Antje Hellwich

@antjehellwich.bsky.social

1017 Followers 1897 Following

Editor-in-Chief MAGNETOM Flash Internet editor MAGNETOM World, the Siemens Healthineers peer-to-peer community for MRI users. https://www.magnetomworld.siemens-healthineers.com

This week is MRSafetyWeek, the perfect opportunity to refresh essential MRI safety knowledge. MRI safety training is a fundamental part of operating an MR system. To support this, we developed a practical Questions & Answers card deck covering key MRI safety topics — from patient screening and emergency procedures to safe patient positioning and everyday best practices. Whether you're onboarding new colleagues, refreshing existing knowledge, or running departmental safety training, these cards are an easy way to support consistent learning. Available in English, French, Spanish, German, and Swedish.
Shoutout to my colleague Heike Weh (Siemens Healthineers) who scanned the images back in 2007 - scanning each body part in the right postion / rotation and then stitching them together in Photoshop. No AI back then - just about one week of work for this cyclist.
At the beginning of clinical MRI, maximum achievable gradient strength (Gmax) were typically in the range of 1 to 2 mT/m amplitude, with rise times of 1 to 2 ms. This article explores how Siemens Healthineers learned to make good gradients. It traces the remarkable technological advances from 1983, when Siemens Medizintechnik, as it was known then, began developing its first MRI system, the MAGNETOM; almost until today, when Siemens Healthineers provide MAGNETOM Cima.X, MAGNETOM Terra.X, and Connectome 2.0 to the clinical and research community. The article allows a fascinating glimpse behind the scenes of MRI engineering. The historic lab books, photographs, and personal recollections are a wonderful reminder that many of today's "standard" technologies were once bold ideas sketched on a laboratory bench.
The clinical benefit of diffusion-weighted imaging (DWI) with isotropic resolution lies in its volumetric information, enabling, for example, improved assessment of the progression of an epidermoid cyst, tumor, or stroke area. Readout-segmented echo-planar imaging (RESOLVE) produces sharp diffusion-weighted images with high spatial resolution, while GRAPPA (GeneRalized Autocalibrating Partially Parallel Acquisition) and Simultaneous Multi-Slice (SMS) acquisition significantly reduce scan times in clinical MRI. High-resolution imaging (1.1 mm isotropic) creates severe susceptibility artifacts in single shot EPI. RESOLVE is therefore a good alternative, but would require very long scan times on systems with low gradient system configurations. With the 3T MAGNETOM Cima.X, the acquisition time for a RESOLVE sequence with an isotropic resolution of 1.1 mm will be reduced to under 10 minutes, making it feasible for clinical examinations. Learn more about this high-resolution DWI technique with isotropic voxel sizes and convincing image quality – also for multiplanar reformations (MPR) – in an acceptable acquisition time.
The tibial tuberosity to trochlear groove (TT-TG) distance is a radiographic measurement used primarily to assess patellar instability (which occurs when the kneecap slips out of place). It measures the lateral offset between the deepest part of the trochlear groove and the point where the patellar tendon attaches to the tibia. The TT-TG measurement is often requested in orthopedic clinics, particularly for patients presenting with: • Recurrent patellar dislocations • Chronic kneecap pain (patellofemoral pain syndrome) • Patellar maltracking If a surgeon is considering an operation to stabilize the kneecap, the TT-TG distance is part of the standard preoperative workup. The measurement is typically performed using computed tomography (CT) or magnetic resonance imaging (MRI). Radiologists take two cross-sectional images: one at the level of the trochlea and one at the level of the tibial tuberosity. They then superimpose the images to measure the horizontal distance between the two points. Historically, CT was the gold standard for this measurement, partially because the modality provides 3D acquisitions. Today, MRI can also perform 3D acquisitions with thin slices comparable to CT, while offering additional advantages, including different contrasts (achieved with T1, T2, PD, fat-sat, or zero echo time imaging PETRA) that allow visualization of cartilage and ligaments without ionizing radiation. Another advantage of MRI is the possibility of reliably measuring the tibial tuberosity to posterior cruciate ligament distance (TT-PCL). This information helps determine if the instability failure is purely due to the tibial bone.
Magnetic resonance neurography of the cervical spine greatly benefits from the signal-to-noise ratio available at 7T. The 3D DESS sequence with water excitation provides excellent visualization of cervical nerve anatomy and intervertebral discs while effectively suppressing surrounding bone and fat tissue. This enables high-contrast images that clearly depict microstructural detail such as intradural rootlets/roots, dorsal root ganglia (DRGs), and the proximal spinal nerves. A key challenge, however, is the relatively long acquisition time required to achieve the necessary submillimeter isotropic resolution. Extended scan durations make the examination more vulnerable to motion artifacts caused by swallowing, breathing, or cerebrospinal fluid pulsation. Substantial reductions in scan time can be achieved through coherent (e.g., GRAPPA, CAIPIRINHA) and incoherent (compressed sensing [CS]) undersampling combined with deep learning (DL) reconstruction. Combining CAIPIRINHA and CS with DL reconstruction may enable even greater acceleration while preserving or improving image quality compared to state-of-the-art CAIPIRINHA or CS. The authors explored a novel joint CS + CAIPIRINHA (CS-CAIPI) data sampling scheme with DL reconstruction for highly accelerated 3D DESS MR neurography at 7T.
Standardized MRI protocols are essential not only for achieving high image quality and diagnostic accuracy, but also for ensuring efficiency, safety, and reproducibility across a diverse and challenging patient population. Unlike human radiology, veterinary imaging must accommodate multiple species, a wide range of sizes, and varied pathologies — all while operating under the constraints of anesthesia, cost, and staffing. By designing thoughtful, size-based, and anatomically grounded protocols that anticipate these demands, veterinary teams can maximize the utility of MRI as a diagnostic tool. The figure shows the standard spinal protocol in dogs. The imaging volume encompassed either the C1–T2 or T2–S1 spinal segments. Sagittal T2W and dorsal T2W STIR sequences were acquired for planning the region of interest (ROI), followed by transverse T2W images through the ROI.
These are real MRI acquisitions, not AI-generated images. Many thanks to my colleague Heike Weh (Siemens Healthineers, Erlangen, Germany), whose expertise made these remarkable images possible.
Sports cardiologists are frequently confronted with a fundamental challenge: distinguishing physiological adaptations of the athlete’s heart from pathological cardiovascular abnormalities. In this context, cardiac magnetic resonance imaging (CMR) has evolved into an indispensable tool in modern sports cardiology. Its unique ability to characterize myocardial structure, function, tissue composition, fibrosis, and inflammation, and the possibility of anatomical evaluation of the coronary arteries make CMR indispensable for the assessment of athletes with suspected cardiovascular disease. Whether evaluating inflammatory heart disease, differentiating athlete’s heart from cardiomyopathy, assessing arrhythmic risk, monitoring aortic adaptation, or characterizing myocardial scar in coronary disease, CMR provides information that directly influences clinical decision-making. As sports cardiology increasingly embraces individualized care and shared decision-making, CMR serves not merely as a diagnostic modality but as a central component of athlete-centered cardiovascular risk assessment. Its ability to combine anatomical, functional, and tissue-specific information makes it the cornerstone imaging technique for modern sports cardiology. The figure shows serial CMR findings in an elite handball player with physiological exercise-induced cardiac remodeling. The baseline CMR study (left column, 2020) and follow-up CMR study (middle and right columns, 2025) demonstrate persistent left ventricular enlargement. End-diastolic four-chamber cine images (left and middle columns) show stable chamber dilatation without focal structural abnormalities. Corresponding LGE images demonstrate no focal myocardial fibrosis. Native T1 mapping (top right) and T2 mapping (bottom right) at follow-up show normal tissue characteristics without evidence of diffuse myocardial fibrosis or myocardial edema, consistent with athlete’s heart.
