Loreen Ruhm
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Loreen Ruhm has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
Anke Henning
High‐Field MR Centre Max Planck Institute for Biological Cybernetics Tübingen Germany
7 shared publications- Nikolai Ivanovich Avdievich
Magnetic Resonance Center Max Planck Institute for Biological Cybernetics Tübingen Germany
6 shared publications - Johanna Dorst
Max-Planck-Institute for Biological Cybernetics
4 shared publications - Klaus Scheffler
High‐Field Magnetic Resonance Center, Max Planck Institute for Biological Cybernetics, Tübingen, Germany
3 shared publications - Andrew Martin Wright
Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, Texas, USA
2 shared publications - Anke Henning
University of Texas Southwestern Medical Center
2 shared publications
Latest Publications
- Open Access
- 21 May 2026
- Published by IOS Press
- Vol. 336, 1068-1069
OMOP Extraction of Medical Text Using LLMs: Preliminary Results
We used medium-sized LLMs to extract medication data from German discharge letters into the OMOP CDM, achieving >85% accuracy and >75% F1-score, demonstrating the feasibility of OMOP extraction with medium-sized models.
- 21 May 2026
- Published by IOS Press
- Vol. 336, 1068-1069
Open AccessOMOP Extraction of Medical Text Using LLMs: Preliminary Results
We used medium-sized LLMs to extract medication data from German discharge letters into the OMOP CDM, achieving >85% accuracy and >75% F1-score, demonstrating the feasibility of OMOP extraction with medium-sized models.
- Open Access
Improving the Effective Spatial Resolution in 1H-MRSI of the Prostate with Three-Dimensional Overdiscretized Reconstructions
In in vivo 1H-MRSI of the prostate, small matrix sizes can cause voxel bleeding extending to regions far from a voxel, dispersing a signal of interest outside that voxel and mixing extra-prostatic residual lipid signals into the prostate. To resolve this problem, we developed a three-dimensional overdiscretized reconstruction method. Without increasing the acquisition time from current 3D MRSI acquisition methods, this method is aimed to improve the localization of metabolite signals in the prostate without compromising on SNR. The proposed method consists of a 3D spatial overdiscretization of the MRSI grid, followed by noise decorrelation with small random spectral shifts and weighted spatial averaging to reach a final target spatial resolution. We successfully applied the three-dimensional overdiscretized reconstruction method to 3D prostate 1H-MRSI data at 3T. Both in phantom and in vivo, the method proved to be superior to conventional weighted sampling with Hamming filtering of k-space. Compared with the latter, the overdiscretized reconstructed data with smaller voxel size showed up to 10% less voxel bleed while maintaining higher SNR by a factor of 1.87 and 1.45 in phantom measurements. For in vivo measurements, within the same acquisition time and without loss of SNR compared with weighted k-space sampling and Hamming filtering, we achieved increased spatial resolution and improved localization in metabolite maps.
Open AccessImproving the Effective Spatial Resolution in 1H-MRSI of the Prostate with Three-Dimensional Overdiscretized Reconstructions
In in vivo 1H-MRSI of the prostate, small matrix sizes can cause voxel bleeding extending to regions far from a voxel, dispersing a signal of interest outside that voxel and mixing extra-prostatic residual lipid signals into the prostate. To resolve this problem, we developed a three-dimensional overdiscretized reconstruction method. Without increasing the acquisition time from current 3D MRSI acquisition methods, this method is aimed to improve the localization of metabolite signals in the prostate without compromising on SNR. The proposed method consists of a 3D spatial overdiscretization of the MRSI grid, followed by noise decorrelation with small random spectral shifts and weighted spatial averaging to reach a final target spatial resolution. We successfully applied the three-dimensional overdiscretized reconstruction method to 3D prostate 1H-MRSI data at 3T. Both in phantom and in vivo, the method proved to be superior to conventional weighted sampling with Hamming filtering of k-space. Compared with the latter, the overdiscretized reconstructed data with smaller voxel size showed up to 10% less voxel bleed while maintaining higher SNR by a factor of 1.87 and 1.45 in phantom measurements. For in vivo measurements, within the same acquisition time and without loss of SNR compared with weighted k-space sampling and Hamming filtering, we achieved increased spatial resolution and improved localization in metabolite maps.
