a magnetic field inhomogeneity map of the system can be obtained. The field map enables estimation of the magnetic susceptibility of tissue to determine concentration of iron or other substances, such as gadolinium.

Better MRI Performance with Improved 3-D UTE Imaging

A UW–Madison researcher has developed a method for three-dimensional UTE imaging to maximize k-space coverage. The approach combines radial- and spiral-based techniques.

In the method, an MRI system establishes a radial magnetic field gradient that increases, then decreases in amplitude over time. While increasing, the sampling trajectory extends outward. While decreasing, an oscillating magnetic field gradient is established. The two field gradients together form a sampling trajectory that spirals outward in a cone.

Single-Shot Vascular MR Imaging Without Contrast Agent

UW–Madison researchers have developed a non-contrast-enhanced MR angiography and venography (MRAV) method that can acquire arterial-specific signals in one imaging slice and venous-specific signals in another slice during a single cardiac cycle.

In the MRAV approach, radio frequency (RF) saturation pulses may be applied to one or more slabs to selectively suppress MR signals flowing into a selected imaging slice. In this way, the pulses may be selected and timed to suppress venous blood signals in an arterial imaging slice, or to suppress arterial blood signals in a venous imaging slice.

The RF saturation pulses and single-shot acquisitions may be timed to occur during substantially steady state inflow into the respective imaging slice.

Accelerated MRI Scanning Using Spectral Sensitivity

UW–Madison researchers have developed an MRI method to accelerate data acquisitions in the presence of severe off-resonances, such as those induced by metallic objects.

After an imaging machine acquires k-space data, the method is used to derive spectral sensitivity information. This information—in the form of spectral images or sensitivity maps—relates specific resonance frequencies to distinct spatial locations in the magnetic field. This can be done because the off-resonance produced by metal implants and other foreign objects depends on factors like the size, shape and position of the object. The data can be spatially encoded to reconstruct an image.

Faster, Better Quality Medical Imaging by Constrained Reconstruction

UW–Madison researchers have developed a modified algorithm for medical image reconstruction that increases reconstruction speed, improves image quality and provides more accurate results. The algorithm constrains images to be consistent with a signal model, which relates image intensity values to free and control parameters such as relaxation time and multiple echo or inversion times, respectively.

The signal model may be analytical or approximate—learned from acquired image data, as is done in the case of time-resolved MRI. The model consistency condition may be enforced using an operator that projects an image estimate onto the space of all functions satisfying the signal model.

Eliminating Encoding Distortion in MRI for Clarity in the Presence of Metal

UW–Madison researchers have developed a new pulse sequence approach for performing spectrally-resolved, 3-D MRI without using a frequency encoding gradient during the scan process. This allows for spectral encoding of signals such that local magnetic field differences—like those around metal substances—can be measured. Moreover, signal separation can be performed to distinguish tissue types and the relaxation rate of transverse magnetization, R2*, can be measured.

The technology comprises a magnet configured to generate a polarizing field around a subject, gradient coils and a radio frequency (RF) system applying and receiving signals. A computer first directs the RF system to produce a pulse that rotates net magnetization about an axis, and the coils establish three phase-encoding gradients along respective perpendicular directions.

Data are acquired as defined by the three gradients by sampling magnetic resonance signals during multiple time points in which no field gradients are established by the MRI system. This process eliminates any artifacts due to frequency encoding and enables accurate, spectroscopic imaging with higher spatial resolution.

Correcting for Patient Motion with T1-Weighted PROPELLER MRI

Researchers have developed a technique that enables motion-corrected images to be acquired with T1 FLAIR contrast by combining PROPELLER and parallel imaging with calibration data shared between blades.

Using data from a shared external calibration blade reduces the number of internal calibration lines needed and enables higher acceleration for each blade. Maintaining short ETL readouts, the high parallel imaging acceleration increases the effective blade width significantly to allow for motion correction.

Quantifying Visceral Fat Using MRI

UW–Madison researchers have developed a new ‘chemical-shift’ imaging technique that distinguishes visceral adipose tissue, measuring the ratio of that tissue to total fat (VTR) of the abdomen and pelvic region, with a single 26-second high-resolution acquisition requiring less than 15 minutes of processing.

By analyzing the difference in resonance frequency between fat, water and fat-fraction, the method rapidly separates the targeted tissue from other material, automatically excluding from measurement air cavities and background noise.

The fat-concentration map that is produced from signal analysis provides an adipose ‘mask,’ or refined representation of all fat tissues, from which a quantitative metric of adipose volume can be determined. An adipose threshold value allows each image pixel to be counted as VAT or non-VAT, calculating the patient’s adipose distribution as a ratio over total fat content.

MRI Water-Fat Separation with Full Dynamic Range Using In-Phase Images

UW–Madison researchers have developed an image reconstruction method for accurate water-fat separation that fits in-phase echo signals to a signal model that characterizes the fat spectrum as having multiple resonance peaks. Signal contributions of water and fat are separated by fitting only those echo signals having water and the main fat spectral peak in-phase with each other to a signal model of the fat spectrum. An image can be produced with the desired amount of signal contributions from water and fat by using the separated signal contributions.

This method simplifies the separation problem by separating the combined water and main fat peak from the secondary fat peaks, the latter of which make up less than thirty percent of the signal. Using known relationships between the remaining water peak and main fat peak, the signal contributions from these two peaks can be reliably separated for ambiguity-free fat quantification.

Ultra-High Frame Rate, Time-Resolved, 4-D MRA

A UW–Madison researcher has developed a system and method for producing time-resolved 3-D medical images of a subject from a time series of 2-D data sets and a time-independent 3-D volume of the subject.