Fluid attenuated inversion recovery (FLAIR) is an essential sequence in brain MRI. By suppressing cerebrospinal fluid (CSF) and free water signals, FLAIR improves the contrast between lesions and brain parenchyma, thereby improving the detection of pathologies adjacent to CSF spaces, such as cerebral infarction, abscess, inflammatory demyelinating diseases, and tumors. However, conventional FLAIR without fat suppression suffers from high fat signal and chemical shift artifacts, which may reduce gray-white matter contrast and mask lesions involving the skull or intracranial structures. Frequency selective fat suppressed FLAIR (FS FLAIR) has been widely adopted to overcome these limitations. Unfortunately, FS FLAIR is highly sensitive to B0 and B1 inhomogeneities, which are exacerbated by tissue susceptibility differences, geometric distortion, and the presence of metal dental implants. As a result, FS FLAIR often fails to provide uniform fat suppression, leading to degraded image quality, signal distortion, and reduced diagnostic confidence — particularly in patients with fixed metal dentures. The authors integrated the Dixon technique with FLAIR (Dixon FLAIR). The Dixon method exploits the chemical shift difference between water and fat, using dual echo acquisition and advanced reconstruction algorithms. It is inherently less sensitive to B0/B1 inhomogeneities and local susceptibility variations, offering robust and uniform fat suppression even in challenging anatomical regions without increasing scan time. This article presents three cases demonstrating the differences in fat-suppression quality between FS-FLAIR and Dixon-FLAIR in brain MRI and the resulting impact on lesion visualization. 📌 Please note that Dixon FLAIR is Work in progress. The application is currently under development and is not for sale in the U.S. and in other countries. Its future availability cannot be ensured.
Clinical applications of dental dedicated MRI span multiple domains of oral and maxillofacial medicine. Temporomandibular joint (TMJ) disorders: MRI is the only non-invasive modality that directly visualizes the articular disc, showing its position, shape, and integrity. Differentiating between muscular dysfunction, disc displacement with or without reduction, and degenerative joint disease is clinically decisive and fundamentally influences the therapeutic approach. Endodontics: MRI enables direct visualization of the pulp. Current vitality tests — thermal or electrical — provide indirect, sometimes ambiguous information. MRI can display inflammatory changes around the apex, enable assessments of vascularization, and support the detection of periapical pathology at early stages before substantial bone loss has occurred. Periodontics: MRI allows visualization of soft tissue inflammation and ligament integrity. Active periodontitis can show characteristic edema and hyperemia on MRI before radiographically detectable bone loss occurs. The extent of soft tissue attachment loss can be assessed non-invasively, complementing clinical probing. Surgical planning: MRI enhances surgical planning through detailed nerve visualization. The inferior alveolar nerve, mental nerve, and lingual nerve can be traced along their courses. This allows more precise planning of implant placement, wisdom tooth extractions, and orthognathic procedures, particularly in cases where the nerve course is unusually close to the planned surgical site or where previous interventions have altered normal anatomy. Pediatric dentistry: Here the absence of ionizing radiation is particularly relevant. Repeated MRI in growing children — to monitor dental development, assess impacted teeth, or follow trauma over time — carries no cumulative radiation burden.
Chronic pulmonary diseases (CPD) represent a major global health burden and are the leading cause of death among lung diseases. Each year, more than 55 million new cases are diagnosed worldwide, and approximately 10% of patients affected by CPDs are children. CPDs comprise a spectrum of diseases that require continuous surveillance. Assessing lung function, particularly perfusion (blood flow) and ventilation (airflow), is essential. Nuclear medicine is currently considered the gold standard for functional lung imaging. However, when repeated follow-up imaging is required, radiation exposure has to be taken into account, especially for vulnerable patient groups such as children. LungMaps is a proton MRI-based application that provides functional lung information without the need for tracers, contrast agents, or ionizing radiation. It is based on the PREFUL method, which has been clinically evaluated across a wide range of patient populations. Although LungMaps is a post-processing application, it relies on a 2D MRI acquisition protocol with sufficient temporal resolution to capture both respiratory- and cardiac-based signal variations in the lungs. Shout-out to Andreas Voskrebenzev, Ph.D.; Filip Klimeš, Ph.D.; and Jens Vogel-Claussen, M.D. (Department of Radiology, Charité Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Germany; Department for Diagnostic and Interventional Radiology, Hannover Medical School, Germany) for sharing their protocols. Download the protocols for: -- 0.55T MAGNETOM Free (syngo MR XA80) -- 1.5T MAGNETOM Sola (syngo MR XA61) -- 3T MAGNETOM Vida (syngo MR XA60)
Effective use of gradient magnetic fields plays an important role in achieving high image quality. Boost Gradient Mode is a unique feature that enables MRI systems to use the maximum gradient strength, allowing further optimization of imaging parameters. First introduced on the 0.55T MAGNETOM Free. Platform, it is now also available on the 1.5T MAGNETOM Flow. Platform. One of the key advantages of Boost Gradient Mode is the ability to shorten echo spacing when optimizing imaging parameters. While shorter echo spacing does not inherently shorten acquisition time, it does provide additional benefits that include improved signal-to-noise ratio (SNR) and reduced blurring and banding artifacts. The clinical cases presented in this article demonstrate the practical impact of Boost Gradient Mode across multiple anatomical regions. This article highlights the feasibility and clinical usefulness of Boost Gradient Mode. By making effective use of maximum gradient performance, it helps to accelerate imaging and improve image quality. Although higher gradient performance increases the likelihood of reaching dB/dt limits, integrated tools such as Boost Assistant provide robust mitigation of these risks.
New pediatric MRI protocols with Deep Resolve are now available, shared by Michael Kean, FISMRT, Chief MR Technologist Children’s MRI and PET Centre, Murdoch Children's Research Institute / Honorary Senior Research Fellow, Department of Paediatrics, University of Melbourne The Royal Children’s Hospital. Included are protocols for: - Brain (under 6 months, headache, developmental disorders, multiple sclerosis, temporal lobe epilepsy, intracranial infection, and stroke) - Cervical spine trauma - Spine (scoliosis and discitis) - Long bone infection - Extremity tumor - Liver Some of the protocols have been optimized for quieter scanning using Quiet Mode. The protocols follow the recommendations published by: Ferraciolli SF, et al. International standardization of pediatric magnetic resonance imaging protocols: creation of the World Federation of Pediatric Imaging MR Protocols Committee. Pediatr Radiol. 2025;55(3):375–383. doi: 10.1007/s00247-024-06154-6. Download the .exar1 files for: - 1.5T MAGNETOM Sola (syngo MR XA61) - 3T MAGNETOM systems (syngo MR XA60)
This MRI of a soccer player was acquired by my colleague Heike Weh.