- Open Access
Mapping of Glutamate Metabolism using 1H FID-MRSI after oral Administration of [1- 13 C]Glc at 9.4 T
Glutamate is the major excitatory transmitter in the brain and malfunction of the related metabolism is associated with various neurological diseases and disorders. The observation of labeling changes in the spectra after the administration of a 13 C labelled tracer is a common tool to gain better insights into the function of the metabolic system. But so far, only a very few studies presenting the labelling effects in more than two voxels to show the spatial dependence of metabolism. In the present work, the labeling effects were measured in a transversal plane in the human brain using ultra-short TE and TR 1H FID-MRSI. The measurement set-up was most simple: The [1- 13 C]Glc was administered orally instead of intravenous and the spectra were measured with a pure 1H technique without the need of a 13 C channel (as Boumezbeur et al. demonstrated in 2004). Thus, metabolic maps and enrichment curves could be obtained for more metabolites and in more voxels than ever before in human brain. Labeling changes could be observed in [4- 13 C]glutamate, [3- 13 C]glutamate+glutamine, [2- 13 C]glutamate+glutamine, [4- 13 C]glutamine, [2,3- 13 C]aspartate and [6- 13 C]N-acetyl aspartate with a high temporal (3.6 min) and spatial resolution (32×32 grid with nominal voxel size of 0.33 μL) in five volunteers.
Open AccessMapping of Glutamate Metabolism using 1H FID-MRSI after oral Administration of [1- 13 C]Glc at 9.4 T
Glutamate is the major excitatory transmitter in the brain and malfunction of the related metabolism is associated with various neurological diseases and disorders. The observation of labeling changes in the spectra after the administration of a 13 C labelled tracer is a common tool to gain better insights into the function of the metabolic system. But so far, only a very few studies presenting the labelling effects in more than two voxels to show the spatial dependence of metabolism. In the present work, the labeling effects were measured in a transversal plane in the human brain using ultra-short TE and TR 1H FID-MRSI. The measurement set-up was most simple: The [1- 13 C]Glc was administered orally instead of intravenous and the spectra were measured with a pure 1H technique without the need of a 13 C channel (as Boumezbeur et al. demonstrated in 2004). Thus, metabolic maps and enrichment curves could be obtained for more metabolites and in more voxels than ever before in human brain. Labeling changes could be observed in [4- 13 C]glutamate, [3- 13 C]glutamate+glutamine, [2- 13 C]glutamate+glutamine, [4- 13 C]glutamine, [2,3- 13 C]aspartate and [6- 13 C]N-acetyl aspartate with a high temporal (3.6 min) and spatial resolution (32×32 grid with nominal voxel size of 0.33 μL) in five volunteers.
- 5 September 2022
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 89 (1) , 29-39
Deuterium metabolic imaging of the human brain in vivo at 7 T
Purpose: To explore the potential of deuterium metabolic imaging (DMI) in the human brain in vivo at 7 T, using a multi‐element deuterium (2H) RF coil for 3D volume coverage.Methods: 1H‐MR images and localized 2H MR spectra were acquired in vivo in the human brain of 3 healthy subjects to generate DMI maps of 2H‐labeled water, glucose, and glutamate/glutamine (Glx). In addition, non‐localized 2H‐MR spectra were acquired both in vivo and in vitro to determine T1 and T2 relaxation times of deuterated metabolites at 7 T. The performance of the 2H coil was assessed through numeric simulations and experimentally acquired B1+ maps.Results: 3D DMI maps covering the entire human brain in vivo were obtained from well‐resolved deuterated (2H) metabolite resonances of water, glucose, and Glx. The T1 and T2 relaxation times were consistent with those reported at adjacent field strengths. Experimental B1+ maps were in good agreement with simulations, indicating efficient and homogeneous B1+ transmission and low RF power deposition for 2H, consistent with a similar array coil design reported at 9.4 T.Conclusion: Here, we have demonstrated the successful implementation of 3D DMI in the human brain in vivo at 7 T. The spatial and temporal nominal resolutions achieved at 7 T (i.e., 2.7 mL in 28 min, respectively) were close to those achieved at 9.4 T and greatly outperformed DMI at lower magnetic fields. DMI at 7 T and beyond has clear potential in applications dealing with small brain lesions.