The 2-D time series of images is obtained using MRI, and then combined with the time-independent 3-D volume to generate a set of time-dependent 3-D volume images of the subject at the frame rate of the acquired 2-D data sets. The system is able to reconstruct multiple sets of 2-D time series of images at different view angles. Reconstruction can be done using HYPR, a processing technique previously developed by the researcher.

Improved MRI Scan Time through Rotating Angle Velocity Encoding

UW–Madison researchers have developed a system and method for performing PC MRI with a substantially reduced acquisition time. Rather than performing separate acquisitions for each velocity encoding, the method allows multiple velocity encodings to be combined. A reconstruction method reconciles inconsistency in the resulting set of Fourier slices by determining the velocity components despite the combination of multiple velocity encodings per readout.

A set of reference projections of a subject is acquired having both stationary spins and non-stationary spins. A set of velocity-sensitive projections is acquired that is encoded to be velocity sensitive along multiple directions per readout. For each projection of the set of velocity-sensitive projections, directional velocity components are determined and a PC image is generated using the directional velocity components and the sets of reference projections and velocity-sensitive projections. Thus, to measure the velocity in two directions, only one set of velocity-sensitive projections is needed, whereas in traditional PC MRA, two are needed.

Deterministic Approach to Generating Optimal Ordering of MRI Measurements

UW–Madison researchers have developed a system and method for generating and ordering a highly uniform point set that defines acquisition parameters for MRI. The ordering of the points is optimized for the particular imaging task at hand. For example, the point set is generated to be antipodally symmetric for diffusion-weighted imaging applications.

The medical imaging system contains a processor configured to generate a point set that defines MRI acquisition parameters and includes points that are uniformly distributed on the surface of a sphere. The processor also determines an order in which the points in the point set are to be temporally arranged by minimizing electrostatic energy potentials. The processor communicates the ordered point set with an MRI system and acquires MRI data in accordance with the MRI acquisition parameters. In the case of k-space sampling, the MRI system is directed to acquire k-space data from a subject using the determined MRI acquisition parameters and the order in which the points in the point set are to be temporally arranged. From the acquired k-space data, an image of the subject is reconstructed. In the case of diffusion imaging, the MRI system is directed to apply the diffusion gradients using the determined MRI acquisition parameters and point ordering. Diffusion parameters can be estimated from the acquired diffusion-weighted images.

Method for Error-Compensated Chemical Species Signal Separation with Magnetic Resonance Imaging

A UW–Madison researcher has developed a method for separating the relative signal contributions of multiple chemical species in which echo signal information containing errors is discarded during signal separation. The method enables production of an image with an MRI system in which relative signal contributions from the chemical species are separated while accounting for errors. It requires using multiple echo signals acquired at different times to form signal models that account for relative signal components for each chemical species. Then, each echo signal that contains errors is identified and discarded from the relative signal components for each chemical species. Finally, an image is produced using the reliable data from the relative signal components of the chemical species.

Method for Quantification of R2 Relaxivity in Magnetic Resonance Imaging

UW–Madison researchers have developed a method for measuring R2* with MRI in which signal decays that occur as a result of macroscopic variations in the main magnetic field of the MRI system are incorporated into a chemical-shift based signal model. The model provides for the mitigation of errors due to macroscopic field variations and allows better signal-to-noise ratio performance compared to existing R2* measurements.

The method samples echo signals occurring at different echo times to acquire MRI image data. For each of the echo signals, a signal model is formed to account for relative signal components for each different chemical species, such as water and fat. Magnetic field inhomogeneity values associated with the MRI system are estimated by fitting the acquired image data to the signal models. This allows the creation of signal models that account for relative signal components for each different chemical species and signal decay resulting from macroscopic variations in the main magnetic field of the MRI system. The method also allows estimation of R2* for at least one of the chemical species by fitting the acquired image data to the signal models.

Improved Images with MRI Acquisition of Multiple Chemical Species

UW–Madison researchers have developed a method for producing a high-resolution image of a subject with a magnetic resonance imaging (MRI) system where the image depicts signal contributions from only one chemical species. A unique set of radial lines is acquired at a sequence of multiple echo times occurring within two or more repetition times (TRs). Odd-numbered echoes are sampled during odd-numbered TRs, and even-numbered echoes are sampled during even-numbered TRs. Images are reconstructed and used to calculate the respective signal contributions of two or more chemical species using a species separation technique such as IDEAL. The signal contributions then are used to produce images that primarily depict only one of the chemical species, such that it is possible to produce separated water and fat images.

This imaging method provides superior separation of water and fat signals while allowing the acquisition of high-resolution image data sets. Additionally, the method provides for effective water-fat separation despite sampling a unique set of radial lines at each echo time.

Method to Reconstruct Motion-Compensated Magnetic Resonance Images with Non-Cartesian Trajectories

UW–Madison researchers have developed a motion-compensated image depiction method for use with magnetic resonance imaging systems. An MRI system is used to acquire a time series of k-space data from a subject by sampling k-space along non-Cartesian trajectories, such as radial or spiral, at a plurality of time frames. The time frames at which motion occurred are identified and used to segment the time series into a plurality of k-space data subsets containing consistent data. 

The k-space data subsets contain k-space data acquired at temporally adjacent time frames that occur between those identified time frames at which motion occurred. Transformation matrices are derived from co-registration of images created from these consistent subsets. Motion correction parameters are determined from the k-space data subsets. The determined parameters are applied to the individual consistent subsets of k-space data, and these corrected data subsets are combined to form a corrected k-space data set from which a motion-compensated image is reconstructed.