CATCH-220 is a deep learning-enhanced, clinically implementable ultrafast MRI protocol that enables comprehensive whole-brain imaging in less than 220 seconds. By combining multiple image contrasts, including perfusion, within a single rapid acquisition, the protocol offers a flexible and scalable solution for a wide range of neuroradiological applications. Its compatibility with standard 1.5T MRI systems and reliance on commercially available technology support its integration into routine clinical workflows, particularly in time-critical settings and for patients requiring repeated examinations. Of course, some limitations should be acknowledged. The protocol does not include a dedicated angiographic sequence, such as time-of-flight (TOF) MR angiography, which may limit its immediate applicability in certain vascular contexts. Although susceptibility-weighted imaging (SWI) can provide indirect vascular information (e.g., the susceptibility vessel sign), it cannot fully replace vascular imaging yet. Beyond acute stroke, the potential applications of CATCH-220 extend to other clinical scenarios. In longitudinal imaging, such as ARIA monitoring in patients receiving anti-amyloid therapies, shorter scan times may improve adherence and workflow efficiency. Similarly, in pediatric and non-cooperative populations, ultrafast protocols can reduce motion artifacts and minimize the need for sedation. Shout-out and thank you to all co-authors!
The Journey of Dental‑Dedicated MRI at a School of Dentistry: The Aarhus Experience by Dr. Jennifer Christensen and Dr. Rubens Spin-Neto (Department of Dentistry and Oral Health, Aarhus University, Aarhus, Denmark). Dental‑dedicated magnetic resonance imaging (ddMRI) has recently emerged as a promising radiation‑free imaging modality for dentomaxillofacial diagnostics. By combining a lower‑field MRI system, a dedicated radiofrequency coil, and optimized pulse sequences, ddMRI enables simultaneous visualization of dental structures, bone, soft tissues, and inflammatory processes. Aarhus University was among the first institutions worldwide to implement and systematically investigate this technology within its school of dentistry. Through a series of experimental ex vivo and clinical studies, the researchers explored sequence optimization, workflow integration, and diagnostic feasibility across multiple dental disciplines. This article summarizes the first years of ddMRI implementation in Aarhus, highlighting technological developments, patient workflow, sequence optimization, and emerging clinical applications. The Aarhus experience illustrates both the potential and current limitations of ddMRI and outlines key research that will be necessary for integrating MRI into routine dentomaxillofacial imaging. The figure shows Proton‑density‑weighted images of impacted inferior third molars. The red arrow shows the submandibular ganglion and its connection to the lingual nerve; the green arrows show the mandibular canal. (Left, axial view; Middle, sagittal view; Right, coronal view)
The authors investigated the imaging of titanium vs. Carbon/PEEK pedicle screws at 0.55T, 1.5T, and 3T, with and without advanced metal artifact reduction techniques. Their findings demonstrate that techniques such as WARP and SEMAC effectively reduce metal artifacts near titanium implants at all three field strengths. In contrast, Carbon/PEEK screws could be imaged using standard #MRI protocols without implant-related image artifact. It is essential to acknowledge that there are numerous considerations involved in imaging with metal implants. Metal artifact reduction strategies often come at the cost of reduced SNR, increased scan times, and higher SAR. Imaging at lower field strengths (e.g., 0.55T) and with non-metal hardware may dramatically simplify protocol development in such scenarios. Beyond excellent soft-tissue contrast, MRI techniques like diffusion-weighted imaging (DWI) and magnetic resonance spectroscopy (MRS) offer valuable functional information that aids in assessing tissue viability, identifying areas of hypoxia, and optimizing dose delivery based on tumor biology. Furthermore, MRI can facilitate monitoring of tumor response and treatment efficacy during and after radiation therapy without concerns over ionizing radiation exposure. This capability empowers clinicians to evaluate treatment response and adapt treatment plans as needed. Whether in treatment planning, functional imaging, or treatment follow-up, addressing the challenge of imaging with spine implants requires close collaboration between radiation oncologists, medical physicists, and radiologists. Optimization of MRI protocols, development of artifact reduction strategies, and accurate integration of MRI data into the treatment planning process are paramount. Despite these challenges, the benefits of using MRI in spine radiation therapy should outweigh the limitations, and ultimately lead to improved treatment outcomes for patients with spinal tumors.
In the last decade, whole-body MRI (WB-MRI) has been established as a standard diagnostic and treatment-monitoring tool for cancer patients with bone disease. Bone metastases are a common finding in late-stage cancer patients and can be associated with poorer prognosis, treatment resistance, and tumor-related adverse skeletal events, such as fractures, spinal cord compression, and the need for bone radiotherapy. Standard WB-MRI protocols, as prescribed in MET-RADS and MY-RADS, focus exclusively on the evaluation of bone marrow involvement, largely because of the inability to visualize the mineralized bone. Thus, WB-MRI has always had the limitation of requiring computed tomography (CT) scans to confidently evaluate the bone structure for surgical interventions and RT planning. Recent advances have focused on generating CT-like images using MRI. Acquiring CT-like images simultaneously with conventional WB-MRI could reduce the need for additional CT appointments, eliminate ionizing radiation doses, and minimize spatial registration errors. Indeed, the creation of synthetic CT-like images for radiotherapy planning purposes is already possible using the syngo.via RT Image Suite. However, these images cannot provide the level of bone matrix detail needed for the radiological diagnosis of bone pathology. Multiple investigators have reported bone MRI using a variety of VIBE and ZTE/UTE approaches. Structures with no signal return show black against a uniformly gray soft tissue background — known as “black-bone” imaging. When the image contrast is inverted, the bone appears white, like that in CT. Deep Resolve reconstruction could cut whole-body diffusion-weighted imaging (WB-DWI) sequences by almost 50% without corresponding declines in image quality. Building on their experience, the authors developed and clinically implemented a proton density (PD)-weighted VIBE sequence using Deep Resolve to demonstrate its added value for bone imaging in WB-MRI examinations.
MRI for local staging and restaging of rectal cancer is a cornerstone of modern multidisciplinary treatment. High resolution MRI is recommended by global organizations, as the quality of the images has a direct impact on radiologists’ ability to accurately assess the tumor. To ensure clinical consistency, we provide protocols which meet the recommendations from ESGAR (European Society of Gastrointestinal and Abdominal Radiology), SAR (Society of Abdominal Radiology, USA) and Australian organizations. These protocols are optimized for varied MAGNETOM systems to deliver the high-contrast, high-resolution detail required for treatment planning. • Protocols for high-resolution T2W: Small FOV, thin-slice (3 mm) sequences with base in-plane resolution 0.6 x 0.6 mm (0.3 x 0.3 x 3.0 mm with Deep Resolve). Sagittal and aligned to the tumor axis for optimized assessment of T-stage, Extramural Venous Invasion, Lymph Node and Mesorectal Fascia involvement. • Protocols for DWI: High b-value diffusion-weighted imaging (b = 800 s/mm^2) to increase sensitivity for post-treatment response. • Large FOV protocols for LM staging: Axial overview of the entire pelvis to identify distant lymphadenopathy. Download the guideline-compliant .exar1 and PDF protocol packages for your 1.5T or 3T system
The effect on the image quality and SNR can be seen in the figure. For conventional GRAPPA acquisitions, the images become noisier. For Deep Resolve acquisitions the noise level is reduced, although very fine image features may be subject to increased blurring.