- 5 September 2022
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 89 (1) , 29-39
Deuterium metabolic imaging of the human brain in vivo at 7 T
Purpose: To explore the potential of deuterium metabolic imaging (DMI) in the human brain in vivo at 7 T, using a multi‐element deuterium (2H) RF coil for 3D volume coverage.Methods: 1H‐MR images and localized 2H MR spectra were acquired in vivo in the human brain of 3 healthy subjects to generate DMI maps of 2H‐labeled water, glucose, and glutamate/glutamine (Glx). In addition, non‐localized 2H‐MR spectra were acquired both in vivo and in vitro to determine T1 and T2 relaxation times of deuterated metabolites at 7 T. The performance of the 2H coil was assessed through numeric simulations and experimentally acquired B1+ maps.Results: 3D DMI maps covering the entire human brain in vivo were obtained from well‐resolved deuterated (2H) metabolite resonances of water, glucose, and Glx. The T1 and T2 relaxation times were consistent with those reported at adjacent field strengths. Experimental B1+ maps were in good agreement with simulations, indicating efficient and homogeneous B1+ transmission and low RF power deposition for 2H, consistent with a similar array coil design reported at 9.4 T.Conclusion: Here, we have demonstrated the successful implementation of 3D DMI in the human brain in vivo at 7 T. The spatial and temporal nominal resolutions achieved at 7 T (i.e., 2.7 mL in 28 min, respectively) were close to those achieved at 9.4 T and greatly outperformed DMI at lower magnetic fields. DMI at 7 T and beyond has clear potential in applications dealing with small brain lesions.
- Open Access
- 29 June 2022
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 88 (4) , 1912-1926
A 32‐element loop/dipole hybrid array for human head imaging at 7 T
Purpose: To improve whole‐brain SNR at 7 Tesla, a novel 32‐element hybrid human head array coil was developed, constructed, and tested.Methods: Our general design strategy is based on 2 major ideas: Firstly, following suggestions of previous works based on the ultimate intrinsic SNR theory, we combined loops and dipoles for improvement of SNR near the head center. Secondly, we minimized the total number of array elements by using a hybrid combination of transceive (TxRx) and receive (Rx) elements. The new hybrid array consisted of 8 folded‐end TxRx‐dipole antennas and 3 rows of 24 Rx‐loops all placed in a single layer on the surface of a tight‐fit helmet.Results: The developed array significantly improved SNR in vivo both near the center (∼20%) and at the periphery (∼20% to 80%) in comparison to a common commercial array coil with 8 transmit (Tx) and 32 Rx‐elements. Whereas 24 loops alone delivered central SNR very similar to that of the commercial coil, the addition of complementary dipole structures provided further improvement. The new array also provided ∼15% higher Tx efficiency and better longitudinal coverage than that of the commercial array.Conclusion: The developed array coil demonstrated advantages in combining complementary TxRx and Rx resonant structures, that is, TxRx‐dipoles and Rx‐loops all placed in a single layer at the same distance to the head. This strategy improved both SNR and Tx‐performance, as well as simplified the total head coil design, making it more robust.
