Thomas Küstner

The use of a generalized reconstruction by inversion of coupled systems (GRICS) approach for generic respiratory motion correction in PET/MR imaging

Respiratory motion is a source of artifacts in multimodality imaging such as PET/MR. Solutions include retrospective or prospective gating. They have however found limited use in clinical practice, since their increased overall acquisition duration to maintain overall image quality. More elaborate methods consist of using 4D MR datasets to extract spatial deformations in order to correct for the respiratory motion in PET. The main drawbacks of such approaches is the relatively long acquisition times associated with 4D MR imaging which is often incompatible with clinical PET/MR protocols. The objective of this work was to overcome these limitations by exploiting a generalized reconstruction by inversion of coupled systems (GRICS) approach. The methodology is based on a joint estimation of motion during the MR image reconstruction process, providing internal structure motion and associated deformation matrices for retrospective use in PET respiratory motion correction. This method was first validated on four MR volunteers and two PET/MR patient datasets by comparing GRICS generated MR images to 4D MR series obtained by retrospective gating. In a second step 4D PET datasets corresponding to acquired 4D MR images were simulated using the GATE Monte Carlo simulation platform. GRICS generated deformation matrices were subsequently used to correct respiratory motion in comparison to the 4D MR image based deformations both for the simulated and the two 4D PET/MR patient datasets. Results confirm that GRICS synchronized MR images correlate well with the acquired 4D MR series. Similarly, the use of GRICS for respiratory motion correction allows an equivalent percentage improvement on lesion contrast, position and size, considering the PET simulated tumors as well as PET real tumors. This work demonstrates the potential interest of using GRICS for PET respiratory motion correction in combined PET/MR using shorter duration acquisitions without the need for 4D MRI and associated specific MR sequences.

MR-based respiratory and cardiac motion correction for PET imaging.

HighlightsPET motion correction from simultaneously acquired MR‐derived motion model.Fast MR acquisition freeing scan time per PET bed for further diagnostic sequences.Clinically feasible setup: streamlined processing in Gadgetron evaluation on a cohort of 36 patients.Publicly available: https://sites.google.com/site/kspaceastronauts. Graphical abstract Figure. No caption available. ABSTRACT Purpose: To develop a motion correction for Positron‐Emission‐Tomography (PET) using simultaneously acquired magnetic‐resonance (MR) images within 90 s. Methods: A 90 s MR acquisition allows the generation of a cardiac and respiratory motion model of the body trunk. Thereafter, further diagnostic MR sequences can be recorded during the PET examination without any limitation. To provide full PET scan time coverage, a sensor fusion approach maps external motion signals (respiratory belt, ECG‐derived respiration signal) to a complete surrogate signal on which the retrospective data binning is performed. A joint Compressed Sensing reconstruction and motion estimation of the subsampled data provides motion‐resolved MR images (respiratory + cardiac). A 1‐POINT DIXON method is applied to these MR images to derive a motion‐resolved attenuation map. The motion model and the attenuation map are fed to the Customizable and Advanced Software for Tomographic Reconstruction (CASToR) PET reconstruction system in which the motion correction is incorporated. All reconstruction steps are performed online on the scanner via Gadgetron to provide a clinically feasible setup for improved general applicability. The method was evaluated on 36 patients with suspected liver or lung metastasis in terms of lesion quantification (SUVmax, SNR, contrast), delineation (FWHM, slope steepness) and diagnostic confidence level (3‐point Likert‐scale). Results: A motion correction could be conducted for all patients, however, only in 30 patients moving lesions could be observed. For the examined 134 malignant lesions, an average improvement in lesion quantification of 22%, delineation of 64% and diagnostic confidence level of 23% was achieved. Conclusion: The proposed method provides a clinically feasible setup for respiratory and cardiac motion correction of PET data by simultaneous short‐term MRI. The acquisition sequence and all reconstruction steps are publicly available to foster multi‐center studies and various motion correction scenarios.

Self‐navigated 4D cartesian imaging of periodic motion in the body trunk using partial k‐space compressed sensing

To enable fast and flexible high‐resolution four‐dimensional (4D) MRI of periodic thoracic/abdominal motion for motion visualization or motion‐corrected imaging.