In the practice of Cardiac MRI, various definitions of the parameter TR (Repetition Time) have evolved, as gated pulse sequences have become increasingly complex. Understandably this has caused significant confusion among many users. In some cases, the original concept of TR seems to have been lost. This article describes the various definitions of TR. While the physical definition of TR will always mean only one thing, the time between successive excitation pulses, a number of variations have resulted from the necessity of controlling a pulse sequence within the cardiac cycle. Hopefully, this brief explanation, along with the accompanying diagrams, will make the topic easier to understand for everyone working in cardiac MRI.
Gliomas are the most common primary intracranial tumors in adults. They originate from central nervous system glial cells and are categorized by the WHO into grades 1 to 4. Common types in adults include glioblastoma (grade 4), astrocytic tumors (grades 1–3), oligodendrogliomas (grades 2, 3), and ependymomas (grades 1–3). Glioblastoma is notably aggressive, with a two-year survival rate below 10%. Achieving the maximal extent of resection has been shown to enhance both progression-free survival and overall survival in the surgical treatment of gliomas. The optimal surgical strategy for gliomas requires balancing aggressive tumor removal with minimizing the risk of new or worsened neurological deficits to ensure better outcomes. Due to the diffuse nature of gliomas, neurosurgeons face a significant challenge in accurately identifying the boundary between tumor tissue and adjacent brain matter. Hence, glioma resection is guided by the anatomical expertise of the surgeon, which can be aided by visualization techniques, making the selection of the appropriate imaging strategies crucial in glioma surgery. The authors provide an overview of the possibilities for visualizing brain-tumor margin, with a particular focus on the input of intraoperative magnetic resonance imaging (ioMRI). They also present a case in which we performed ioMRI with machine learning-based radiomics. Shoutout to the co-authors: Baran Atli, M.D.; Martin Aichholzer, M.D.; Andreas Gruber, M.D.
Wake-up strokes account for 14%–28% of all ischemic strokes and present a unique diagnostic challenge. Patients awaken with neurological deficits, making it impossible to determine the onset time of the stroke. Historically, this uncertainty excluded these patients from thrombolytic therapy, as traditional protocols required confirmed symptom onset within 4.5 hours. Today, advanced neuroimaging is changing this paradigm. The shift from rigid time windows to flexible tissue windows, guided by advanced MRI techniques has transformed wake-up strokes from untreatable conditions to intervention opportunities. Successful intervention requires integrated emergency response, rapid imaging capabilities, expert interpretation, streamlined protocols, and robust monitoring. Innovations in medical technology — including high-field MRI systems, accelerated sequences, automated analysis software, and algorithms — continue to expand access to evidence-based treatment. For healthcare systems, investing in advanced imaging infrastructure and specialized teams directly translates into improved outcomes. For patients, stroke symptoms require immediate medical attention regardless of onset time. Wake-up strokes no longer automatically exclude patients from treatment. Through continued collaboration among clinicians, researchers, and medical-technology innovators, the 14%–28% of stroke patients whose symptoms begin during sleep can increasingly receive interventions that preserve brain tissue and function.
Deep learning reconstruction (DLR) is a double-edged sword for radiomics. While it addresses bottlenecks in MRI workflows, its feature perturbations demand rigorous validation. Radiomics, which extracts quantitative features from medical images for clinical decision-making, has gained in popularity. Yet it faces challenges caused by its sensitivity to acquisition parameters, reconstruction algorithms, and post-processing techniques. DLR enables acceptable image quality and faster acquisitions, and it is therefore expected to be widely adopted in the future. However, the influence of DLR and the combined effects of DLR and accelerated acquisition on radiomics features have not yet been investigated. It is necessary to evaluate in advance the potential changes that the new technique could cause to radiomics features. This study provides a basis for future research into establishing radiomics models using images from an accelerated HASTE sequence with DLR. The authors investigated the impact of DLR and accelerated acquisitions on the robustness of radiomics features in abdominal scans acquired using a HASTE sequence. By adhering to standardized protocols and investing in robustness testing, the radiology community can harness the full potential of DLR without compromising the diagnostic promise of radiomics. Shoutout to the co-authors: Yang Song, Minda Lu, and Dominik Nickel
Ultra-high-field MRI (7 Tesla and above) enables submillimeter spatial resolution that creates new opportunities for detailed neuroanatomical assessment and lesion characterization. However, long acquisition times are a major barrier to routine clinical use. The authors illustrate how deep learning-based reconstruction enables higher acceleration of high-resolution brain imaging while preserving diagnostic image quality and thereby supporting the broader clinical usability of 7T MRI. Shoutout to the co-authors: Thomas Yu, Jocelyn Philippe, Gabriele Bonanno, Kevin Battistini, Caterina Bernetti, Roland Wiest, Marwan El-Koussy, Natalia Pato Montemayor, Dominik Nickel, Patrick Liebig, Robin M. Heidemann, Felix T. Kurz, Jean-Philippe Thiran, Tom Hilbert, Tobi Kober, Gian Franco Piredda
Explore the ISMRM 2026 edition MAGNETOM Flash with an Editorial Comment on “Ubuntu in the global conversation around MRI access” by Ernesta Meintjes, PhD (University of Cape Town (UCT), Cape Town, South Africa). As a peer-to-peer journal with more than 30 years in the MR community, MAGNETOM Flash reflects that progress in MRI is driven by community and collaboration. Inspired by the ISMRM motto “Ubuntu – I am because we are” this issue has a focus on expanding access to MRI, evolving guidelines, and the real-world impact of deep learning reconstruction. 👏 A big thank you to all contributors!
The 2024 revisions of the international diagnostic criteria for multiple sclerosis introduce an expanded role for MRI by incorporating two novel imaging biomarkers: the central vein sign (CVS) and paramagnetic rim lesions (PRL). Recent MRI guidelines developed by the MAGNIMS-CMSC-NAIMS consortia provide practical recommendations for implementing the MRI sequences required for accurate biomarker detection and reliable identification of MS. Given the need for rapid clinical adoption of these newly integrated imaging biomarkers, the article outlines practical strategies to support the integration of appropriate susceptibility MRI sequences into routine diagnostic workflows. The use of an optimized susceptibility-weighted imaging (SWI) protocol is recommended to ensure sensitive detection of CVS, while filtered phase images are suggested for improved detection of PRL. When available, advanced 3D echo planar imaging (EPI) sequences are recommended to acquire submillimeter isotropic images, enabling multiplanar evaluation of T2*-weighted and filtered phase images and increasing diagnostic confidence in the assessment of CVS and PRL. Shoutout to all authors and co-authors 👏 Sreekanth Madhusoodhanan Nair, Jin Jin, Chang Gao, Nader Binesh, Thomas Benkert, Marcel Maya
Neurological disorders remain a growing public health challenge in low- and middle-income countries (LMICs), particularly in sub-Saharan Africa, where limited resources, workforce shortages, and infrastructure constraints continue to impact access to care. Neuroimaging, especially MRI, plays a critical role in improving diagnosis and clinical management, yet its adoption in these regions is constrained by infrastructure, financing, and training gaps. The Scan With Me (SWiM) program addresses these challenges through a structured, train-the-trainer model designed to strengthen MRI capacity. Since 2023, the program has trained 233 MRI technologists across 25 LMICs, delivering extensive instruction, hands-on observerships, and peer-to-peer collaboration to optimize imaging protocols and improve clinical practice. In parallel, technological innovation is expanding access to MRI. The installation of a 0.55T MAGNETOM Free.Max system in Rwanda demonstrates how simplified infrastructure requirements, reduced energy consumption, and flexible deployment can enable high-quality diagnostic imaging in resource-constrained environments. Together, scalable training initiatives and innovative MRI solutions can empower healthcare professionals, democratize access, enhance diagnostic precision and clinical decision-making, and ultimately improve patient outcomes across resource-constrained settings. Shoutout and thank you to the authors and co-authors!
