Contact
Prof. Dr. Anja Braune
Head Positron Emission Tomography
a.braune
hzdr.de
Phone: +49 351 260 2621
Previous Research Topics
Effects of radiotherapy on normal tissue
Glioblastoma multiforme is the most common primary malignant brain tumor in adults. The standard therapy is maximal surgical resection followed by radiotherapy (RT) with concurrent and adjuvant chemotherapy using temozolomide. Both RT and chemotherapy are, however, associated with risks of cognitive deficits and structural and hemodynamic changes in the normal brain tissue. A detailed understanding of the relation between the received dose during RT and the changes in brain tissue are of obvious relevance for treatment planning as well as for diagnostic evaluation of the disease progression and for determining tissue specific dose tolerance levels.
We have shown the decrease of perfusion in the healthy brain tissue following radiochemotherapy using a non-invasive ASL perfusion imaging. We are currently working on analyzing the structural changes and on comparing these effects between 3D-CRT and IMRT photon therapy and proton-therapy.
Key publications:Partial Volume Effects (PVE)
Despite the recent developments, the spatial resolution of ASL MRI perfusion measurement has remained relatively low with the majority of voxels containing a mixture of perfusion signals from gray matter (GM), white matter (WM), and CSF, a phenomenon referred to as partial volume effects. Since GM perfusion is reported to be 2 — 4.5 times higher than WM perfusion [9, 10] and CSF is not perfused, the measured ASL signal in a given voxel is dependent on the fractional contributions of GM and WM to the voxel, i.e., its tissue composition. Consequently, differences in measured perfusion across regions within a subject or for the same region across subjects could, to a varying degree, be attributable to differences in tissue composition variation rather than actual changes in perfusion. This is an important topic for perfusion imaging in, both, normal tissue and tumor.
We are continuously working on the improvement of methods for partial volume effects correction in ASL. We have several publications on automatic detection of hypo-perfusion and on improved method for obtaining of partial volume maps — a key component to the partial volume correction algorithm. We are currently working on identifying the issues related to geometric deformations and effective resolution and their correction for improved accuracy of partial volume correction.
Key publications:- Petr et al., ISMRM 2016
- Petr et al., Human brain mapping 35.(4), 1179–1189, 2014
- Petr et al., Magnetic Resonance in Medicine 70.(6), 1535–1543, 2013
Translation to clinical research and application
We are a key contributor to the development of the software platform ExploreASL. Initiated through the EU-funded COST-action "ASL In Dementia", ExploreASL is a collaborative framework of researchers and clinical investigators. It was already used to process ~4000 ASL images from all major MRI vendors and ASL sequences, and a large variety of patient populations. The ultimate goal is to combine data from a large number of studies and identify common and different perfusion patterns to enhance knowledge of ASL image analysis and of the role of perfusion and structural changes in neurodegenerative pathophysiology.
Our recent work focused on improving the ASL image processing across different vendors and ASL implementations in order to increase repeatability of ASL across different centers. We have also developped a new biomarker that is extracted from ASL data that is more sensitive to changes in cerebrovascular reactivity than the perfusion itself and has thus promising applications in the tumor research. Our current acitvity focuses on adjusting the processing pipeline to work efficiently work with tumor patients and to overcome several issues in co-registration and perfusion quantification that are specific for this type of patients.
Key publications:- Mutsaerts, Petr et al., Journal of cerebral blood flow and metabolism, in press, 2017
- Mutsaerts, Petr et al., Journal of magnetic resonance imaging, in press, 2017
- Mutsaerts, Petr et al., ESMRMB, 2017
Assessment of in-vivo quantification accuracy
Positron emission tomography (PET) allows to quantify regional tracer uptake and to derive parameters such as standardized uptake values (SUV) and tracer kinetic transport constants which help to objectify and improve prediction and monitoring of therapy outcome in various tumor diseases. For many years now, PET has been successfully used in combined PET and computed tomography (CT) systems (PET/CT) for such purposes. More recently, PET and magnetic resonance imaging (MRI) have been combined in hybrid PET/MR imaging devices. This combination has the advantage that MRI offers improved soft tissue contrast in comparison with CT. However, PET/MR complicates accurate quantification of the PET data. Consequently, the quantification capabilities of combined PET/MR systems need to be investigated thoroughly. Most of the work in this context has focused on standard phantom measurements but such measurements do not provide direct information regarding quantification accuracy in patient studies.