- 29 June 2022
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 88 (4) , 1912-1926
Open AccessA 32‐element loop/dipole hybrid array for human head imaging at 7 T
Purpose: To improve whole‐brain SNR at 7 Tesla, a novel 32‐element hybrid human head array coil was developed, constructed, and tested.Methods: Our general design strategy is based on 2 major ideas: Firstly, following suggestions of previous works based on the ultimate intrinsic SNR theory, we combined loops and dipoles for improvement of SNR near the head center. Secondly, we minimized the total number of array elements by using a hybrid combination of transceive (TxRx) and receive (Rx) elements. The new hybrid array consisted of 8 folded‐end TxRx‐dipole antennas and 3 rows of 24 Rx‐loops all placed in a single layer on the surface of a tight‐fit helmet.Results: The developed array significantly improved SNR in vivo both near the center (∼20%) and at the periphery (∼20% to 80%) in comparison to a common commercial array coil with 8 transmit (Tx) and 32 Rx‐elements. Whereas 24 loops alone delivered central SNR very similar to that of the commercial coil, the addition of complementary dipole structures provided further improvement. The new array also provided ∼15% higher Tx efficiency and better longitudinal coverage than that of the commercial array.Conclusion: The developed array coil demonstrated advantages in combining complementary TxRx and Rx resonant structures, that is, TxRx‐dipoles and Rx‐loops all placed in a single layer at the same distance to the head. This strategy improved both SNR and Tx‐performance, as well as simplified the total head coil design, making it more robust.
- Open Access
Dynamic observation of 2 H labeled compounds in the human brain with 1 H versus 2 H magnetic resonance spectroscopy at 9.4T
The metabolic pathway of [6,6’- 2 H 2 ]-labeled glucose was investigated with two different techniques. The first technique used direct detection of deuterium applying Deuterium Metabolic Imaging (DMI). The second technique used the indirect detection of deuterium with proton MR spectroscopy (MRS) called Quantitative Exchanged-label Turnover (QELT) MRS. For the first time, time-resolved data was acquired for both techniques in the same healthy human subjects and directly compared. The time-curves were used in a kinetic model to estimate rates of the metabolic pathway of glucose. Two different kinetic models were compared. One included only DMI data, the second one combined DMI and QELT. For the first model, a tricarboxylic acid (TCA) cycle rate of 0.69 ± 0.10 μmol·min -1 ·g -1 was determined. For the second model, the estimated TCA cycle rate was 0.68 ± 0.12 μmol·min -1 ·g -1 . In addition, the rate of glutamine synthesis from glutamate could be estimated with model 2 (0.51 ± 0.15 μmol·min -1 ·g -1 ). The sensitivity of both methods was evaluated and compared to alternative techniques.
Open AccessDynamic observation of 2 H labeled compounds in the human brain with 1 H versus 2 H magnetic resonance spectroscopy at 9.4T
The metabolic pathway of [6,6’- 2 H 2 ]-labeled glucose was investigated with two different techniques. The first technique used direct detection of deuterium applying Deuterium Metabolic Imaging (DMI). The second technique used the indirect detection of deuterium with proton MR spectroscopy (MRS) called Quantitative Exchanged-label Turnover (QELT) MRS. For the first time, time-resolved data was acquired for both techniques in the same healthy human subjects and directly compared. The time-curves were used in a kinetic model to estimate rates of the metabolic pathway of glucose. Two different kinetic models were compared. One included only DMI data, the second one combined DMI and QELT. For the first model, a tricarboxylic acid (TCA) cycle rate of 0.69 ± 0.10 μmol·min -1 ·g -1 was determined. For the second model, the estimated TCA cycle rate was 0.68 ± 0.12 μmol·min -1 ·g -1 . In addition, the rate of glutamine synthesis from glutamate could be estimated with model 2 (0.51 ± 0.15 μmol·min -1 ·g -1 ). The sensitivity of both methods was evaluated and compared to alternative techniques.