Neuroimaging findings in neurodegenerative disorders are often complex, as imaging findings in patients with mild signs and symptoms are frequently subtle and ambiguous. In many cases, by the time imaging findings become obvious, patients have already manifested clinically, and the diagnosis is already established or at least highly suspected. The radiation-free PCASL-MRI approach helps, based on the perfusion profile, to achieve an early and accurate diagnosis prior to gross morphological alterations when standardized with PET in simultaneous PET-MRI acquisitions. PCASL and PET changes preceded structural atrophy patterns and could aid in establishing early clinical diagnosis. The combination of PET and PCASL boosted the sensitivity of structural MRI and PET by synergistically diagnosing disease conditions. The synergistic effect of PCASL and FDOPA boosted the sensitivity and specificity in classifying disorders with dopaminergic deficit into IPD/APD phenotypes without the need for additional FDG PET or D2-receptor imaging. The authors highlight the role of simultaneous PCASL PET-MRI in the workup of complex neurodegenerative conditions with FDG in centers that do not have a cyclotron facility for producing non-FDG tracers. In clear-cut clinical neurodegenerative disorders, where structural changes are equivocal, an additional PCASL sequence can help in early diagnosis without the need for PET studies.
In functional MRI (fMRI) neurofeedback, a closed-loop brain-computer interface enables participants to learn how to self-regulate brain activity using real-time feedback. This method is gaining traction in both scientific and clinical settings. Several 3T MRI studies have shown the promise of this approach as an add-on treatment for depression, helping patients improve emotion regulation and self-efficacy. Conventional neurofeedback cannot distinguish specific emotions. It is therefore unclear whether a patient is engaging in a positive or negative emotional mental state. To address this ambiguity, a new semantic neurofeedback approach maps individual emotions onto a two-dimensional space, visualizing the current emotional state as a moving point on this map. This technique allows participants to actively navigate their emotional space, offering clearer and potentially more effective therapy for depression. A 7-Tesla proof-of-concept study shows promise for this new real-time fMRI neurofeedback method.
NATIVE (non-contrast angiography of the arteries and veins) is a contrast-free MR angiography technique designed to visualize vascular structures throughout of the body. Tailored protocols for use in renal arteries and peripheral vessels are available within the protocol tree provided by Siemens Healthineers. syngo NATIVE TrueFISP is based on the TrueFISP (true fast imaging with steady state precession) sequence, a balanced steady-state gradient echo technique. The sequence can be made selective for arteries or veins by appropriate positioning of the inversion pulse, which can be positioned independently from the imaging volume. The most robust method of performing syngo NATIVE TrueFISP is with respiratory triggering. With BioMatrix Technology, the respiratory sensor in the BioMatrix spine coil improves the workflow and increases efficiency.
Beyond the Clock: The Pivotal Role of Advanced MRI in Revolutionizing the Diagnosis and Management of Brainstem Stroke by Anirudda Deshpande, MD, DM, DNB, FRCP, SCE Neurology; et al. (Dept. of Care of the Elderly & Stroke, Altnagelvin Hospital, Londonderry, Northern Ireland, UK). The brainstem is a compact neural structure that houses critical life-sustaining nuclei and white matter tracts. When ischemic injury occurs in this region, diagnosis becomes particularly challenging. Its complex anatomy and the non-specific nature of presenting symptoms often lead to diagnostic delays or misdiagnosis. Computed tomography is notoriously insensitive to acute brainstem ischemia. This review highlights the indispensable role of advanced magnetic resonance imaging (MRI) as the gold standard for diagnosing brainstem stroke. The authors describe how specific MRI sequences — particularly high-resolution diffusion-weighted imaging — enable precise anatomical localization, facilitate differential diagnosis, and provide prognostic information. The article also explores the emerging potential of advanced techniques such as diffusion tensor imaging and high-resolution vessel wall imaging in refining our understanding of brainstem stroke pathophysiology and in guiding future therapeutic strategies. Advanced MRI has changed the landscape of brainstem stroke management. It has evolved from a mere diagnostic tool to a comprehensive guidance system for acute triage, etiological classification, and prognostic stratification. By leveraging a multimodal protocol on a technologically advanced MRI platform, clinicians can overcome the inherent challenges of brainstem anatomy, ensure accurate and timely diagnosis, and provide patients with highly informed and effective care pathway. The continued integration of these sophisticated imaging techniques into clinical protocols is essential for further improving outcomes in this complex patient population.
Stereotactic Radiotherapy Cranial Protocol by Simon Atkins, BAppSc (Hons), MSc, DipIPEM(S), CMPS, and MACPSEM (ICON Cancer Centre Australia). The goal was to establish a unified imaging protocol across ICON centers, ensuring consistency while remaining feasible across different baseline systems (syngo MR XA30, XA50, and XA60). A key objective was to create a workflow that could be implemented with minimal radiotherapy-specific expertise required from on-site radiographers. Physician requirements focused on achieving 1 mm isotropic 3D imaging for SRS treatments. In response, T1 SPACE and T2 SPACE sequences were prioritized due to their superior spatial resolution and high-quality three-dimensional datasets compared with conventional T1 MPRAGE and T2 TSE acquisitions. The MPRAGE and T2 TSE sequences were retained as optional components to accommodate site-specific preferences. To support consistent implementation across sites, the protocol also includes guidance images and explanatory notes to assist radiographers during planning, with particular emphasis on maintaining critical parameters and avoiding unintended modifications.
Female pelvic disorders have a profound impact on women’s quality of life and overall health. Among the most prevalent and clinically significant are endometriosis, ovarian, uterine, and cervical cancers. Accurate and timely diagnosis is essential for improving patient outcomes and quality of life. MRI plays a pivotal role, offering detailed anatomical and functional information that supports diagnosis, staging, and treatment planning. By aligning with ESUR recommendations, the new uterus and ovaries workflows introduced in the syngo MR XB10 software provide standardized, high-quality imaging protocols tailored to gynecological needs. Additionally, Deep Resolve – our deep learning-based image reconstruction – can further accelerate scan times while maintaining high image quality, enhancing patient comfort and workflow efficiency.