Magnetic Resonance Dynamic Imaging Sequence for Accelerated Pseudo-Random Data Magnetic Resonance Imaging

UW–Madison researchers have developed a system and method for improving image data acquisition and processing for time-resolved MRI. The method includes an acquisition sequence configured to acquire an undersampled set of magnetic resonance data. The undersampled data set has a pseudo-random sampling pattern within a data space, which is influenced by other pseudo-random sampling patterns within the data space arising from the acquisition of additional undersampled sets of magnetic resonance data over time.

In some embodiments of the proposed method, the pseudo-random sampling patterns of the data sets interleave to yield a desired sampling pattern. Each sampling location of the desired sampling pattern is sampled at least once, and the sampling locations towards the center of the data space are sampled with greater frequency than those further from the center of the data space. The data are combined with parallel image reconstruction and/or a Fourier transform, resulting in a high-quality image with improved temporal and spatial resolution.

Accelerated Pseudo-Random Data Magnetic Resonance Imaging

UW–Madison researchers have developed a system and method for improved accelerated magnetic resonance data collection and image synthesis utilizing parallel image reconstruction. A set of magnetic resonance data is acquired over a time frame of interest with a non-uniform sampling pattern on a predefined area of data space. Parallel image reconstruction is used to generate an initial image from the magnetic resonance data. The predefined area of data space comprises a uniform pattern of locations that are to be sampled and locations that are not to be sampled. The non-uniform sampling pattern comprises a pseudo-random sampling pattern that samples locations toward the center of the data space with greater frequency than those further from the center. Parallel imaging reconstruction is performed on a coil-by-coil basis to suppress aliasing artifacts and is calibrated specifically for the uniform sampling pattern. Back projection reconstruction may be used to generate a reconstructed image for the time frame of interest if additional acquisition sequences at respective time frames are used to acquire additional sets of magnetic resonance data.

“DR-PICCS” – Dose Reduction Using PICCS Image Reconstruction Algorithms

UW-Madison researchers have developed a method using existing “PICCS” (prior image constrained compressed sensing) image reconstruction algorithms to reduce radiation dose while attaining quality images with high SNR. Multiple slices of an image volume are collected and then averaged together to create a single thick slice, known as the “prior image,” with high SNR but lacking detailed anatomical structures. The PICCS algorithm then is used to reconstruct each image slice with the original slice thickness using the prior image. The resulting final image has the equivalent image noise variance level of the prior image, the high spatial resolution of the acquired image is preserved and the anatomical features will be detailed.

Noise variance in the final reconstructed image is improved by a factor of approximately the number of slices included in the prior image. Using this method, the patient receives a reduced dose of radiation while the radiologist acquires final images of the same quality currently attained with higher radiation levels.

Water-Fat Signal Separation for Improved MRI Image Reconstruction

UW-Madison researchers have developed an algorithm for separating the nuclear magnetic resonance (NMR) signal contributions from at least two different species such as fat and water to improve image reconstruction methods. The process uses a field map constructed with an algorithm as the initial input. This initial map is used to generate weighting maps that indicate the likelihood of a water-fat swap at each pixel based on spectral differences and the local field map gradient at each pixel. Then the field map values for select pixels are recalculated based on the weighting maps. The process is repeated iteratively until the weighting maps indicate that no pixels should be recalculated.  Images reconstructed using these algorithms are of substantially better quality with minimal water-fat swaps.

Method for Improved Fat-Water Signal Separation in Phase Contrast Magnetic Resonance Imaging

UW-Madison researchers have developed an improved method of PC MRI that allows for accurate fat and water signal separation. The method simultaneously performs several measurements of echo-time and velocity encoding so that data is acquired from both an on-resonance water signal and an off-resonance fat signal. A set of images is reconstructed by estimating a common magnitude image and a plurality of phase images from the acquired phase contrast image data. Accurate flow velocity quantification is enabled by accurately separating the fat signal from the water images. The resulting image is created with minimal noise and improved quality.

Cardiac Image Reconstruction with Improved Temporal Resolution

UW–Madison researchers now have developed a ‘prior image’ method for reconstructing dynamic and undersampled data. The method is applicable to a number of different modalities including CT, X-ray C-arm imaging, MDCT, magnetic resonance imaging (MRI), positron emission tomography (PET) and single photon emission CT (SPECT).

Specified for each system, in general the method combines image data from current and prior time frames, like heartbeat phases. A limited amount of additional image data is incorporated into the consistency condition imposed during prior-image constrained-image reconstruction.

Magnetic Resonance Imaging of Diffusion and T2 Using Multi-Echo Projection Acquisition

UW-Madison researchers have developed a method for simultaneously imaging the ADC and T2* in a single breath-hold with only one hyperpolarized gas contrast agent dose in the lungs or other body airspace.  The method uses a multi-echo projection acquisition based pulse sequence that varies the inter-echo spacing and the diffusion weighting to successfully separate the effects of diffusion and T2* decay in the MR signals.  This separation improves the reliability of the measurements for ADC and T2* to allow for single breath-holds to be used.  The use of one breath-hold also reduces the required contrast agent dose to just one.

Post-Processing MRI Fat Suppression Method to Enhance Image Quality and Improve Medical Diagnostics

UW–Madison researchers have developed a post-processing technique to improve fat and water suppression in images reconstructed from VIPR-SSFP data acquired by the previous method. The improved processing method, termed Dual Acquisition Phase Difference SSFP, acquires two echo signals which are combined in a second cancellation step after initial VIPR-SSFP data reconstruction. First, the phase of one echo is shifted and combined with the second echo in a process that transforms the radial data to Cartesian coordinates as in the previously developed method, which is known as Linear Combination SSFP. The second step involves application of a phase mask, derived from the echoes’ phase difference, to the reconstructed image. The additional phase difference mapping provides fat signal cancellation across a wide range of off-resonance frequencies centered about the fat resonance peak.