- Open Access
31 P Transversal Relaxation Times and Metabolite Concentrations in the Human Brain at 9.4T
A method to estimate phosphorus ( 31 P) transversal relaxation times (T 2 ) of coupled spin systems is demonstrated. Additionally, intracellular and extracellular pH (pH ext , pH int ) and relaxation corrected metabolite concentrations are reported. Echo time (TE) series of 31 P metabolite spectra were acquired using STEAM localization. Spectra were fitted using LCModel with accurately modeled Vespa basis sets accounting for J-evolution of the coupled spin systems. T 2 s were estimated by fitting a single exponential two-parameter model across the TE series. Fitted inorganic phosphate frequencies were used to calculate pH, and relaxation times were used to determine the brain metabolite concentrations. The method was demonstrated in the healthy human brain at a field strength of 9.4T. T 2 relaxation times of ATP and NAD are the shortest between 8 ms and 20 ms, followed by T 2 s of inorganic phosphate between 25 ms and 50 ms, and PCr with a T 2 of 100 ms. Phosphomonoesters and –diesters have the longest T 2 s of about 130 ms. Measured T 2 s are comparable to literature values and fit in a decreasing trend with increasing field strengths. Calculated pHs and metabolite concentrations are also comparable to literature values.
Open Access31 P Transversal Relaxation Times and Metabolite Concentrations in the Human Brain at 9.4T
A method to estimate phosphorus ( 31 P) transversal relaxation times (T 2 ) of coupled spin systems is demonstrated. Additionally, intracellular and extracellular pH (pH ext , pH int ) and relaxation corrected metabolite concentrations are reported. Echo time (TE) series of 31 P metabolite spectra were acquired using STEAM localization. Spectra were fitted using LCModel with accurately modeled Vespa basis sets accounting for J-evolution of the coupled spin systems. T 2 s were estimated by fitting a single exponential two-parameter model across the TE series. Fitted inorganic phosphate frequencies were used to calculate pH, and relaxation times were used to determine the brain metabolite concentrations. The method was demonstrated in the healthy human brain at a field strength of 9.4T. T 2 relaxation times of ATP and NAD are the shortest between 8 ms and 20 ms, followed by T 2 s of inorganic phosphate between 25 ms and 50 ms, and PCr with a T 2 of 100 ms. Phosphomonoesters and –diesters have the longest T 2 s of about 130 ms. Measured T 2 s are comparable to literature values and fit in a decreasing trend with increasing field strengths. Calculated pHs and metabolite concentrations are also comparable to literature values.
- Open Access
- 1 December 2021
- Published by Elsevier in NeuroImage
- Vol. 244, 118639
Deuterium metabolic imaging in the human brain at 9.4 Tesla with high spatial and temporal resolution
- 1 December 2021
- Published by Elsevier in NeuroImage
- Vol. 244, 118639
Open AccessDeuterium metabolic imaging in the human brain at 9.4 Tesla with high spatial and temporal resolution
- Open Access
- 4 July 2021
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 86 (5) , 2368-2383
3D 31P MRSI of the human brain at 9.4 Tesla: Optimization and quantitative analysis of metabolic images
Purpose: To present 31P whole brain MRSI with a high spatial resolution to probe quantitative tissue analysis of 31P MRSI at an ultrahigh field strength of 9.4 Tesla.Methods: The study protocol included a 31P MRSI measurement with an effective resolution of 2.47 mL. For SNR optimization, the nuclear Overhauser enhancement at 9.4 Tesla was investigated. A sensitivity correction was achieved by applying a low rank approximation of the γ‐adenosine triphosphate signal. Group analysis and regression on individual volunteers were performed to investigate quantitative concentration differences between different tissue types.Results: Differences in gray and white matter tissue 31P concentrations could be investigated for 12 different 31P resonances. In addition, the first highly resolved quantitative MRSI images measured at B0 = 9.4 Tesla of 31P detectable metabolites with high SNR could be presented.Conclusion: With an ultrahigh field strength B0 = 9.4 Tesla, 31P MRSI moves further toward quantitative metabolic imaging, and subtle differences in concentrations between different tissue types can be detected.