The endovascular treatment of intracranial aneurysms has become a standard therapeutic option over the past three decades. Its success is closely linked to the quality of angiographic imaging, which not only enables diagnosis and procedural planning, but also provides the real-time guidance necessary for safe device deployment. Optimal imaging in this context does not merely mean standard anterior-posterior (AP) or lateral projections, but rather the identification of individualized “working projections” that provide unobstructed visualization of the aneurysm sac, the neck, and the parent vessels. Traditionally, such projections were identified empirically, often requiring multiple 2D acquisitions in a trial-and-error fashion. The introduction of 3D rotational angiography and 3D digital subtraction angiography (3D-DSA) has dramatically improved this process, allowing operators to plan projections on volumetric datasets before attempting them in the angiography suite. However, a persistent limitation remains: the mechanical restrictions of C-arm systems. When an operator selects the theoretically optimal projection, it may not be physically achievable because of collision risks with the patient or the angiography table. This mismatch between imaging theory and clinical reality is a frequent bottleneck in neurointerventions. To address this gap, researchers at University Hospital Magdeburg have developed a prototype solution: automated calculation of optimized head positions. By adjusting the orientation of the patient’s head, the anatomical relationship between the intracranial vessels and the C-arm geometry can be modified, creating feasible working projections that would otherwise be unattainable. Work in progress. Currently under development and not for sale in the U.S. and in other countries. Its future availability cannot be ensured.
Interested in how open platforms are accelerating clinical research and innovation? Don’t miss these expert talks: 🎤 Translating research into seamless clinical evaluation using Open Recon by Till Hülnhagen, PhD (Siemens Healthineers, Germany) 🎤 Unlocking Innovation: gammaSTAR’s integration with the Open Sequence interface by Daniel Christopher Hoinkiss, Dr. rer. nat. (Fraunhofer MEVIS, Germany)
By combining accessible, works-anywhere solutions with advanced technologies, this SCMR edition of MAGNETOM Flash underscores a shared mission: to democratize CMR while continuing to define its future across diverse healthcare settings worldwide. The issue covers CMR in light the 2025 ESC Guidelines on Myocarditis and Pericarditis; microstructural imaging; valvular heart disease; whole-chest, whole-heart, and whole-aortic imaging; congenital and pediatric CMR; AI in CMR, cardiac MRI at 0.55T, and much more. We sincerely appreciate all contributors! We also truly appreciate the work of everyone involved in bringing this issue to life!
Cardiac Amyloidosis Through the Lens of Cardiac MRI by Diego Artemio Valadez-Villegas, MD; et al. (Instituto Nacional de Cardiología Ignacio Chávez, Ciudad de México). Cardiac magnetic resonance (CMR) plays a central role in evaluation of heart failure by clarifying both etiology and prognosis. One of its most important applications is the assessment of cardiac amyloidosis (CA), a progressive and underdiagnosed disease caused by the accumulation of amyloid fibrils in the extracellular space of the myocardium, altering contractility, impairing relaxation, and causing restrictive physiology associated with high morbidity and mortality. CA is most commonly due to either light-chain (AL) amyloidosis, derived from clonal plasma cell dyscrasia and associated with poor outcomes, or transthyretin (ATTR) amyloidosis, related to misfolded transthyretin protein synthesized in the liver. Until recently, diagnosis relied on biopsy, but this approach is invasive, less sensitive in ATTR, and limited in early disease. CMR has therefore emerged as a noninvasive, reproducible, whole-heart technique capable of characterizing myocardial composition and distinguishing CA from other cardiomyopathies.
Survival of patients with congenital heart disease (CHD) has dramatically improved over the last 30 years. Over 97% of patients now reach adulthood, as a result, the adult CHD population has outgrown the pediatric population. However, despite improved survival, CHD patients cannot be considered cured. Increased life expectancy exposes them to potential complications such as arrhythmias, stroke, heart failure, valvular heart lesions, and pulmonary arterial hypertension. Due to their high risk of complications, patients with CHD need life-long specialized follow-up. Imaging plays a central role in diagnosis and surveillance, particularly for valvular lesions, shunt quantification and assessment of cardiac volumes, and function. While transthoracic echocardiography (TTE) remains the first-line modality, it has several limitations in CHD, particularly in the assessment of right-heart lesions, which are frequently encountered in the CHD population. Cardiac magnetic resonance (CMR) overcomes these limitations and is considered the gold standard for non-invasive evaluation of ventricular volumes, myocardial structure, and flow evaluation. As the number of adult CHD patients and consequently the demand for congenital CMR is steadily increasing, rapid access to CMR has become increasingly challenging due to long waiting times. To address this issue and expand the number of examinations performed per week, our center introduced a “short CMR” strategy in 2024. In this approach, selected patients undergo an accelerated protocol tailored to the clinical indication and the need for contrast administration. Shoutout and sincere thanks to the co-authors: Fabienne Dirbach, Marco Müller, Christopher Roy, Paolo Garelli, Jürg Schwitter, Matthias Stuber
Artificial intelligence (AI) is emerging as a powerful ally in cardiac MRI, addressing many of the challenges that previously limited its efficiency and accessibility. By automating and optimizing steps from protocol planning and image acquisition to reconstruction, analysis, and integration with clinical data, AI can make cardiac MRI faster, more consistent, and more widely available. Far from replacing clinicians, AI supports them by reducing repetitive tasks, improving reproducibility, and enabling the extraction of advanced diagnostic and prognostic information. An important aspect of this evolution is the integration of cardiac MRI into a multimodality framework where it is combined with other imaging techniques such as echocardiography or CT, and with clinical, biological, and electrophysiological data. This approach paves the way for advanced concepts like the digital twin – a virtual model of the patient’s heart that can guide diagnosis and therapy planning, further enhancing precision and personalization in cardiovascular care. The authors explore how AI is transforming their cardiac MRI practice in four main domains: 1. Planning and acquisition: including automated plane prescription and parameter optimization 2. Image reconstruction: accelerating acquisitions and improving image quality 3. Image analysis and post-processing: enabling rapid and consistent quantification 4. Development of diagnostic and prognostic tools: integrating imaging with multisource and multimodal patient data
To address the challenges of prolonged scan times and reliance on patient breath-holding, a range of free-breathing CMR techniques have emerged over the past decade. These include respiratory navigator gating, self-gating, motion correction, compressed sensing, and AI-assisted sequence optimization. While impactful, most of these approaches were limited to specific sequences and did not systematically transform the overall CMR workflow. In this work, the authors introduce a fully integrated free-breathing CMR protocol that enables intelligent, rapid examinations without the need for breath-holding. Compressed Sensing (CS) and motion correction (MOCO) are core techniques implemented in the full free-breathing CMR protocol. Conceptually, CS leverages the inherent sparsity of CMR data to enable accurate image reconstruction from undersampled k-space, thereby substantially reducing acquisition time without compromising diagnostic fidelity. Motion correction utilizes advanced retrospective algorithms to compensate for respiratory and cardiac motion, effectively reducing motion-related artifacts and maintaining image quality. The core innovation lies in the deep integration of these two methodologies across several critical CMR sequences, leading to a qualitative leap in acquisition efficiency and diagnostic robustness. Full free-breathing CMR examinations are performed on a 3T MAGNETOM Vida using an 18-channel body coil. Free-breathing CMR is not only a technological innovation, but also a milestone in making advanced cardiovascular imaging more equitable and patient-centered. By eliminating reliance on breath-holding, this approach delivers faster, smarter, and more widely accessible diagnostic services. Shoutout to the coauthors: Wenli Zhou, Kai Yang, Gang Yin, Jing An, Xinling Yang, Xiaoming Bi, Jianing Pang, Kelvin Chow
Shoutout to Heike Weh for scanning the tree!