The improved dual acquisition technique allows for reconstruction of fat or water suppressed images in shorter scan time, at higher resolution or at higher signal-to-noise ratio. Utilizing the SSFP method to reduce scan times will increase patient comfort and throughput as well as minimize motion artifacts. The improved resolution and signal-to-noise ratio also enhance the quality of MR images, which will facilitate the applicability of this technology in standard medical diagnostics, especially breast exams.

Real-Time Progressive Medical Image Reconstruction Method for Time-Resolved Data

UW–Madison researchers have developed a method for medical image reconstruction that delivers quality images from time-resolved image data in real time. The method provides accurate images with increased signal-to-noise ratio and temporal resolution while minimizing a patient’s exposure to X-ray radiation.

The method starts with two related images that can be neighboring images from a time series image set or images from the same phase point during repeated motion, such as breathing or the beating of the heart. The images are subtracted to get an image of the difference between the two, which undergoes a sparsifying transformation to reconstruct the final image of interest. All of this is done in real time to achieve more dynamic medical imaging, such as image guided interventional procedures.

High Temporal Resolution Cardiac CT Imaging with Slowed Gantry Speed

UW-Madison researchers have developed an application of the PICCS image reconstruction method (see WARF reference number P08127US) for producing a time series of images with a higher temporal resolution than the temporal resolution at which the image data was acquired. In cardiac imaging, this allows use of slow gantry rotation for improved image resolution instead of continuing to increase the speed of gantry rotation, which is mechanically challenging.

For high temporal resolution cardiac imaging, a “cone-beam” arrangement such that the focal spot of the X-ray source and the detector define a cone-shaped beam of X-rays is used. The gantry rotation time is counter-intuitively slowed to about 10 seconds. This rotation time enables a single breath-hold for most cardiac patients, which reduces motion during imaging. During the cone-beam CT data acquisition, the ECG-signal will be recorded as 400 to 600 views of the cone-beam projection are simultaneously acquired during each gantry rotation. The acquired data will be used to reconstruct a “prior” image containing the heart that does not contain dynamic information and possibly contains motion-induced streaks. Next, the acquired projection data are “gated” using the ECG data so that there is one projection per heart beat and images can be reconstructed using data from each gated “window”. This allows the PICCS algorithm to accurately reconstruct each cardiac phase. The resulting images have an ultra-high temporal resolution about 20 times better than images obtained using state-of-the-art CT scanners with increased gantry speeds.

Prior Image Constrained Compressed Sensing (PICCS)

UW-Madison researchers have developed a method for reconstructing a high quality image from undersampled image data that is applicable to a number of imaging modalities including CT, MRI and positron emission tomography (PET).

A seed image is acquired by using a prescan, reconstructing a high signal-to-noise (SNR) ratio image from data acquired through any modality, or from a fully sampled data set. This prior seed image is used to iteratively reconstruct a final output image from an undersampled data set taken of the same anatomical structure as the seed image. The high SNR seed image guides a mathematical manipulation of the data set, resulting in a high quality image constrained by the original high SNR image. The method typically requires only two to five iterations to achieve clinically useful images, resulting in a convergence speed much faster than any known iterative image reconstruction methods.

Non-Invasive Magnetic Resonance Thermometry in the Presence of Water and Fat

UW-Madison researchers have developed a method to use images obtained with water-fat signal separation techniques to non-invasively measure temperature change in tissue. The temperature in tissue containing both water and fat can be obtained by calibrating the image reconstruction using the relative frequency between water and fat as an internal reference, avoiding the challenges of reference images in MRI thermometry. The IDEAL water-fat image separation technique (see WARF reference number P090389US01) also could be used to avoid corruption of MRI temperature measurement by removing the fat signal and maintaining only the dependent water signal. These methods allow the generation of a complete temperature map of tissues that contain both fat and water.

More Accurate Methods for Detecting and Quantifying Fat from Magnetic Resonance Images

UW-Madison researchers have developed two methods for improving the accuracy of fat measurements from MR images.  These methods remove the effects of noise during the quantification process.  Large amounts of noise leads to an over- or underestimate of the amount of fat.  The removal of this noise results in a more accurate fat image.

The two methods are known as the phase constrained and signal normalization methods.  The phase constrained method assumes water and fat to have a common, equal phase.  This method calculates the fat fraction directly from real water and fat components, as opposed to complex signal components, to remove noise bias.  The signal normalization method takes an image signal when the water and fat are in phase.  Two or more other image signals are obtained and normalized using the previous in phase image signal to remove any noise bias.

Improved MRI with Radio Frequency Coil Decoupling Circuit to Enhance Image Quality

UW-Madison researchers have developed an improved method and apparatus for a decoupling circuit for individual RF coil elements in a MRI system.  The decoupling circuit includes a power amplifier and matching networks that provide control of current amplitudes.  In the new device, eight individual RF coil segments, or rungs, have an input matching network, an independent, controllable current source and an output load line matching network.  The matching networks and associated components combine to minimize input impedance and maximize output impedance of currents produced by other rungs.  A transmission line transformer also is used to dampen the quality factor of the output matching network and improve system bandwidth.  The improved method for decoupling segments in a RF coil will allow each coil segment to be accurately driven without interference from currents induced by neighboring coil segments.

