Georg Schmitz

Ruhr-Universität Bochum, Bochum, North Rhine-Westphalia, Germany

Are you Georg Schmitz?

Claim your profile

Publications (120)116.67 Total impact

  • Dimitri Ackermann · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: The imaging of microvessels and the quantification of their blood flow is of particular interest in the characterization of tumor vasculature. The imaging resolution (50-200 μm) of high frequency ultrasound (20-50 MHz) is not sufficient to image microvessels (~10 μm) and Doppler sensitivity is not high enough to measure capillary blood flow (~1mm/s). For imaging of blood flow in microvessels our approach is to detect single microbubbles (MBs), track them over several frames and to estimate their velocity. First, positions of MBs will be detected by separating B-mode frames in a moving foreground and a static background. For the crucial task of association of these positions to tracks we implemented a modified Markov Chain Monte Carlo Data Association (MCMCDA) algorithm, which can handle a high number of MBs. False alarms, the detection, initiation and termination of MBs tracks are incorporated in the underlying model. To test the algorithms performance an ultrasound imaging simulation of a vessel tree with flowing MBs was set up (resolution 148 μm). The trajectories and flow velocity in the vessels with a lateral distance of 100 μm were reconstructed with super-resolution. In a phantom experiment, a suspension of MBs was pumped through a tube (diameter 0.4 mm) at speeds of 2.2, 4.2, 6.3 and 10.5 mm/s and was imaged with a Vevo2100 system (Visualsonics). The estimated MBs' mean speeds were 2.1, 4.7, 7 and 10.5 mm/s. To demonstrate the applicability for in vivo measurements, a tumor xenograft bearing mouse was imaged by this approach. The tumor vasculature was visualized with higher resolution than in a maximum intensity projection image and the velocity values were in the expected range of 0 to 1 mm/s.
    No preview · Article · Nov 2015 · IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Our objective was to determine the lowest diagnostically effective dose for E-selectin-targeted poly n-butyl cyanoacrylate (PBCA)-shelled microbubbles and to apply it to monitor antiangiogenic therapy effects. Procedures: PBCA-shelled microbubbles (MBs) coupled to an E-selectin-specific peptide were applied in mice carrying MLS or A431 carcinoma xenografts scaling down the MB dosage to the lowest level where binding could be examined with a 18-MHz small animal ultrasound transducer. Differences in E-selectin expression in the two carcinoma xenografts were confirmed by enzyme-linked immunosorbent assay (ELISA). In addition, MLS tumor-bearing mice under antiangiogenic therapy were monitored using E-selectin-targeted MBs at the lowest applicable dose. Therapy effects on tumor vascularization were verified by immunohistological analyses. Results: The minimally required dosage was 7 × 10(7) MBs/kg body weight. This dosage was sufficient to enable E-selectin detection in high E-selectin-expressing MLS tumors, while low E-selectin-expressing A431 tumors required almost 2.5-fold higher doses. At the dose of 7 × 10(7) MBs/kg body weight, a decrease in E-selectin MB binding under antiangiogenic therapy could be assessed (being significant after 3 days of treatment; p < 0.0001), which was in line with the significant drop in E-selectin-positive area fractions that was found histologically (p < 0.05). Conclusions: Molecular ultrasound imaging with our E-selectin-targeted MB and therapy monitoring was possible down to a dose of 7 × 10(7) MBs/kg body weight (equates to 66 μg PBCA/kg and 4.6 mg PBCA/70 kg). Improvements in choice of targets, MB composition, and other MB detection methods may improve sensitivity and lead to reliable detection results of clinically transferrable MBs at even lower dosage levels.
    No preview · Article · Sep 2015 · Molecular imaging and biology: MIB: the official publication of the Academy of Molecular Imaging
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To enhance the regional antitumor activity of the vascular-targeting agent truncated tissue factor (tTF)-NGR by combining the therapy with low-energy ultrasound (US) treatment. For the in vitro US exposure of human umbilical vein endothelial cells (HUVECs), cells were put in the focus of a US transducer. For analysis of the US-induced phosphatidylserine (PS) surface concentration on HUVECs, flow cytometry was used. To demonstrate the differences in the procoagulatory efficacy of TF-derivative tTF-NGR on binding to HUVECs with a low versus high surface concentration of PS, we performed factor X activation assays. For low-energy US pretreatment, HT1080 fibrosarcoma xenotransplant-bearing nude mice were treated by tumor-regional US-mediated stimulation (ie, destruction) of microbubbles. The therapy cohorts received the tumor vessel-infarcting tTF-NGR protein with or without US pretreatment (5 minutes after US stimulation via intraperitoneal injection on 3 consecutive days). Combination therapy experiments with xenotransplant-bearing nude mice significantly increased the antitumor activity of tTF-NGR by regional low-energy US destruction of vascular microbubbles in tumor vessels shortly before application of tTF-NGR (P < .05). Mechanistic studies proved the upregulation of anionic PS on the outer leaflet of the lipid bilayer of endothelial cell membranes by low-energy US and a consecutive higher potential of these preapoptotic endothelial cells to activate coagulation via tTF-NGR and coagulation factor X as being a basis for this synergistic activity. Combining retargeted tTF to tumor vessels with proapoptotic stimuli for the tumor vascular endothelium increases the antitumor effects of tumor vascular infarction. Ultrasound treatment may thus be useful in this respect for regional tumor therapy. © 2015 by the American Institute of Ultrasound in Medicine.
    Full-text · Article · Jul 2015 · Journal of ultrasound in medicine: official journal of the American Institute of Ultrasound in Medicine
  • Source

    Full-text · Article · Jul 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Quantitative Cardiovascular Magnetic Resonance (CMR) techniques have gained high interest in CMR research. Myocardial T2 mapping is thought to be helpful in diagnosis of acute myocardial conditions associated with myocardial edema. In this study we aimed to establish a technique for myocardial T2 mapping based on gradient-spin-echo (GraSE) imaging. The local ethics committee approved this prospective study. Written informed consent was obtained from all subjects prior to CMR. A modified GraSE sequence allowing for myocardial T2 mapping in a single breath-hold per slice using ECG-triggered acquisition of a black blood multi-echo series was developed at 1.5 Tesla. Myocardial T2 relaxation time (T2-RT) was determined by maximum likelihood estimation from magnitude phased-array multi-echo data. Four GraSE sequence variants with varying number of acquired echoes and resolution were evaluated in-vitro and in 20 healthy volunteers. Inter-study reproducibility was assessed in a subset of five volunteers. The sequence with the best overall performance was further evaluated by assessment of intra- and inter-observer agreement in all volunteers, and then implemented into the clinical CMR protocol of five patients with acute myocardial injury (myocarditis, takotsubo cardiomyopathy and myocardial infarction). In-vitro studies revealed the need for well defined sequence settings to obtain accurate T2-RT measurements with GraSE. An optimized 6-echo GraSE sequence yielded an excellent agreement with the gold standard Carr-Purcell-Meiboom-Gill sequence. Global myocardial T2 relaxation times in healthy volunteers was 52.2 ± 2.0 ms (mean ± standard deviation). Mean difference between repeated examinations (n = 5) was −0.02 ms with 95% limits of agreement (LoA) of [−4.7; 4.7] ms. Intra-reader and inter-reader agreement was excellent with mean differences of −0.1 ms, 95% LoA = [−1.3; 1.2] ms and 0.1 ms, 95% LoA = [−1.5; 1.6] ms, respectively (n = 20). In patients with acute myocardial injury global myocardial T2-RTs were prolonged (mean: 61.3 ± 6.7 ms). Using an optimized GraSE sequence CMR allows for robust, reliable, fast myocardial T2 mapping and quantitative tissue characterization. Clinically, the GraSE-based T2-mapping has the potential to complement qualitative CMR in patients with acute myocardial injuries.
    Full-text · Article · Feb 2015 · Journal of Cardiovascular Magnetic Resonance
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Multispectral photoacoustic laser diode systems have multiple wavelengths available simultaneously. In addition to multispectral imaging, this can be exploited to increase the signal to noise ratio (SNR) by combining these wavelengths to form a combined image, but at the loss of spectral information. Here, a novel signal processing concept is introduced, which optimizes the SNR in the reconstructions of single wavelength data from combined acquisitions while simultaneously permitting to obtain a higher SNR fused image from the same data. The concept is derived for an arbitrary number of wavelengths; it is also applicable at low pulse repetition frequencies. The concept is applied in an experiment using two wavelengths, verifying the theoretical results.
    Preview · Article · Jan 2015 · Optics Express
  • [Show abstract] [Hide abstract]
    ABSTRACT: We report an experimental finding of photoacoustic signal enhancement from finite sized DNA–gold nanoparticle networks. We synthesized DNA-functionalized hollow and solid gold nanospheres (AuNS) to form finite sized networks, which were characterized by means of optical extinction spectroscopy, dynamic light scattering, and scanning electron microscopy in transmission mode. It is shown that the signal amplification scales with network size for networks comprising either hollow or solid AuNS as well as networks consisting of both types of nanoparticles. The laser intensities applied in our multispectral setup (λ = 650 nm, 850 nm, 905 nm) were low enough to maintain the structural integrity of the networks. This reflects that the binding and recognition properties of the temperature-sensitive cross-linking DNA-molecules are retained.
    No preview · Article · Oct 2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: To evaluate the use of the recently proposed ultrafast B1 (+) mapping approach DREAM (Dual Refocusing Echo Acquisition Mode) for a refinement of patient adaptive radiofrequency (RF) shimming. Materials and methods: Volumetric DREAM B1 (+) calibration scans centered in the upper abdomen were acquired in 20 patients and three volunteers with written informed consent at a clinical dual source 3 Tesla (T) MR system. Based on these data, RF transmit settings were optimized by central-slice based RF-shimming (CS-RF shim) and by a refined, multi-slice adaptive approach (MS-RF shim). Simulations were performed to compare flip angle accuracy and B1 (+) homogeneity (cv = stddev/mean) achieved by CS-RF shim versus MS-RF shim for transversal and coronal slices, and for volume shimming on the spine. Results: By MS-RF shim, mean deviation from nominal flip angle was reduced to less than 11% in all slices, all targets, and all subjects. Relative improvements in B1 (+) cv (MS-RF shim versus CS-RF) were up to 14%/39%/47% in transversal slices/coronal slices/ spine area. Conclusion: Volumetric information about B1 (+) can be used to further improve the accuracy and homogeneity of the B1 (+) field yielding higher diagnostic confidence, and will also be of value for various quantitative methods which are sensitive to flip angle imperfections.