Photon-counting CT (PCCT) represents a major step toward bridging the historical gap between cardiac CT and CMR. Its ability to provide high-resolution anatomical imaging, quantitative spectral data, and myocardial tissue characterization within a single, time-efficient acquisition brings cardiac CT closer to the comprehensive diagnostic capability traditionally reserved for CMR. However, these modalities should not be seen as competing but as highly complementary. CMR remains the reference standard for advanced tissue characterization, myocardial fibrosis assessment, and complex flow quantification, while PCCT offers unparalleled spatial resolution, reduced acquisition times, and simultaneous evaluation of coronary arteries, myocardium, and valves. Together, PCCT and CMR form a synergistic imaging approach, combining structural and functional insights to refine diagnosis, guide management, and ultimately improve patient care in cardiovascular disease. Shoutout to the co-authors: Adrien De Minteguiaga, M.D.; Christos Gkizas, M.D., Ph.D.; Aimée Rodriguez Musso, M.D.; Mehdi Haidar, M.D.
MRI-Guided Planning for Pelvic Brachytherapy By Florian Putz, M.D. (Department of Radiation Oncology, Universitätsklinikum Erlangen, FAU Erlangen-Nürnberg, Erlangen, Germany) At University Hospital Erlangen, Germany, MRI plays a central role in the planning of pelvic brachytherapy. After applicator implantation, a planning MRI is performed on the 1.5T MAGNETOM Sola RT Pro Edition, which is conveniently located within the radiation oncology department. This setup enables streamlined workflow between the procedure area and the HDR treatment rooms. The MR-in-RT working group at University Hospital Erlangen shares their MRI protocols for pelvic brachytherapy. Developed in accordance with the GEC-ESTRO recommendations, these protocols include 3D sequences such as T2-SPACE and T1-VIBE Dixon. Together, they provide excellent visualization of the target volume, organs at risk, and applicators, supporting accurate and reproducible treatment planning.
Diffusion MRI of the human heart provides unique insight into myocardial microstructure but has been hampered by cardiac and respiratory motion, short T2 of the heart muscle, and limited gradient strength. Recent advances in ultra-strong gradient technology not only help to overcome these technical challenges but also allow higher diffusion weighting (i.e., b-values) with clinically compatible echo times. The authors demonstrate how this enabled in vivo diffusion kurtosis imaging (DKI) and q-space trajectory imaging (QTI) in the beating human heart, therefore moving beyond the Gaussian assumptions of diffusion tensor imaging (DTI). These advances may pave the way for more sensitive biomarkers of pathological changes of the myocardium and bring microstructural imaging closer to clinical application. Key points • Ultra-strong gradients (300 mT/m) make cardiac diffusion MRI feasible at higher b-values. • In vivo cardiac diffusion kurtosis imaging and q-space trajectory imaging (QTI) were demonstrated with clinically compatible echo times. • Kurtosis and QTI metrics reveal non-Gaussian diffusion, offering access to new imaging biomarkers of myocardial microstructure. • Translation to clinical systems is within reach with new 200 mT/m gradient scanners. Shoutout and thank you to the co-authors: Lars Mueller, Ph.D.; Jürgen E Schneider, Ph.D. (Biomedical Imaging Science Department, Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds, Leeds, UK) Derek K Jones, Ph.D. (Cardiff University Brain Research Imaging Centre (CUBRIC), School of Psychology, Cardiff University, Cardiff, UK) Filip Szczepankiewicz, Ph.D. (Department of Medical Radiation Physics, Lund University, Lund, Sweden) Fabrizio Fasano, Ph.D. (Siemens Healthineers)
In multiple sclerosis, care relies on MRI for diagnosis and longitudinal management. Standardized, cross-institutional MRI protocols maximize clinical utility by aligning with contemporary criteria. They enable early, accurate diagnosis and capture prognostic markers, flagging subclinical activity that may signal therapeutic ineffectiveness or safety concerns, and support equitable access to high-quality, innovative treatment. General University Hospital in Prague presents optimized therapeutic MS imaging protocols for 3T systems from Siemens Healthineers. The suite of protocols — diagnostic, monitoring, and safety — covers the brain, spinal cord, and optic nerves to provide a comprehensive assessment of central nervous system disease activity, including dissemination in space and time, negative prognostic markers, and treatment-relevant activity. For brain MRI, 3D FLAIR, high-resolution 3D MPRAGE, 3D DIR, 3D EPI, and 2D T2WI and DWI delineate lesions in white and gray matter. Contrast-enhanced imaging is performed immediately following gadolinium administration to highlight new or active lesions. Spinal cord evaluation employs sagittal T2 STIR — in line with guidance from the Magnetic Resonance in MS (MAGNIMS) network — and proton density sequences, followed by post-contrast T1 to detect inflammation and structural damage. Optic nerve assessment includes coronal and transverse T2WI with fat suppression (FS) and, after contrast administration, T1WI with FS to visualize inflammation and demyelination associated with optic neuritis. These harmonized protocols provide a reliable framework to detect, localize, and monitor MS lesions across the central nervous system in accordance with the 2021 and 2024 MAGNIMS international consensus recommendations and the diagnostic McDonald criteria. Shout out to Cinzia Gussoni, Michala Belasova, Jaroslav Leitmann, André Fischer
This issue covers a broad spectrum of innovation: from the potential of ultra-high-gradient MRI to the advantages of lower-field MRI for imaging claustrophobic patients and those with implants; from ultra-fast pediatric brain MRI to standardized pelvic protocols; and from improving access to care with mobile MRI to tackling the real-life challenges of operating MRI systems in geographically challenging or remote locations. The motto of this year’s RSNA — “Imaging the Individual – Fuel Curiosity, Ignite Inspiration – Fresh Insights, New Frontiers – Global Reach, Limitless Potential” — is fully reflected in the articles featured in this edition.
To date, 7T TOF-MRA has shown value in the evaluation of cerebrovascular malformations, atherosclerosis, and neurodegenerative diseases. However, it also faces challenges such as susceptibility artifacts at the air-bone interface of the sella turcica region, and physiological motion (e.g., vascular pulsation), which lead to suboptimal imaging of the petrous segment of the internal carotid artery (ICA). This paper aims to optimize imaging of the petrous ICA through optimized acquisition parameters. The petrous segment of the ICA is located in the parasellar region, where the large susceptibility difference between air and bone leads to susceptibility artifacts. These artifacts affect the homogeneity of both the B0 and B1 fields, which can result in asymmetry of TOF signals between the two sides. Even B0 shimming and volume-specific parallel-transmit (pTX) B1 shimming could not fully resolve the signal loss at the ICA. The optimization strategy was revised by altering the flow compensation scheme and increasing the acquisition bandwidth, therefore shortening the echo time (TE) and reducing the influence of T2* effects. Furthermore, changing the phase-encoding direction redirected pulsation artifacts away from critical vascular structures. Together, these approaches minimized the impact of susceptibility and pulsation artifacts in this region and led to a better display of the ICA on the TOF-MRA. The figures show a 77-year-old male with atherosclerosis. Visual inspection of the original TOF-MRA images revealed the following findings: 1. Shorter TE values exhibit superior efficacy in suppressing vascular pulsation artifacts. 2. Reduced TE improved homogeneity of the blood flow signal. 3. Under conditions of pronounced pulsation artifacts, P>>A (posterior-to-anterior) phase-encoding orientation effectively prevents cross-vessel interference induced by pulsation-related artifacts.