Recently, a strategy for direct assessment of the actual in vivo quantification accuracy of a PET system has been proposed by us and applied to two different PET/CT systems and one stand-alone PET system. In this work, a comparison was performed between tracer activity in the bladder as measured by PET and urine samples measured in a cross-calibrated well-counter. It was shown that there is a systematic modest underestimate across different PET and PET/CT systems.
We could further demonstrate that our PET/MR system underestimates true activity concentration in the pelvic region more serverely than the previously investigated three PET and PET/CT systems. The most probable cause for the increased underestimate seems to be a number of shortcomings (e.g. inaccurate truncation compensation) of present-day MR-based attenuation correction.
Key publications:Evaluation of MR-based attenuation correction
To correctly quantify tracer concentrations in positron emission tomography (PET), attenuation correction is mandatory since the emitted annihilation photons can interact with the tissue prior to reaching the PET detectors and are thus scattered or absorbed in the body. The fraction of photons that reaches the detectors without any interaction naturally depends on the tissue type and amount of tissue between point of emission and detector.
To analyze the quantitative accuracy of MR-based attenuation correction (MRAC), we conducted a patient study in cooperation with the Department of Nuclear Medince at the University Hospital Dresden. In this study, patients were examined first in a stand-alone PET scanner (which uses 68-Ge sources for a direct measurement of the attenuation). Subsequently, all patients were examined in a combined PET/MRI. In this way, we were able to reconstruct the emission data from the PET/MRI with two different methods for attenuation correction. First, we used the vendor-provided MR-based attenuation correction and second we used the measured transmission form the standalone PET. A comparison of the two reconstructed PET images showed that MRAC-based deviations in the reconstructed PET activity concentration are in the range from -15% to +15% if the tissue-type-segmentation works correctly. In case of segmentation errors, however, much more serious deviations of the order of 50% can occur.
Key publications:- Schramm et al., IEEE Trans. Med. Imaging 32, 2056-2063, 2013
- Schramm et al., MAGMA 25, 115-126, 2013
Improved MR-based attenuation correction
Because the vendor-provided tissue type segmentation results in incorrect attenuation images in the presence of MR image artifacts (e.g. resulting from metal implants), we developed a new improved segmentation methode that is able to compensate for these artifacts. This algorithm uses a delineation of a dedicated MR image and a separate delineation of the PET image. An example showing the results of this approach in a patient with hip and knee endoprotheses is shown.
Key publications:- Schramm et al., Medical Physics 42, 6468-6476, 2015
- Schramm et al., Phys. Med. Biology 59, 2713-2726, 2014
Clinical Translation
To be able to evaluate the practical benefit of our developed methods in a clinical environment, we have developed a software platform (pyPRS) which allows to use our aforementioned methods in a clinical setting. This includes semi-automatically reconstructing PET/MR data (which is affected by inaccurate MR-based attenuation correction) with our improved MR segmentation, providing the clinicians with a corrected image dataset with improved quantitative accuracy.
Key publications:Tube of response High resolution OSEM Reconstruction (THOR)
A central task in iterative PET image reconstruction is adequate modeling of the system matrix and its efficient computation. System matrix modeling has a crucial influence on the resulting image quality. The elements of the system matrix describe the transformation between measured projection data and image space. They are defined by the geometrical probability that a photon pair emitted from a voxel is detected in a certain detector pair. Conventionally, this probability is considered to be proportional to the length of intersection between the corresponding line of response (LoR) and the voxel. This approach possesses relatively low computational costs and is suitable for on-the-fly calculation, where "on-the-fly" denotes the calculation of elements of the system matrix during the reconstruction process rather than pre-calculating and storing them on hard disk or, possibly, in memory. However, in its basic form this approach neglects the finite size of the detector crystals and other resolution degrading effects and thus does not provide a realistic model of the given physical system.
One of the ways to improve system modeling is to replace the length of intersection of an LoR by the volume of intersection of a tube of response (ToR) since this allows to take the finite detector size explicitly into account. We proposed a model for the on-the-fly volume of intersection system matrix calculation which was then integrated into an in-house implementation of a fully 3D list-mode reconstruction (Tube of response High resolution OSEM Reconstruction - THOR) based on the maximum-likelihood expectation-maximization algorithm. The reconstruction includes all necessary corrections for detector normalization and attenuation as well as random and scattered coincidences. THOR features flexible scanner support with Siemens ECAT HR+ and Philips Ingenuity TF PET/MR systems currently implemented. Utilization of THOR for reconstruction of the images from Philips Ingenuity PET/MR scanner allows to increase reconstructed images resolution to < 4 mm compared to ≈ 6 mm resolution when using the vendor's reconstruction.
