Jonathan Stelter

Purpose: To develop and validate a novel self‐gated motion‐robust 3D T1‐weighted gradient echo imaging framework based on a Cartesian acquisition with spiral profile ordering (CASPR) for abdominal imaging under unstable breath‐hold conditions. Methods: The CASPR trajectory was evaluated using frequency encoding in the anterior‐to‐posterior (AP) direction for respiratory motion robustness. The CASPR's intrinsic k‐space center oversampling was exploited to directly extract a self‐gated respiratory signal from the acquired data. A soft‐gating motion compensation framework was integrated into the reconstruction pipeline. The method was validated in 10 healthy volunteers (five males and five females) and compared to a Cartesian acquisition with pseudorandom undersampling. Results: Both CASPR and Cartesian acquisition with pseudorandom undersampling exhibited motion artifacts under unstable breath‐hold conditions without retrospective correction. The proposed self‐gated CASPR framework enabled reliable extraction of the respiratory signal directly from the AP readout. Soft gating significantly improved the image quality of motion‐corrupted scans. Conclusion: This study introduces a novel self‐gated CASPR‐based framework for respiratory motion compensation in 3D abdominal MRI, in cases of incomplete and delayed breath‐holds. The method enables the extraction of the self‐navigator signal along the anterior‐to‐posterior axis and compensates for motion using a soft‐gating technique.

Brown adipose tissue (BAT) is a metabolically active tissue in humans, located primarily within the supraclavicular adipose tissue (scAT), that can be activated by cold or high‐caloric meal consumption. While the changes of proton density fat fraction (PDFF) upon cold activation are well investigated, there is a knowledge gap regarding PDFF and dynamics upon meal‐induced BAT activation. Therefore, the purpose of this study was to establish a robust pipeline to quantify PDFF and changes in scAT in fasting control and meal intervention experiments. To that end, a motion‐robust radial stack‐of‐stars trajectory has been employed with a regularized iterative reconstruction algorithm, graph‐cut‐based water‐fat separation, affine image registration and automatic neural‐network‐based region of interest (ROI) segmentation. Next to scAT, ROIs of subcutaneous adipose tissue (SAT) and muscle have been created to compare the results to reference tissues. Multi‐echo gradient echo acquisitions were repeated in a total of ten volunteers before a break, in which the volunteers consumed an iso‐caloric meal in the meal intervention experiment. After the break, repeated multi‐echo gradient echo measurements were used to dynamically track the changes of both PDFF and . In the meal intervention experiment, scAT PDFF showed a stronger decrease of compared to the fasting control experiment, in which the decrease below baseline was similar to that in SAT. Furthermore, scAT exhibited a strong decrease to below baseline after meal intervention, while no decrease was observed in the fasting control experiment or in SAT. Considering the observed dynamic effects and their differences to the fasting control experiment, a PDFF decrease of and a decrease of could be attributed to meal‐induced activation of scAT.

Purpose: To develop and validate a repeatable and reproducible approach, QuantoRAGE, for simultaneous whole‐brain T ₁ and T ₂ mapping using adiabatic magnetization preparation. Methods: QuantoRAGE is a 3D FLASH‐based sequence using an adiabatic T ₂ ‐prepared inversion followed by two readout blocks. Sequence repetitions with different preparation durations and inversion times provide T ₁ ‐ and T ₂ ‐weighted images. Quantitative parameters are voxel‐wise estimated by matching signal evolutions to a dictionary generated from extended phase graph simulations. To accelerate the method, the protocol was optimized using Cramér‐Rao lower bound analysis, and a neural‐network was trained on sequence‐specific dictionaries for fast matching. The optimized protocol was evaluated in phantom and healthy volunteers at 3 T and compared against reference MP2RAGE T ₁ and multi‐echo spin‐echo T ₂ measurements. In vivo repeatability and reproducibility were quantified using scan‐rescan experiments. Feasibility of QuantoRAGE at 7 T was tested in vivo. Results: At 3 T, T ₁ and T ₂ phantom quantification agreed with nominal values (T ₁ : r = 0.98, T ₂ : r = 0.99). In eight healthy volunteers, high repeatability (bias ± SD, T ₁ : 14.1 ± 30.2 ms, T ₂ : 1.2 ± 1.7 ms) and reproducibility (T ₁ : 18.9 ± 32.6 ms, T ₂ : 1.1 ± 2.1 ms) were observed across brain regions, with T ₁ estimates comparable to reference methods ( r = 0.96, slope = 0.74), and T ₂ underestimated ( r = 0.79, slope = 0.44). Fast matching produced results consistent with dictionary‐matching and significantly reduced parameter estimation time. At 7 T, the adiabatic preparation resulted in homogeneous T ₁ and T ₂ maps at 0.8‐mm isotropic resolution. Conclusion: QuantoRAGE enabled reproducible simultaneous T ₁ and T ₂ mapping at 3 T, providing an efficient framework for longitudinal and cross‐sectional quantitative MRI studies and showed feasibility at 7 T.

A free‐breathing six‐echo radial stack‐of‐stars acquisition is evaluated as an alternative to Cartesian breath‐hold scans for pancreatic fat quantification at 3T. Pancreatic fat deposition was assessed in 18 participants with obesity enrolled in a randomized controlled lifestyle intervention study (LION) using both a Cartesian breath‐hold scan and a free‐breathing high‐resolution stack‐of‐stars sequence (scan times: 16 s vs. 2 min 55 s). Pancreatic segmentations were initially generated using the VibeSegmentator and subsequently refined manually to reduce the amount of included visceral fat. The Intra‐observer reproducibility of the manual refinement was assessed after 3 months using Dice similarity score. SNR for both scan methods was quantified using a large water phantom. Mean segmented volume and mean pancreatic proton density fat fraction (PDFF) values between radial and Cartesian scans were compared. PDFF and volume changes between baseline and post‐intervention were investigated for both techniques. Segmented pancreatic volumes and PDFF measurements demonstrated excellent agreement between the radial and Cartesian sequences (Volume: R ² = 0.97, ICC = 0.96; PDFF: R ² = 0.98, ICC = 0.98). Compared to the Cartesian scans, the radial scans exhibited a minor positive volume bias at higher pancreatic volumes (linear regression: 0.85 x + 9.4 cm ³ ; mean volume bias: 5.8 cm ³ , 95% LoA: −7.1 to 18.7 cm ³ ), alongside a proportional PDFF bias leading to slight overestimation at lower values and underestimation at higher values (linear regression: 0.88 x + 2.1%; mean PDFF bias −0.2%, 95% LoA: −3.0 to 2.7%). Post‐intervention, both datasets captured significant decreases in pancreatic PDFF (mean loss 2.3% ± 2.1%, Cartesian p < 0.001, radial p < 0.01) and volume (mean loss 8.4 cm ³ ± 7.7 cm ³ ; both p < 0.001). The present study demonstrates the feasibility of a free‐breathing radial stack‐of‐stars sequence for pancreatic fat quantification at 3T and its application in adults with obesity undergoing diet‐induced weight loss.