The new method and apparatus may be used for both whole-body and local MRI systems to produce a uniform magnetic field B1 and improve the quality of MR images.  By improving the quality of MR images, MRI will become a more accurate and useful tool in medical diagnostics and other applications like material science.

Magnetic Resonance Imaging Diffusion Weighted Preparatory Sequence to Remove Patient Motion Effects

UW–Madison researchers have developed an improved preparatory DWI technique that removes the effect of patient motion.  Specifically, the method uses a MR pulse sequence that provides diffusion-weighting to NMR signals without sensitivity to patient motion.  An initial pulse sequence is performed prior to the imaging pulse sequence that diffusion weights the spin magnetization moment of the water molecules.  The initial pulse sequence uses a gradient waveform that removes all the distorting phase shifts resulting from velocity motion of the tissue.  However, the phase shifts due to higher order motion such as accelerations and other motions characteristic of diffusing spins are not removed.  Therefore, DWI can be used at the area of interest without the effects of patient motion.

Synthesis of Magnetic Resonance Imaging Contrast Agents Using Ring-Opening Metathesis Polymerization

UW-Madison researchers have developed sensitive and versatile polymer-based contrast agents for MRI. The contrast agents were synthesized using ring opening metathesis polymerization (ROMP), a powerful synthesis method for creating biologically-active polymers. During this process, HOPO-based chelating moieties were integrated into benzonorbornadiene backbone units to create a polymer capable of binding gadolinium at multiple sites.

Voltage Standing Wave Suppression Safety Improvement for MR-Guided Therapeutic Interventions

UW-Madison researchers have developed a medical device for MR-guided therapeutic interventions that uses optimized cable traps to restrict dangerous heat buildup while also allowing the device to retain its flexibility.  Instead of wrapping the cable along the entire length of the device, small cable traps are set up at short distances along the length.  High impedance in the cable traps reduces the current on the outside of the device, while allowing no disruption in the signals transmitted to the device.  The voltage standing wave then is suppressed and minimizes patient risk.  This optimized cable trap safety feature can be implemented in any interventional MR imaging (MRI) device in which the conductance phenomenon exists.

Radio Frequency (RF) Coil for an Improved MRI System

UW–Madison researchers have developed a TEM resonator coil with drive circuitry that can be used in an MRI system to transmit a uniform magnetic field or receive NMR signals. It differs from previous TEM resonators because the device itself is not a resonance structure and its multiport excitation suppresses unwanted resonant modes.

This device consists of a coil that has a cylindrical shield encircling a central axis, which supports multiple pairs of opposing conductive legs arranged symmetrically around the central axis. Terminal susceptance elements, most commonly capacitors, are connected between the legs and the Faraday shield. Each pair of legs is connected to a current balun, which maintains equal currents in the leg pairs. A tune and match circuit, which matches impedances, is connected between the balun and conductive legs. The coil also can be utilized to operate at multiple Larmor frequencies at once while not creating unwanted excitation modes.

Real Time 3-D Tracking and Imaging System and Method for MR Guided Endovascular Intervention Therapy

UW-Madison researchers have developed a system and method capable of 3-D tracking and imaging in real time for minimally invasive endovascular therapy. The system is comprised of a multi-mode medical device system (see WARF reference number P05330US), which is guided through the patient’s vasculature using its tracking and imaging capabilities to produce high quality, localized images. This device is coupled to an MR scanner, which uses the multiple-echo Vastly Under-sampled Imaging with Projections (VIPR) data acquisition sequence to acquire real-time 3-D images of the area of interest using external coils. External imaging, internal tracking and internal imaging are provided simultaneously by this entire system in real time.

Characterization of Receiver Demodulation for Correcting Off-Axis MR Imaging Degradation

UW-Madison researchers have developed an MRI pre-scan calibration procedure that allows easier demodulation corrections. The procedure measures the hardware timing error to 0.1 ms accuracy between the real-time frequency demodulation hardware and the data acquisition subsystem.

By knowing the timing delay, the actual demodulation phase applied to each raw data point by the hardware can be calculated rapidly. A phase correction can then be applied before the image is reconstructed to account for the phase errors created by the delay. Alternatively, knowledge of the correct delay can be used to adjust the hardware prior to image acquisition; however, this is less accurate as current hardware can only be adjusted in quantized interval of 1 or 2 ms.

Image Reconstruction Method for Motion Encoded Magnetic Resonance Images

UW-Madison researchers have developed a new method for reconstructing motion encoded magnetic resonance (MR) images from undersampled data.  A highly constrained backprojection method is used to reconstruct MR images using a composite image made up of interleaved projection views.

The composite image allows the method to produce good quality images with far less data, reducing scan time for PC applications.  The highly constrained backprojection reconstruction method weights image pixels to increase the image quality at areas in which the composite image pixels intersect structures in the subject, instead of simply assuming the pixels should be weighted evenly.  Increasing the quality of this composite image, by taking a series of undersampled images and interleaving them, directly increases the reconstructed image quality and also maintains the ability to quantitatively measure blood flow.

Image Reconstruction Method for Functional Magnetic Resonance Imaging

UW-Madison researchers have developed a new method for acquiring and reconstructing functional magnetic resonance (MR) images.  A series of views are combined into a composite image that is employed in a highly constrained backprojection method to reconstruct fMRI image with an increased signal-to-noise ratio (SNR) and fewer artifacts.