MR Image Reconstruction Using a Combination of Compressed Sensing and Partial Fourier Acquisition: ESPReSSo

A Cartesian subsampling scheme is proposed incorporating the idea of PF acquisition and variable-density Poisson Disc (vdPD) subsampling by redistributing the sampling space onto a smaller region aiming to increase k-space sampling density for a given acceleration factor. Especially the normally sparse sampled high-frequency components benefit from this sampling redistribution, leading to improved edge delineation. The prospective subsampled and compacted k-space can be reconstructed by a seamless combination of a CS-algorithm with a Hermitian symmetry constraint accounting for the missing part of the k-space. This subsampling and reconstruction scheme is called Compressed Sensing Partial Subsampling (ESPReSSo) and was tested on in-vivo abdominal MRI datasets. Different reconstruction methods and regularizations are investigated and analyzed via global (intensity-based) and local (region-of-interest and line evaluation) image metrics, to conclude a clinical feasible setup. Results substantiate that ESPReSSo can provide improved edge delineation and regional homogeneity for multidimensional and multi-coil MRI datasets and is therefore useful in applications depending on well-defined tissue boundaries, such as image registration and segmentation or detection of small lesions in clinical diagnostics.

Combined unsupervised-supervised classification of multiparametric PET/MRI data: application to prostate cancer.

Multiparametric medical imaging data can be large and are often complex. Machine learning algorithms can assist in image interpretation when reliable training data exist. In most cases, however, knowledge about ground truth (e.g. histology) and thus training data is limited, which makes application of machine learning algorithms difficult.

Systematic Evaluation of Amide Proton Chemical Exchange Saturation Transfer at 3 T: Effects of Protein Concentration, pH, and Acquisition Parameters.

ObjectiveThe goal of this work was to systematically evaluate the reproducibility of amide proton transfer chemical exchange saturation transfer (APT-CEST) at 3 T and its signal dependence on pH, protein concentration, and acquisition parameters. An in vitro system based on bovine serum albumin (BSA) was used, and its limitations were tested by comparing it to in vivo measurements. The contribution of small endogenous metabolites on the APT-CEST signal at 3 T was also investigated. In addition, the reliability of different z-spectrum interpolations as well as the use of only a few frequency offset data points instead of a whole z-spectrum were tested. Materials and MethodsWe created both a BSA phantom at different concentrations and pH values and a metabolite phantom with different small molecules. Chemical exchange saturation transfer data were acquired using a 2-dimensional fast spoiled gradient-echo sequence with pulsed CEST preparation at different saturation durations and power levels. Healthy volunteer measurements were taken for comparison. Z-spectra were interpolated using a 24th-order polynomial (Poly), an eighth-order Fourier series (Fourier), and a smoothing Spline (sSpline) algorithm. To evaluate reduced data sets, only 6 to 14 frequency offsets of the z-spectrum were used and interpolated via a cubic Spline. Region of interest (ROI) evaluations were used to investigate the reproducibility of amide magnetization transfer ratio asymmetry [MTRasym(3.5 ppm)] and to analyze the MTRasym and z-spectra. ResultsInterscan standard deviations of MTRasym(3.5 ppm) were always below 0.3%. MTRasym(3.5 ppm) increased when the BSA concentrations increased and decreased when the pH increased. The amine MTRasym signal of small molecules was very small compared with BSA and was only detectable using short saturation times and higher power levels. The MTRasym(3.5 ppm) between BSA concentration steps and between nearly all pH steps was significantly different for all 3 fitting methods. The Fourier and sSpline methods showed no statistically significant differences; however, the results for the Poly method were significantly higher at some concentrations and pH values. Using only few frequency offsets resulted in less significant differences compared with fitting the complete z-spectrum. In general, MTRasym(3.5 ppm) of gray matter, white matter, and ventricle ROIs from volunteer scans increased with an increase in saturation power and partially decreased with an increase in saturation duration. Intra-ROI covariances of MTRasym(3.5 ppm) revealed the highest variations for Poly, whereas using reduced spectral data resulted in an increased signal variation. ConclusionsAmide proton transfer–CEST imaging is a highly reproducible method in which absolute signal differences of approximately 0.5% are detectable in principle. For in vivo applications, Fourier or sSpline interpolations of z-spectra are preferable. Using reduced data sets delivers similar results but with increased variation and therefore decreased (pH/concentration) differentiation capability. Differentiation capability increases with increases in the saturation duration and power level. The results from the in vitro BSA system cannot be directly transferred to the in vivo situation due to different chemical environments resulting in, for example, higher asymmetric macromolecular cMT effects in vivo. Amine signals from small molecules are unlikely to contribute to APT-CEST at 3 T (except for creatine); however, signals can be enhanced by using short saturation times and higher power levels.