- 4 July 2021
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 86 (5) , 2368-2383
Open Access3D 31P MRSI of the human brain at 9.4 Tesla: Optimization and quantitative analysis of metabolic images
Purpose: To present 31P whole brain MRSI with a high spatial resolution to probe quantitative tissue analysis of 31P MRSI at an ultrahigh field strength of 9.4 Tesla.Methods: The study protocol included a 31P MRSI measurement with an effective resolution of 2.47 mL. For SNR optimization, the nuclear Overhauser enhancement at 9.4 Tesla was investigated. A sensitivity correction was achieved by applying a low rank approximation of the γ‐adenosine triphosphate signal. Group analysis and regression on individual volunteers were performed to investigate quantitative concentration differences between different tissue types.Results: Differences in gray and white matter tissue 31P concentrations could be investigated for 12 different 31P resonances. In addition, the first highly resolved quantitative MRSI images measured at B0 = 9.4 Tesla of 31P detectable metabolites with high SNR could be presented.Conclusion: With an ultrahigh field strength B0 = 9.4 Tesla, 31P MRSI moves further toward quantitative metabolic imaging, and subtle differences in concentrations between different tissue types can be detected.
- Open Access
- 25 February 2021
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 86 (1) , 581-597
Unshielded bent folded‐end dipole 9.4 T human head transceiver array decoupled using modified passive dipoles
Purpose: To develop an unshielded dipole transceiver array for human head imaging at 9.4 Tesla and to improve decoupling of adjacent dipole elements, a novel array design with modified passive dipole antennas was developed, evaluated, and tested.Methods: The new array consisted of 8 bent folded‐end dipole elements placed in a single row and surrounding the head. Adjacent elements of RF transceiver arrays are usually decoupled by introducing circuits electrically connected to elements. These methods are difficult to use for dipole arrays because of the distant location of the adjacent antennas. A recently developed decoupling technique using passive dipoles is simple and does not require any electrical connection. However, common parallel passive dipoles can produce destructive interference with the RF field of the array itself. To minimize this interference, we placed the passive dipoles perpendicularly to the active dipoles and positioned them at the ends of the array. We also evaluated the effect of different passive dipoles on the array transmit performance. Finally, we optimized the array transmit performance by varying the length of the dipole folded portion.Results: By rotating the passive dipoles 90º and moving them toward the ends of the array, we minimized the destructive interference to an acceptable level without compromising decoupling and the transmit efficiency.Conclusion: While keeping the benefits of the passive dipole decoupling method, the new modified dipoles produce substantially less destructive interference with the RF field of the array than the common design. The constructed transceiver array demonstrated good decoupling and whole‐brain coverage.
- 25 February 2021
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 86 (1) , 581-597
Open AccessUnshielded bent folded‐end dipole 9.4 T human head transceiver array decoupled using modified passive dipoles
Purpose: To develop an unshielded dipole transceiver array for human head imaging at 9.4 Tesla and to improve decoupling of adjacent dipole elements, a novel array design with modified passive dipole antennas was developed, evaluated, and tested.Methods: The new array consisted of 8 bent folded‐end dipole elements placed in a single row and surrounding the head. Adjacent elements of RF transceiver arrays are usually decoupled by introducing circuits electrically connected to elements. These methods are difficult to use for dipole arrays because of the distant location of the adjacent antennas. A recently developed decoupling technique using passive dipoles is simple and does not require any electrical connection. However, common parallel passive dipoles can produce destructive interference with the RF field of the array itself. To minimize this interference, we placed the passive dipoles perpendicularly to the active dipoles and positioned them at the ends of the array. We also evaluated the effect of different passive dipoles on the array transmit performance. Finally, we optimized the array transmit performance by varying the length of the dipole folded portion.Results: By rotating the passive dipoles 90º and moving them toward the ends of the array, we minimized the destructive interference to an acceptable level without compromising decoupling and the transmit efficiency.Conclusion: While keeping the benefits of the passive dipole decoupling method, the new modified dipoles produce substantially less destructive interference with the RF field of the array than the common design. The constructed transceiver array demonstrated good decoupling and whole‐brain coverage.