Key publications:- Lougovski et al., Phys. Med. Biology 60 (10), 2015
- Lougovski et al., Phys. Med. Biology 59 (3), 2014
Improved Scatter Correction
Utilization of Time-Of-Flight (TOF) information provides large benefits in terms of reduced image noise and improved convergence rate for PET image reconstruction. However, Scatter Correction (SC) becomes much more complicated compared to the non-TOF case. Existing TOF-SC algorithms are either much slower than their non-TOF counterparts or introduce noticeable reconstruction artifacts.
The accelerated TOF-SC algorithm proposed by our group demonstrates a similar scatter correction precision compared to well known TOF-Single Scatter Simulation (TOF-SSS) by Watson while performing 1.5 times faster in terms of overall reconstruction time. The newly proposed method does not cause any kind of halo artifacts under high contrast conditions which are clearly pronounced in vendor reconstruction.
Key publications:MLAA-based Attenuation Correction
As is well known, quantitative combined PET/MR imaging depends on accurate MR-based attenuation correction (MRAC). While a mostly satisfactory state of affairs has been reached today, problems persist regarding segmentation errors including unsatisfactory bone identification and residual systematic differences in comparison to PET/CT. Alternative or complementary strategies for attenuation correction (AC), therefore, are of considerable relevance. In this context, Maximum Likelihood reconstruction of Attenuation and Activity (MLAA) is one of the most promising approaches.
The TOF version of the MLAA algorithm was implemented for the Philips Ingenuity-TF PET/MR scanner by incorporation into the THOR reconstruction tool. For evaluation, different brain and whole body studies were reconstructed. We could show that MLAA is capable to reconstruct skull and air cavities in head and lungs in whole body attenuation maps. Moreover, metal implants in one patient which caused severe artifacts in MR-derived attenuation maps were successfully identified by MLAA.
Key publications:Compensation of head motion
Apart from respiratory motion, involuntary or voluntary patient movement during a PET acquisition is another source of image blurring. While in case of whole body motion motion compensation methods are substantially complicated due to the non-rigid nature of motion in that case, motion of the head during a brain acquisition can be corrected quite accurately. One well-known method for correction of head motion is the so-called MAF (multiple acquisition framing) method which splits PET data into seperate (acquisition) time frames based on the significance of occurred motion. This has the disadvantage, however, that one needs to compromise between accuracy of the motion correction and the minimum required statistical accuracy (i.e. acquisition time) of the separate acquisition frames.
We, therefore, developed a motion compensation method which is applied directly to the raw list-mode data of a PET measurement instead. By spatially correcting each single coincidence event with the motion information gained by extern motion tracking devices we were able to show that applying a list-mode based motion compensation can outperform the standard MAF method for correcting head motion affected PET data.
Key publications:- Langner, Ph.D. thesis, 2009
- Langner et al., EANM, 2008
- Bühler et al., IEEE Trans. Med. Imaging, 1176-1185, 2004
Compensation of respiratory motion
For accurate quantification of tumor metabolism and tumor size reliable delineation of the true tumour boundaries is required. Patient motion leads to blurring of the acquired PET images which adversely affects this task. This is especially true for breathing induced motion. In such cases PET images of the chest will contain intrinsic motion artefacts that can make it almost impossible to, e.g., precisely delineate lung tumours. We are, therefore, working on motion compensation methods which allow to minimize the adverse effects of patient motion.
Regarding respiratory motion we have developed different motion compensation approaches which allow to track patient motion either using optical tracking systems or by using belt-based systems to register respiratory motion in a clinical setup. This allows to not only display and evaluate the effects of patient motion on a PET acquisition but also apply motion compensation methods such as amplitude-based gating which splits PET data based on phase information gathered from detailed motion analyses. Furthermore, we are working on development of motion compensation methods which allow to map data of different gates to a single motion compensated image (using non-rigid motion transforms) without loosing the statistical precision a single gate solution imposes.
In fact, we could show that our motion compensation methods are working in a clinical setup and that there is a huge potential for improving quantitative assessment of the tumour metabolism if our methods are directly applied.
Motion Compensation in clinical application
In close collaboration with the University Hospital Dresden our motion compensation methods have been integrated into clinical routine. In over 1000 patient measurements we were able to apply our head motion compensation techniques and provide nuclear physicians with motion compensated PET images as well as information on the significance of occurred patient motion. Furthermore, in case of respiratory motion we have implemented our method in a way that medical physics experts are able to compensate for respiratory motion and generate corrected images for evaluation by the resonsible physicians. For ease of use in the clinical context dedicated graphical user interfaces have been developed as well.
Key publications:
