Constantin von Deuster

Constantin von Deuster

Auto-generated profile, based on publicly available information

Constantin von Deuster

Swiss Innovation Hub (SIH), Siemens Healthineers International AG Zurich Switzerland

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 July 2020 in Magnetic Resonance in Medicine
Verified
Journal Article
Closed access
PDF file
GPGian Franco Piredda
Gian Franco Piredda
THTom Hilbert
Tom Hilbert
Erick Jorge Canales-RodríguezErick Jorge Canales-Rodríguez
Erick Jorge Canales-Rodríguez
MPMarco Pizzolato
Marco Pizzolato
CDConstantin von Deuster
Constantin von Deuster
TKTobias Kober
Tobias Kober

12 March 2021 in Investigative Radiology
Verified
Journal Article
Closed access
CDConstantin von Deuster
Constantin von Deuster
SSStefan Sommer
Stefan Sommer
CGChristoph Germann
Christoph Germann
NHNatalie Msc Hinterholzer
Natalie Msc Hinterholzer
RHRobin M. Heidemann
Robin M. Heidemann
DNDaniel Nanz
Daniel Nanz
Objectives: Through-slice chemical shift artifacts in state-of-the-art turbo-spin-echo (TSE) images can be significantly more severe at 7 T than at lower field strengths. In musculoskeletal applications, these artifacts appear similar to bone fractures or neoplastic bone marrow disease. The objective of this work was to explore and reduce through-slice chemical shift artifacts in 2-dimensional (2D) TSE imaging at 7 T. Materials and Methods: This prospective study was approved by the local ethics board. The bandwidths of the excitation and refocusing radiofrequency (RF) pulses of a prototype 2D TSE sequence were individually modified and their effect on the slice profiles and relative slice locations of water and fat spins was assessed in an oil-water phantom. Based on these results, it was hypothesized that the combination of matched and increased excitation and refocusing RF pulse bandwidths (“MIB”) of 1500 Hz would enable 2D TSE imaging with significantly reduced chemical shift artifacts compared with a state-of-the-art sequence with unmatched and moderate RF pulse bandwidths (“UMB”) of 1095 and 682 Hz. A series of T1-weighted sagittal knee examinations in 10 healthy human subjects were acquired using the MIB and UMB sequences and independently evaluated by 2 radiologists. They measured the width of chemical shift artifacts at 2 standardized locations and graded the perceived negative effect of chemical shift artifacts on image quality in the bones and in the whole gastrocnemius muscle on a 5-point scale. Similar knee, wrist, and foot images were acquired in a single subject. Signal-to-noise ratios in the femoral bone marrow were computed between the UMB and MIB sequences. Results: Phantom measurements confirmed the expected spatial separation of simultaneously affected water and fat slices between 40% and 200% of the prescribed slice thickness for RF pulse bandwidths between 2500 and 500 Hz. Through-slice chemical shift artifacts at the bone-cartilage interface were significantly smaller with MIB than with UMB (location 1: 0.35 ± 0.20 mm vs 1.27 ± 0.27 mm, P < 0.001; location 2: 0.25 ± 0.13 mm vs 1.48 ± 0.46 mm, P < 0.001; intraclass correlation coefficient = 0.98). The negative effect of chemical shift artifacts on image quality was significantly smaller with MIB than with UMB (bone: 2 ± 0 vs 4 ± 1, P < 0.004 [both readers]; muscle: 3 ± 0 vs 2 ± 0, P < 0.004 [both readers]; κ = 0.69). The signal-to-noise ratio of the UMB and MIB sequences was comparable, with a ratio of 99 ± 7%. Images acquired using the UMB sequence displayed numerous artifactual hyperintensities and diffuse, as well as locally severe, fat signal loss in all examined regions, whereas the MIB sequence consistently yielded high image quality with bright T1-weighted fat signal and excellent depiction of fine tissue structures. Conclusions: On 7 T systems, the selection of high and matched RF bandwidths for excitation and refocusing pulses for 2D TSE imaging without fat suppression showed consistently better image quality than state-of-the-art sequences with unmatched lower RF pulse bandwidths.

11 July 2026 in Journal of Magnetic Resonance Imaging
Verified
Journal Article
Open access
PDF file
JDJeanette C. Deck
Jeanette C. Deck
SGSophia Samira Goller
Sophia Samira Goller
GKGeorg W. Kajdi
Georg W. Kajdi
AMAdrian A. Marth
Adrian A. Marth
PZPatrick Zingg
Patrick Zingg
RSReto Sutter
Reto Sutter
Background: Slice encoding for metal artifact correction (SEMAC) is used in MRI for total hip arthroplasties for metal artifact reduction. However, ripple artifacts often arise near the implant, impairing diagnosis in affected regions. Hypothesis: Applying a wavelet domain filter (WD‐Filter) might reduce ripple artifacts and improve image quality. Study Type: Retrospective. Subjects: 100 patients with primary total hip arthroplasty (female: 51, mean age: (64 ± 13) years; male: 49, mean age: (65 ± 13) years) who underwent clinically indicated MRI using compressed sensing SEMAC. Field Strength/Sequences: 1.5 T; compressed sensing‐accelerated turbo spin‐echo (TSE)‐based SEMAC sequence. Assessment: Three readers rated 250 ripple artifacts before and after applying the WD‐filter. 4‐point Likert scales were used for the intensity of ripple artifacts (Likert‐RA, “none”: 1, “mild”: 2, “moderate”: 3, “severe”: 4) and the image quality (Likert‐IQ, “poor”: 1, “moderate”: 2, “good”: 3, “excellent”: 4). The relative reduction of the signal intensity standard deviation () was measured in a region of interest (ROI) containing the ripple artifact. Statistical Tests: Inter‐reader agreement was assessed with Fleiss' kappa, with 95% confidence intervals. Differences between pre‐ and post‐filter ratings (Likert‐RA and Likert‐IQ) were evaluated using the Wilcoxon signed‐rank test, while was compared using a two‐sided paired t ‐test. A p value of less than 0.05 indicated statistical significance. Results: Prior to WD‐Filter application, ripple artifacts were rated as “severe”, as “moderate”, and as “mild”. After filtering, artifacts were rated as “none”, and as “mild”. Upon filter application, Likert‐RA and Likert‐IQ significantly improved from to and to , respectively, with a large effect size ( r > 0.89) and “almost perfect” inter‐rater agreement (kappa range: 0.86–0.96). The filtered images showed a significantly reduced compared to the unfiltered images (relative change: ). Data Conclusion: The WD‐Filter effectively reduced ripple artifacts in MRI of total hip arthroplasty and improved image quality. Evidence Level: 3. Technical Efficacy: Stage 3.

24 July 2026 in Investigative Radiology
Verified
Journal Article
Open access
FAFrederik Abel
Frederik Abel
CDConstantin von Deuster
Constantin von Deuster
AWAndreas J. Walch
Andreas J. Walch
DPDominik Paul
Dominik Paul
MNMarcel Dominik Nickel
Marcel Dominik Nickel
RSReto Sutter
Reto Sutter
Objective: To evaluate the potential of combined coherent and incoherent undersampling with deep learning (DL) reconstruction for highly accelerated high-resolution three-dimensional (3D) double-echo steady-state (DESS) MRI of the cervical nerves and rootlets at 7T. Methods: In this prospective study, asymptomatic volunteers underwent 7T MRI of the cervical spine. A high-resolution, 3D DESS sequence (reconstructed voxel size 0.35 mm³ isotropic) with regular 2-fold GRAPPA undersampling served as reference. The sequence was repeated with coherent CAIPIRINHA undersampling alone and combined CAIPIRINHA + incoherent compressed sensing (CS)-based undersampling with DL reconstruction and varying acceleration factors (DL-CAIPIRINHA: R=2 to 12 and DL-CS-CAIPI: R=4 to 16), while all other imaging parameters remained constant. Two fellowship-trained musculoskeletal radiologists assessed image quality, noise, reconstruction and motion artifacts, and edge sharpness of the spinal cord, intradural rootlets/roots, and exiting spinal nerves, including dorsal root ganglia (DRGs), using 5-point Likert scales, with higher scores indicating better image quality or fewer artifacts. Statistics included Friedman and post hoc Wilcoxon signed-rank tests (Benjamini-Hochberg corrected) and κ statistics. Results: Thirty-two volunteers (mean age 30.9±7.2 y; mean BMI 23.4±2.8 kg/m²; 16 females) were included. Image quality significantly improved with 2- and 4-fold DL-CAIPIRINHA (means 4.4 to 4.6) and 4- and 8-fold DL-CS-CAIPI (3.8 to 3.9) compared with GRAPPA (3.2; all P <0.001), and remained comparable with 12-fold DL-CS-CAIPI ( P =0.42) despite more than 6-fold reduction in scan time. All DL-based reconstructions reduced noise ( P <0.001). Reconstruction artifacts were more pronounced for 12-fold DL-CAIPIRINHA (1.6) than for 12- and 16-fold DL-CS-CAIPI (3.1 to 3.3; P <0.001). Motion artifacts decreased with acceleration factors ≥4 ( P <0.001). Sharpness of the spinal cord, intraspinal rootlets/roots, and DRGs was highest with 4-fold DL-CAIPIRINHA (4.2 to 4.7) vs. GRAPPA (2.9 to 3.2; P <0.001). Inter-reader agreement was almost perfect across all imaging features (κ=0.85 to 0.95). Conclusions: Coherent CAIPIRINHA undersampling with DL reconstruction improves image quality over conventional GRAPPA while enabling 50% faster imaging in 7T cervical spine MRI. Adding incoherent CS undersampling permits substantially higher acceleration with preserved image quality and fewer reconstruction artifacts. The novel integration of coherent and incoherent undersampling via combined CAIPIRINHA and CS with DL reconstruction provides a robust framework for preserving image quality at acceleration factors beyond 8, enabling efficient high-resolution cervical spine MRI with scan time reductions exceeding 6-fold.