    No preview · Article · Oct 2014 · Journal of Magnetic Resonance Imaging
  • Leili Salehi · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: Nonlinear ultrasound diffraction tomography reconstructs material parameters of a medium from scattered sound waves taking into account multiple scattering. The reconstruction algorithm based on the (KM) reconstructs the spatially varying speed of sound (SoS), equivalent to a spatially varying compressibility, in the time domain. On the other hand, the Contrast Source Inversion method (CSI) reconstructs the SoS from the refractive index of an inhomogeneous object in a known background in the frequency domain. In this work, the reconstruction results for the SoS applying both methods to the same object in the same medium are compared. The results show smaller errors for Kaczmarz method but higher convergence rate for the Contrast Source Inversion algorithm.
    No preview · Conference Paper · Sep 2014
  • Stefanie Dencks · Hesty Susanti · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: Sonography is the standard technique for monitoring local punctures with needles. Using convex arrays for deeper puncture targets, an increasingly poorer needle visibility is expected. However, it is observed that the needle reappears for large incidence angles of the ultrasound wave to the needle. It is investigated, whether this effect is caused by resonant scattering of the needle. Measurements and simulations demonstrate that resonant scattering occurs for angles of incidence larger than the Rayleigh-angle caused by the excitation of helical guided waves propagating circumferentially and axially. This effect could be exploited to improve needle visibility by adaptive beamforming.
    No preview · Conference Paper · Sep 2014
  • Markus C. Hesse · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: While linear pulse-echo ultrasound imaging methods (e.g. B-mode (BM), delay-and-sum (DAS) beamforming, synthetic aperture focusing technique (SAFT), filtered backpropagation (FBP), etc.) solely reach qualitative images mainly showing tissue boundaries assuming single scattering, nonlinear diffraction tomographic reconstruction methods have been proposed to reconstruct quantitative distributions of various acoustic tissue parameters as attenuation, compressibility, mass density or speed of sound under multiple scattering. In the present contribution, we analyze numerically the image reconstruction robustness of a previously suggested nonlinear simultaneous compressibility and mass density reconstruction algorithm under plane wave excitation to raw data with various noise levels. The image reconstruction results obtained for an unidirectional pulse-echo breast imaging application using raw data sets with maximum signal-to-noise ratios (SNRs) of 30 dB, 40 dB, 50 dB and 60 dB demonstrate the method's robustness up to 50 dB.
    No preview · Conference Paper · Sep 2014
  • Martin F. Schiffner · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: We introduced the fast multipole method (FMM) into our concept for plane wave pulse-echo ultrasound imaging (UI) to reduce the memory consumption and the computational costs associated with the numerical solution of the underlying regularized linear inverse scattering problem (ISP). For an example of typical size and in comparison to the conventional approach, we showed that the FMM requires less than 0.25% of the memory and less than 24% of the number of complex-valued multiplications. The FMM thus enables the numerical solution of the regularized (e.g. by compressed sensing) linear ISP on standard personal computers. It significantly improves the applicability of inverse scattering strategies in practical UI.
    No preview · Conference Paper · Sep 2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: The aim of a photoacoustic image reconstruction is to locate acoustic sources that originate from laser irradiation. Many conventional one-step methods suffer from artifacts caused by simplifications in their derivation. For example, the limited view of most detection systems is usually not considered. Recently, several iterative reconstruction algorithms have been proposed that are able to approximate the original distribution more accurately. Some of them have been shown to reduce the influence of a limited view. However, the efficiency of these algorithms is usually only assessed on a circular detection setup. Also, acoustic medium heterogeneities are often neglected. We propose a new iterative reconstruction algorithm that can be applied to linear array acquisitions and considers known medium heterogeneities. The approach is based on the Kaczmarz method as commonly used in Computed Tomography.