Background/Objectives: Lung nodules present a common diagnostic challenge, particularly when benign and malignant lesions exhibit similar imaging characteristics. Standard evaluation relies on computed tomography (CT), positron emission tomography (PET), or biopsy, all of which have limitations. Quantitative magnetic resonance (MR) relaxometry using native longitudinal relaxation time (T1) and transverse relaxation time (T2) mapping offers a radiation-free alternative reflecting tissue-specific differences. Methods: This prospective, single-center study included 64 patients with 76 histologically or radiologically confirmed lung lesions (25 primary lung cancers, 28 metastases, 9 granulomas, and 14 pneumonic infiltrates). The patients underwent T1 and T2 mapping at 3T. Two independent readers quantified the mean values for each lesion. The pre-specified primary endpoints were (1) benign versus malignant and (2) primary lung cancer versus pulmonary metastases. Results: Significant differences in T1 and T2 values were observed across lesion types. Benign lesions exhibited high T2 values (mean 213.6 ms) and low T1 values (mean 836.6 ms), whereas malignant tumors exhibited lower T2 values (~77–78 ms) and higher T1 values (~1460–1504 ms, p < 0.001). Binary classification yielded 95.7% accuracy (sensitivity 93.8% for malignant, specificity 100% for benign) in an internal 70/30 hold-out validation (no external dataset), with consistent performance confirmed by patient-level and nested cross-validation (balanced accuracy ≈ 0.92–0.94). However, malignant subtypes could not be reliably distinguished (p > 0.05), and multiclass accuracy was 60.9%. Conclusions: Quantitative MR relaxometry allows accurate, radiation-free differentiation of benign and malignant lung lesions and may help reduce unnecessary invasive procedures.

Purpose: To develop a method for abdominal simultaneous 3D water () and () mapping with isotropic resolution using a free‐breathing Cartesian acquisition with spiral profile ordering (CASPR) at 3 T. Methods: The proposed data acquisition combines a Look‐Locker scheme with the modified BIR‐4 adiabatic preparation pulse for simultaneous and mapping. CASPR is employed for efficient and flexible k‐space sampling at isotropic resolution during free breathing. The imaging pipeline includes subspace reconstruction, water‐fat separation, and ‐specific dictionary matching. The proposed method was validated in a water‐fat relaxometry phantom using spin echo‐based reference techniques and was compared with MOLLI and GRASE mapping in 10 volunteers. The method's flexibility was assessed at isotropic resolutions of 2.5, 3, and 3.5 mm, with corresponding scan times of 7:19, 5:23, and 3:52 min. Additionally, the method was applied to 9 oncological patients with abdominal pathologies. Results: Phantom experiments demonstrated good agreement between the proposed method and spin echo‐based reference techniques across a wide range of and values. The volunteer study assessed in vivo quantification performance and demonstrated the flexibility of the proposed method at different spatial resolutions. The 3 mm method was successively performed in patients with abdominal pathologies, whereby lesions with diameters below 1 cm could be assessed. Conclusion: The proposed Look‐Locker‐based method using CASPR enables efficient, flexible, and simultaneous and mapping at isotropic resolution in a fixed scan time during free breathing.

Purpose: To develop a self‐navigation approach to estimate respiratory motion for motion‐corrected liver mapping using a Look‐Locker acquisition with radial stack‐of‐stars trajectory. Methods: The proposed method derives 1D field‐map profiles from the oversampled k‐space center to estimate a normalized breathing curve and the variation amplitude for each slice and coil. drift and contrast variations, inherent to the Look‐Locker acquisition, were modeled and corrected by fitting and demodulating drift and offset terms. The breathing curve was employed to bin data into motion states for motion‐resolved reconstruction, followed by water‐specific mapping. Simulations with an anatomical body model and in vivo experiments with a Look‐Locker multi‐echo gradient echo sequence were performed to validate the technique. The estimated normalized breathing curve was compared with magnitude‐ and phase‐based self‐navigation approaches using principal component analysis. Results: The proposed self‐navigation reliably estimated the normalized breathing curve and the variation amplitude in simulations and in vivo. variation amplitudes increased with greater tissue displacement, with median values across slices and coils ranging from 4 to 15 Hz at 3 T in volunteers. Motion‐resolved reconstruction using the estimated breathing curve reduced motion artifacts and improved image and mapping quality compared to motion‐averaged reconstruction. Conclusion: self‐navigation allows estimation of respiratory motion in acquisitions with varying contrast and quantifies the variation amplitude, providing a possible surrogate signal for tissue displacement and enabling self‐gated liver mapping using a Look‐Locker approach.