Automated reference-free detection of motion artifacts in magnetic resonance images

ObjectivesOur objectives were to provide an automated method for spatially resolved detection and quantification of motion artifacts in MR images of the head and abdomen as well as a quality control of the trained architecture.Materials and methodsT1-weighted MR images of the head and the upper abdomen were acquired in 16 healthy volunteers under rest and under motion. Images were divided into overlapping patches of different sizes achieving spatial separation. Using these patches as input data, a convolutional neural network (CNN) was trained to derive probability maps for the presence of motion artifacts. A deep visualization offers a human-interpretable quality control of the trained CNN. Results were visually assessed on probability maps and as classification accuracy on a per-patch, per-slice and per-volunteer basis.ResultsOn visual assessment, a clear difference of probability maps was observed between data sets with and without motion. The overall accuracy of motion detection on a per-patch/per-volunteer basis reached 97%/100% in the head and 75%/100% in the abdomen, respectively.ConclusionAutomated detection of motion artifacts in MRI is feasible with good accuracy in the head and abdomen. The proposed method provides quantification and localization of artifacts as well as a visualization of the learned content. It may be extended to other anatomic areas and used for quality assurance of MR images.

3D motion flow estimation using local all-pass filters

Fast and accurate motion estimation is an important tool in biomedical imaging applications such as motion compensation and image registration. In this paper, we present a novel algorithm to estimate motion in volumetric images based on the recently developed Local All-Pass (LAP) optical flow framework. The framework is built upon the idea that any motion can be regarded as a local rigid displacement and is hence equivalent to all-pass filtering. Accordingly, our algorithm aims to relate two images, on a local level, using a 3D all-pass filter and then extract the local motion flow from the filter. As this process is based on filtering, it can be efficiently repeated over the whole image volume allowing fast estimation of a dense 3D motion. We demonstrate the effectiveness of this algorithm on both synthetic motion flows and in-vivo MRI data involving respiratory motion. In particular, the algorithm obtains greater accuracy for significantly reduced computation time when compared to competing approaches.

Active learning for magnetic resonance image quality assessment

In medical imaging, the acquired images are usually analyzed by a human observer and rated with respect to a diagnostic question. However, this procedure is time-demanding and expensive. Further more, the lack of a reference image makes this task challenging. In order to support the human observer in assessing image quality and to ensure an objective evaluation, we extend in this paper our previous no-reference magnetic resonance (MR) image quality assessment system with an active learning loop to reduce the amount of necessary labeled training data. We employ two different active learning query strategies based on uncertainty sampling. Since the classification task is performed on 2D image slices, but the human observer labels complete 3D image volumes, we present a method to select representative 3D images instead of independant 2D image slices. The performance is evaluated on in-vivo MR image data.

Combining Compressed Sensing with motion correction in acquisition and reconstruction for PET/MR

In the field of oncology, simultaneous Positron-Emission-Tomography/Magnetic Resonance (PET/MR) scanners offer a great potential for improving diagnostic accuracy. However, to achieve a high Signal-to-Noise Ratio (SNR) for an accurate lesion detection and quantification in the PET/MR images, one has to overcome the induced respiratory motion artifacts. The simultaneous acquisition allows performing a MR-based non-rigid motion correction of the PET data. It is essential to acquire a 4D (3D + time) motion model as accurate and fast as possible to minimize additional MR scan time overhead. Therefore, a Compressed Sensing (CS) acquisition by means of a variable-density Gaussian subsampling is employed to achieve high accelerations. Reformulating the sparse reconstruction as a combination of the inverse CS problem with a non-rigid motion correction improves the accuracy by alternately projecting the reconstruction results on either the motion-compensated CS reconstruction or on the motion model optimization. In-vivo patient data substantiates the diagnostic improvement.