The composite image allows the method to produce good quality images with far less data, reducing scan time.  The highly constrained backprojection reconstruction method weights image pixels to increase the image quality at areas where the composite image pixels intersect structures in the subject, instead of simply assuming the pixels should be weighted evenly.  Increasing the quality of the composite image by taking a series of undersampled images and interleaving them increases the reconstructed image quality.

Image Reconstruction Method for Cardiac Gated Magnetic Resonance Imaging

A UW-Madison researcher has developed a new method for reconstructing cardiac gated MR images and specifically for improving the quality of highly undersampled cardiac phase images.  A series of views for one image from a specific cardiac phase are combined into a composite image.  Then a highly constrained backprojection method, using the composite image, allows for the reconstruction of 2-D and 3-D images for each phase.  A highly sampled composite image also can be constructed from multiple undersampled images from a certain cardiac phase to increase the signal-to-noise ratio of the final images.

Highly Constrained Image Reconstruction for Magnetic Resonance Spectroscopy

UW-Madison researchers have developed a new method for reconstructing magnetic resonance images using multiple projecting view sets, acquired by pulse sequences, with different MR parameters for each set.  Projection views acquired using varying magnetic resonance parameters are interleaved to form a composite image.  A highly constrained backprojection reconstruction uses the composite image to reconstruct images from each set of projection views.

The image reconstruction method improves the quality of images with far fewer projection views.  By improving the composite image through the use of multiple interleaved projection views, the reconstructed MR images can be improved.  The method also can substantially improve MRS image quality, which directly relates to improved spectra of metabolites in tissue.

Diffusion Tensor Imaging Using Highly Constrained Image Reconstruction Method

UW-Madison researchers have developed a new method for collecting diffusion weighted data and reconstructing DWI images.  A highly constrained backprojection method reconstructs each DWI image using a composite image made up of interleaved projection views.

The method is able to produce good quality images with far less data, reducing overall scan time.  The highly constrained backprojection reconstruction method weights image pixels to increase the image quality at areas in which the composite image pixels intersect structures in the subject.  Increasing the quality of this composite image directly increases the reconstructed image quality.

Image Reconstruction Method for Computed Tomography and Magnetic Resonance Cardiac Imaging

A UW–Madison researcher has developed a new method for reconstructing highly undersampled images at specific cardiac phases for both X-ray computed tomography (CT) and magnetic resonance imaging (MRI). The method uses a highly constrained backprojection method and requires a composite image that is enhanced using the previously proposed method.

The highly constrained backprojection reconstruction method weights image pixels to increase the image quality at areas in which the composite image pixels intersect structures in the subject, instead of simply assuming the pixels should be weighted evenly like previous techniques. Increasing the quality of this composite image directly increases the reconstructed image quality. The composite image can be enhanced further by subtracting the stationary tissue that surrounds the heart.

Removal of Chemical Shift Artifacts in Magnetic Resonance Images with Alternating Readout Gradients

A UW–Madison researcher has developed a method to remove chemical shift artifacts from MR images.  Images are acquired using opposite polarity readout gradients to completely suppress chemical shift artifacts.  The method uses a two-point acquisition to produce two images, one in phase and one out of phase.  The k-space data is acquired from left to right and then from right to left.  Using the chemical shift, the images are realigned to represent either water or fat.  This removes normal chemical shift artifacts, and also is suitable for MR imaging using higher magnetic field strengths with larger chemical shifts and lower bandwidths to improve signal-to-noise ratios.

Highly Constrained Backprojected Reconstruction (HYPR) for Magnetic Resonance Images

A UW-Madison researcher has developed a new method for reconstructing magnetic resonance images using an improved backprojection method.  The method uses a composite image and an assumption of an inhomogeneous backprojected signal to weight the distribution of the backprojected views to reconstruct images.

The composite image can be obtained from either the MRI scan or from previous data to enhance undersampled data sets.  The method can be applied to contrast enhanced magnetic resonance angiography (CEMRA), which would subtract out unwanted tissue in the composite image to further enhance the effectiveness of the present invention.  Multiple composite images can even be utilized to reconstruct changing images such as in time-resolved angiography.

Highly Constrained Image Reconstruction for Medical Imaging Applications

A UW–Madison researcher has developed a new method for reconstructing medical images from projection views of a subject. A backprojection technique is used that does not assume homogeneity in the backprojected signal. A composite image is reconstructed, and then this composite image is used to highly constrain the image reconstruction process to provide more image detail where needed.

This image reconstruction method reduces scan time and radiation dose, and provides higher resolution for time-resolved studies. Acquiring a highly sampled composite image will increases the signal-to-noise ratio (SNR) of the undersampled reconstructed images. This method can be used to improve the reconstruction of medical images.

Improved Lung Disease Diagnostic Using Gas Contrast Agent Diffusion Weighted MRI to Segment the Lumen

UW-Madison researchers have developed an improved method of segmenting the airways to diagnose lung disease.  This method involves segmenting the lumen of the large airways, which consists of the void space in the bronchi.  It uses diffusion weighted MR images after inhalation of a hyperpolarized gas contrasting agent.  As with segmented imaging of blood vessels, observation of varying contrast agent diffusion allows differentiation between tissue volume elements. 

In the new method, a NMR responsive gas such as hyperpolarized helium is used as the contrasting agent in the lung.  Then, diffusion weighted image (DWI) data of the lung is acquired via the MRI system.  From the DWI data an apparent diffusion coefficient (ADC) image can be reconstructed to map how the NMR responsive gas diffuses in the lung.  A segmentation image is produced from the ADC image, allowing the radiologist to differentiate between the void space comprising the airway tree and the lung tissue volume elements.  The segmentation image then is used to segment the final diagnostic resonance image of the airways. 