    No preview · Conference Paper · Sep 2014
  • Martin F Schiffner · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: In pulse-echo ultrasound imaging (PEUI) of soft tissues, the scattered sound field is governed by spatial fluctuations of the two mechanical parameters compressibility and mass density. Spatial fluctuations in compressibility act as isotropic monopole radiators while spatial fluctuations in mass density act as anisotropic dipole radiators. Conventional strategies for linear image reconstruction in PEUI, e.g., delay-and-sum, minimum variance, and synthetic aperture focusing, exclusively account for monopole scattering. This neglect of the inhomogeneous mass density might be accompanied by a loss of diagnostically relevant information, e.g., the detection of tissue abnormalities. In this study, we formulate a linear inverse scattering problem to recover separate, space-resolved maps of the spatial fluctuations in both mechanical parameters from measurements of the scattered acoustic pressure. The physical model accounts for frequency-dependent absorption and dispersion in accordance with the time causal model. The computational costs are effectively reduced by the usage of the fast multipole algorithm. The concept is evaluated using simulated and experimentally obtained radio frequency data.
    No preview · Article · Apr 2014 · The Journal of the Acoustical Society of America
  • Markus C Hesse · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: While established linear pulse-echo ultrasound imaging concepts like B-mode, synthetic aperture (SA) focusing and delay-and-sum (DAS) beamforming solely yield qualitative images showing tissue boundaries, nonlinear reconstruction methods have been proposed to compute quantitative maps of the tissue's material parameter compressibility. In the present contribution, we utilize a previously proposed nonlinear compressibility reconstruction approach and investigate numerically the method's sensitivity to various noise levels in the raw data under cylindrical and plane wave excitation for an unidirectional pulseecho breast imaging application. The reconstruction results demonstrate the approach's robustness up to 30 dB signal-to-noise ratio (SNR) in the data for both excitation modi.
    No preview · Article · Sep 2013 · Biomedizinische Technik/Biomedical Engineering
  • Stefanie Dencks · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: When applying quantitative ultrasound (QUS) measurements to bone for predicting osteoporotic fracture risk, the multipath transmission of sound waves frequently occurs. In the last 10 years, the interest in separating multipath QUS signals for their analysis awoke, and led to the introduction of several approaches. Here, we compare the performances of the two fastest algorithms proposed for QUS measurements of bone: the modified least-squares Prony method (MLSP), and the space alternating generalized expectation maximization algorithm (SAGE) applied in the frequency domain. In both approaches, the parameters of the transfer functions of the sound propagation paths are estimated. To provide an objective measure, we also analytically derive the Cramér-Rao lower bound of variances for any estimator and arbitrary transmit signals. In comparison with results of Monte Carlo simulations, this measure is used to evaluate both approaches regarding their accuracy and precision. Additionally, with simulations using typical QUS measurement settings, we illustrate the limitations of separating two superimposed waves for varying parameters with focus on their temporal separation. It is shown that for good SNRs around 100 dB, MLSP yields better results when two waves are very close. Additionally, the parameters of the smaller wave are more reliably estimated. If the SNR decreases, the parameter estimation with MLSP becomes biased and inefficient. Then, the robustness to noise of the SAGE clearly prevails. Because a clear influence of the interrelation between the wavelength of the ultrasound signals and their temporal separation is observable on the results, these findings can be transferred to QUS measurements at other sites. The choice of the suitable algorithm thus depends on the measurement conditions.
    No preview · Article · Sep 2013 · IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control
  • Source
    Markus C Hesse · Leili Salehi · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: In diagnostic ultrasound imaging, the image reconstruction quality is crucial for reliable diagnosis. Applying reconstruction algorithms based on the acoustic wave equation, the obtained image quality depends significantly on the physical material parameters accounted for in the equation. In this contribution, we extend a proposed iterative nonlinear one-parameter compressibility reconstruction algorithm by the additional reconstruction of the object's inhomogeneous mass density distribution. The improved iterative algorithm is able to reconstruct inhomogeneous maps of the object's compressibility and mass density simultaneously using only one conventional linear transducer array at a fixed location for wave transmission and detection. The