The new method improves upon conventional technologies by using the ADC to produce a diagnostic image of the airway tree without using ionizing radiation.  The method also may allow for cataloging airway tree structures in terms of lung ventilation, an important tool in diagnosing obstructive lung diseases.  With an improved segmentation and less invasive nature, the new MRI method surpasses conventional lung imaging techniques and will enhance the diagnosis and treatment of patients with respiratory diseases and cancers.

Magnetic Resonance Imaging of Metabolites

A UW–Madison researcher has developed a method for imaging in vivo metabolites that dramatically reduces scan time.

The metabolite images are produced with an MRI system using a priori information about their resonant peaks and relative sizes. This reduces the amount of NMR data needed for proper spectral resolution. Also, the NMR signal can be modeled with an equation of relatively few unknowns. Using this model and NMR data acquired at a plurality of echo time (TE), the metabolite at each image pixel can be calculated and imaged.

Multi-Mode Medical Tracking and Visualizing System for MR Guided Interventional Procedures

UW-Madison researchers have developed a multi-mode MR system that incorporates imaging and tracking coils into one device for therapeutic endovascular interventions.  The device includes two coils, one for tracking and one for imaging.  A switch connecting the two allows use of the tracking coil to move the device to the area of interest and then use of the imaging coil to acquire high resolution images of the area.  Only one coil connection to the MR scanner is needed because of the switch.

Real-Time Phase Error Correction for Off-Axis MRI Systems to Improve Efficiency and Image Quality

UW-Madison researchers have developed an improved method for off-axis MRI system calibration of non-Cartesian k-space data.  In this method a pulse sequence is performed to acquire k-space samples from which calibration phase data can be calculated.  A subsequent pulse sequence is performed while applying a frequency modulation that shifts the system axis to attain modified k-space samples centered on the region of interest, the heart for example.  Then a second calibration phase is calculated from the modified k-space and ideal phase data is calculated from the gradient waveform.  The phase error then is calculated from the first and second calibration and ideal phase data.  Finally, the timing error of acquired signals is calculated using the phase error and applied to the NMR data to correct image degradation.

This method minimizes artifacts from phase errors introduced by real-time demodulation hardware in the MRI system by measuring the timing delays that cause phase errors.  The correction can be made prospectively by offsetting the timing error during data acquisition or retrospectively by phase correction of the data after acquisition.  The improved method allows for more accurate processing of non-Cartesian imaging techniques, making MRI faster and more accurate in medical diagnostic imaging and other applications.

Method and System for Analyzing the Flow of Cerebrospinal Fluid

A UW-Madison researcher has developed improved algorithms and software for visualizing and quantifying cerebrospinal fluid flow (CSF) with magnetic resonance imaging (MRI). The algorithms also show promise for the detection and delineation of cancers.

The key advantage of this approach is that it offers spatial-temporal mapping, allowing physicians to measure CSF velocities at various locations in the central nervous system during different periods in the cardiac cycle. It makes diagnosis more accurate by detecting unusually high CSF velocities and by visualizing CSF movement into and out of the cranium at the same point in time.

Method for Improved Efficiency and Image Quality of Parallel MRI using Radial Acquisition Trajectory

UW-Madison researchers have developed an improved method for pMRI using a radial acquisition trajectory.  In this method, undersampled k-space data is acquired in parallel using samples along a radial trajectory.  The undersampled data then is used to reconstruct coil images, which in turn are used to produce coil sensitivity maps.  The coil sensitivity maps and undersampled data are utilized to calculate reference reconstruction coefficients for a coil by matrix inversion.  Additional reference reconstruction coefficients may be estimated by interpolation between those coefficients previously calculated.  Then all reconstruction coefficients and acquired k-space data is used to estimate missing k-space data and complete the k-space data set for each coil.  Individual coil images then are constructed from the completed k-space data sets and combined to produce the final image.

The improved method for pMRI will reduce the time required to produce an image after acquisition of the MR data and improve the quality of the images by minimizing motion artifacts.  The method is helpful for imaging organs in motion, such as the heart and lungs, and imaging patients who may otherwise need to be restrained, such as children or the mentally disabled.  Overall, the improved method of pMRI with radial acquisition trajectories will reduce imaging time and improve image quality, making MRI a more convenient and accurate tool in the medical field and other applications.

Parallel Magnetic Resonance Imaging Method Using a Radial Acquisition Trajectory

UW-Madison researchers have developed a post-processing algorithm that quickly compiles a high-quality composite of radial trajectory magnetic resonance images. In radial acquisition of an MRI, data points are more frequent in the central region. In the outer region, data is undersampled and must be estimated on the basis of training data from a preliminary scan. This algorithm generates training data based on the densely sampled central region rather than a preliminary scan.

Mechanical Force Detection of Magnetic Fields Using Heterodyne Demodulation

UW-Madison researchers have developed a new, high-frequency magnetic field detector that uses mechanical force to directly detect a magnetic field. The device incorporates a coil-on-cantilever design, with a conductive loop placed on a vibrating cantilever beam with one fixed end. The frequency of the alternating current in the loop is chosen so the magnitude of the cantilever’s mechanical vibration at its mechanical resonant frequency reflects the magnitude of the magnetic field. The vibration of the cantilever is then detected by reflecting a laser beam off a reflective portion at the cantilever’s free end.

Phase Contrast MRI with Dual Velocity-Encoded Projection Reconstruction Acquisition

UW-Madison researchers have now developed a dual VENC projection reconstruction acquisition strategy to prevent aliasing of blood velocity during phase contrast MRI studies. The method forms a series of imaging time frames that are each composed of a large set of projections acquired at a low velocity encoding value (low VENC). Interleaved with these time frames is a much smaller set in which projection data is acquired at a high velocity encoding value (high VENC). The high VENC acquisitions are then used to identify and correct regions of aliasing in the final angiographic image. In essence, by enabling the use of lower VENC values without the fear of causing aliasing artifacts, this strategy can provide images with enhanced velocity-to-noise ratios over those possible with other PC techniques.

Magnetic Resonance Imaging with Fat Suppression

UW-Madison researchers have now developed a technique to combine the echoes acquired at the beginning and end of the TR in dual-echo VIPR. This new method suppresses either fat or water in a single acquisition, rather than two, thus significantly reducing the amount of scan time needed. The technique’s main use will be to dampen lipid signals and reduce banding artifacts in images created by steady-state free precession (SSFP), a rapid imaging technique of growing clinical interest.

MRI Method for Assessing Myocardial Viability

Two UW-Madison researchers have now created a technique for retrospectively selecting the optimal TI value during MRI scans of heart tissue. Their invention takes advantage of the under-sampled projection reconstruction (PR) acquisition technique, which collects k-space data as a series of radial projections through the center of k-space. Because PR acquisition intrinsically over-samples the center of k-space, TI can be retrospectively selected by employing a sliding-window technique with a temporal aperture varying with the radial distance.

Method to Suppress Background Tissues in Time-Resolved Magnetic Resonance Angiography

UW-Madison researchers have developed a method for removing static background tissues from a time-resolved series of MRA images without the need for a mask image or operator intervention. Based on a previous invention, called Vastly under-sampled Isotropic PRojection (VIPR), this method is a non-linear algorithm that leaves the high spatial frequency component of image data unaltered, while analyzing the low spatial frequency component pixel by pixel. A matrix equation that uses temporal information provided by VIPR at lower spatial frequencies is solved to identify pixels coming from static and linearly increasing signal. These pixels are then attenuated, resulting in suppression of fat and other background tissue in the final, diagnostic image.

Isotropic Imaging of Vessels with Fat Suppression

The UW-Madison researchers have now extended an eddy current correction method to acquire a second radial line during the rephasing portion of the gradient waveform. Thus, this technology doubles the amount of data that can be acquired during any VIPR imaging sequence, resulting in greatly enhanced signal-to-noise ratios and image sharpness.

Three-Dimensional Phase Contrast Imaging Using Interleaved Projection Data

A UW-Madison researcher has developed a technique, called SUPERVENC for Spectral imaging with Undersampled Projections and interleaved Velocity ENCoding, that allows capture of three-dimensional PC images in the same amount of time as TOF techniques. It also greatly simplifies VENC selection and provides isotropic resolution, advances that could make quantitative PC imaging superior to TOF methods for MR angiography applications.

Enhanced Method for Vessel Segmentation to Improve Magnetic Resonance Angiography Images

UW–Madison researchers have developed an improved MRA method that uses automated vessel segmentation. The method involves injecting the patient with a contrast agent and rapidly acquiring a series of NMR images during a time-resolved phase. Arterial and venous voxels then are automatically identified in the images from which the contrast enhancement reference curves are calculated. Finally, voxels are segmented using the calculated contrast enhancement reference curves.

Magnetic Resonance Angiography Using Floating Table Projection Imaging

A UW-Madison researcher has developed an improved technique for obtaining whole body MRA images as a patient is moved through the FOV of an MRI scanner. As the patient travels through the scanner on a table moving at approximately the same speed as a bolus of contrast agent moving through the patient’s vasculature, image data are continuously acquired for each sub-region of the patient’s body. To offset the effects of table motion, each data set is phase corrected to a reference position. These data are then used to construct an image of each sub-region. To speed image capture while maintaining image quality, the invention employs Mistretta’s Vastly Undersampled Imaging with Projections (VIPR) technique (see link below) to acquire image data.

Automatic Determination of the Arterial Input Function in Dynamic Contrast-Enhanced MRI

A UW-Madison researcher has developed a method of automatically determining the AIF from MRI images that eliminates reliance on offline workstations, trained personnel and the lengthy delay normally associated with MRI CBF measurements.

Frequency-Narrowed High-Power Diode Laser System with External Cavity

UW–Madison researchers have developed a high-power diode laser array system that uses an external cavity to narrow the spectral width, changing the output power from a broad spectrum to a very narrow spectrum. The light from each emitter is collimated, reflected off a diffraction grating and imaged back onto the emitter. This causes each diode to preferentially lase at the wavelength that is fed back. The result is a diode laser array that uses a much greater portion of the laser's output power.

Frequency-Narrowed High-Power Diode Laser Array Method and System

UW–Madison researchers have developed a high-power diode laser array system that uses an external cavity to narrow the spectral width, changing the output power from a broad spectrum to a very narrow spectrum. The light from each emitter is collimated, reflected off a diffraction grating and imaged back onto the emitter. This causes each diode to preferentially lase at the wavelength that is fed back. The result is a diode laser array that uses a much greater portion of the laser's output power.

Brain and Deep Tissue Visualization by Diffusion Tensor Imaging

UW–Madison researchers have developed a post-processing algorithm that accepts diffusion weighted images acquired with diffusion weighted gradients in any 3-D orientation and with any combination of eigenvalues as input.

The code calculates the diffusion tensors for each voxel and provides as output several types of maps, including trace, fractional anisotropy, volume ratio, absolute value color and vector maps. The algorithm also provides a description of the mean diffusion properties of the region of interest and details of the diffusion characteristics of selected voxels.