derived approach is based on an acoustic wave equation including spatial compressibility and mass density variations, and utilizes the Kaczmarz method for iterative material parameter reconstruction. We validate our algorithm numerically for an unidirectional pulse-echo breast imaging application, and thus generate simulated measurements acquired from a numerical breast phantom with realistic compressibility and mass density values. Applying these measurements, we demonstrate with two reconstruction experiments the necessity to calculate the mass density in case of tissues with significant mass density inhomogeneities. When reconstructing spatial mass density variations, artefacts in the breast's compressibility image are reduced resulting in improved spatial resolution. Furthermore, the compressibility relative error magnitude within a diagnostically significant region of interest (ROI) decreases from 3.04% to 2.62%. Moreover, a second image showing the breast's inhomogeneous mass density distribution is given to provide additional diagnostic information. In the compressibility image, a spatial resolution moderately higher than the classical half-wavelength limit is observed.
    Preview · Article · Aug 2013 · Physics in Medicine and Biology
  • Leili Salehi · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: Nonlinear ultrasound diffraction tomography reconstructs material parameters of a medium from transmitted and scattered sound waves taking into account multiple scattering. A few nonlinear approaches propose algorithms to reconstruct the spatially varying speed of sound (SoS) or alternatively the spatially varying compressibility and mass density. These methods consider longitudinal wave propagation, but not the shear wave propagation which is important in elastic media like in muscle tissue or bone. Here, we present an extension of a nonlinear reconstruction algorithm based on the Kaczmarz method to reconstruct bulk and shear moduli in isotropic solids simultaneously. The morphology of the solid inclusion can be imaged using this approach. This is a first step towards the reconstruction of hard tissues like bone and cartilage possibly enabling the assessment of the cortical thickness of bones.
    No preview · Conference Paper · Jul 2013
  • Markus C. Hesse · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: While established linear pulse-echo ultrasound imaging concepts like synthetic aperture (SA) focusing and delay-and-sum (DAS) beamforming solely image tissue features under single scattering, nonlinear reconstruction methods have been proposed to compute quantitative maps of the tissue's material parameters (e.g. compressibility, mass density, speed of sound) under multiple scattering. In the present contribution, we apply a previously proposed nonlinear simultaneous compressibility and mass density reconstruction algorithm and investigate numerically the image reconstruction quality in contrast to linear SA under cylindrical wave (cw) excitation and linear DAS under plane wave (pw) excitation. Using raw data acquired from a Shepp-Logan phantom (SLP) with typical soft tissue compressibility and mass density values, nonlinear reconstruction using cylindrical wave excitation provides high-resolution images with a mean magnitude of relative error of about 4.27% and 3.18% within a region of interest (ROI) in the compressibility and mass density image, outperforming the image quality reached under plane wave excitation. Applying identical raw data, SA and DAS with both predefined and adapted apodization weights yield less-detailed image reconstructions solely showing tissue boundaries. Furthermore, calculating full width at half maximum (FWHM) resolutions of all methods, the nonlinear approach mainly yields smaller axial and lateral resolutions in contrast to SA and DAS.
    No preview · Conference Paper · Jul 2013
  • Source
    Martin F. Schiffner · Georg Schmitz
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigated the separate recovery of spatial fluctuations in compressibility and mass density by plane wave pulse-echo ultrasound imaging based on sparse recovery (SR). Using simulated radio frequency (RF) data obtained from a sparse and a nonsparse object, we demonstrated that the recovery of space-resolved maps of both material parameters with small relative root mean-squared errors (RMSEs) was feasible. For a signal-to-noise ratio of 40 dB, the relative RMSEs were smaller than 4%. For noiseless RF data, recovery was flawless. Using real RF data acquired from a human common carotid artery (in vivo), the presented concept yielded two space-resolved maps emphasizing different features of the object.
    Full-text · Conference Paper · Jul 2013

Publication Stats

615 Citations
116.67 Total Impact Points

Institutions

  • 2005-2015
    • Ruhr-Universität Bochum
      • Faculty of Electrical Engineering and Information Technology
      Bochum, North Rhine-Westphalia, Germany
  • 2011
    • RWTH Aachen University
      • Department of Experimental Molecular Imaging
      Aachen, North Rhine-Westphalia, Germany
  • 2007
    • Philips
      • Philips Research
      Eindhoven, North Brabant, Netherlands