Magnetic resonance imaging (MRI) plays a vital role in medical imaging, yet both healthcare in general and imaging in particular have a considerable environmental impact. MRI systems are highly energy-intensive, posing challenges for healthcare sustainability and environmental preservation. One approach to reducing MRI’s energy footprint is the use of artificial intelligence – specifically, deep learning. While deep learning applications can optimize multiple aspects of MRI operations, their ability to accelerate imaging protocols is especially relevant to sustainability. With advanced denoising techniques, deep learning maintains diagnostic image quality while significantly shortening the time required for each scan. Since faster scans use less energy, deep learning has the potential to minimize greenhouse gas emissions from MRI systems.
This article addresses claustrophobia in #MRI patients and presents strategies to improve the overall patient experience. With its ultra-wide 80 cm bore, the 0.55T MAGNETOM Free.Max scanner is designed to create a less confining environment and enhance comfort. Its compact design and shorter magnet length allow for more flexible patient positioning. Additional measures such as using Contour coils, music during the scan, and cushions for positioning can further increase comfort. The MAGNETOM Free.Max Head/Neck coil has a 12-channel design with 12 integrated pre-amplifiers with three rungs of 4 elements. The coil is tilted to 9 degrees to give patients the feeling of lying on a pillow. The upper part, with 6 elements, can be easily removed. The lower part, also with 6 elements, is usable without the upper part for highly claustrophobic patients. Even with the upper part removed, the AutoAlign option still works. Deep learning-based image reconstruction with Deep Resolve Boost helps denoise the image and achieve good quality. Protocol adjustments, such as selecting prescan normalize, increasing the FOV, and changing the iPAT factor and average, are recommended to increase SNR when scanning without the upper part of the Head/Neck coil.
Advances in Ultra-High-Gradient MRI: High-Resolution Imaging for Accurate Diagnosis of Pancreatic Cystic Lesions by Liang Zhu, M.D.; et al. (Peking Union Medical College Hospital, Beijing, China). Pancreatic cystic lesions (PCLs) are increasingly detected in clinical practice, largely due to the widespread use of computed tomography (CT) and magnetic resonance imaging (MRI). These lesions encompass a broad spectrum of diseases with similar imaging features, ranging from non-neoplastic cysts and benign tumors to premalignant and frankly malignant neoplasms. Accurate diagnosis is therefore challenging. Benign lesions may be misinterpreted as malignancies, leading to overtreatment, while early-stage cancer may go undetected, delaying appropriate treatment. MRI offers distinct advantages for the evaluation of PCLs, including noninvasiveness, absence of ionizing radiation, superior soft-tissue contrast, and multiparametric imaging capabilities. Moreover, high-resolution MRI enabled by an ultra-high-gradient system and deep-learning reconstruction delineates subtle internal structures of PCLs and provides critical diagnostic clues. 👏 Shoutout to the co-authors: Bowen Wu, Yitong Lu (Department of Radiology, Peking Union Medical College Hospital, Beijing, China) Yueluan Jiang, Xiaoye Wang (Siemens Healthineers Ltd., Beijing, China)
Among the various functional imaging techniques, arterial spin labeling (#ASL) has emerged as a powerful tool, not only in neuroscience and cognitive research but also in clinical practice. By using water as an endogenous tracer, ASL delivers quantitative perfusion information with high spatial specificity. Over the years, ASL has demonstrated its clinical value in diagnosing and monitoring a range of neurological conditions, from stroke and brain tumors to neurodegenerative disease, and has been extensively applied at 3T and 1.5T field strengths. The advent of ultra-high-field (UHF) 7T MRI has opened new opportunities for ASL while also presenting significant technical challenges. Increased magnetic field strength accentuates B0 and RF field inhomogeneities, increases SAR, and markedly reduces transverse relaxation time (T2). These challenges, less pronounced at 3T or lower fields, demand systematic innovations and modifications to the ASL sequence to ensure robust and reliable performance at 7T. The figure shows a 45-year-old woman with pathologically confirmed oligodendroglioma. A mass with predominantly mild heterogeneous hyperintensity on T2-weighted imaging was observed in the right frontal lobe, extending into the knee of the corpus callosum, with small areas of hypointensity. Susceptibility-weighted imaging (SWI) sequences revealed signs of bleeding or calcification within the mass, and significant enhancement was noted. ASL sequences demonstrated hyperperfusion in most regions of the lesion.
GRASP MRI: A Decade of Innovation from Bench to Bedside by Li Feng, Kai Tobias Block, Hersh Chandarana, and Daniel K Sodickson (Bernard and Irene Schwartz Center for Biomedical Imaging, Department of Radiology, NYU Grossman School of Medicine, New York, NY, USA / Center for Advanced Imaging Innovation and Research (CAI2R), Department of Radiology, New York University Grossman School of Medicine, New York, NY, USA). This article presents an overview of GRASP MRI, covering its technical foundation, major advances, and clinical applications. The authors share their decade-long journey at New York University (NYU), from the inception and development of GRASP MRI to its successful translation into routine clinical use. The review begins with a historical overview that outlines the motivation and early development of GRASP MRI. This is followed by a detailed summary of the technical framework, clinical implementation, and impact. The subsequent two sections then highlight different methodological extensions and variants of GRASP MRI and their applications. Finally, the review concludes with a discussion of current limitations of this technique and its future directions.
Ultra-Fast Pediatric Brain MRI for Toddlers and Young Children Using Deep Learning Acceleration by Sebastian Altmann, M.D.; Vanessa Schöffling, M.A.; Ahmed Othman, M.D., MHBA (Johannes Gutenberg University (JGU) Medical Center Mainz, Department of Neuroradiology, Center for Imaging, Minimally Invasive and Molecular Therapy, Mainz, Germany). The authors describe the successful clinical implementation of an ultra-fast, T2-weighted, deep learning-enhanced MRI protocol tailored to pediatric patients aged from 6 months* to 6 years. They present a detailed overview of their workflow and practical recommendations for patient preparation, positioning, and scan execution, all aimed at optimizing outcomes in both routine follow-ups and emergency pediatric imaging scenarios. *Please note that MR scanning has not been established as safe for imaging fetuses and infants less than two years of age. The responsible physician must evaluate the benefits of the MR examination compared to those of other imaging procedures.
Mobile MRI: Revolutionizing Access to Advanced Imaging by Jakob Krebs Christensen (Agito Medical A/S, Nørresundby, Denmark). The evolution of mobile MRI units represents a remarkable convergence of medical imaging technology and transportation engineering, driven by the fundamental challenge of bringing diagnostic capabilities to underserved populations and remote locations. Today’s mobile MRI units serve diverse populations worldwide, from rural communities in high-income nations to underserved regions in low- and middle-income countries. These systems have proven particularly valuable in disaster response scenarios, military medical support, and population health screening programs. The concept of mobile units continues to evolve with advances in artificial intelligence, power and helium optimization, and remote scanning. The industry is increasingly focusing on patient comfort and accessibility. Patient-centered technology, such as wide-bore systems, in-system entertainment, low-acoustic noise scanning, lightweight coils, and free-breathing scanning, will continue to be an important goal.
Conventional high-field MRI scanners make it challenging to examine body regions containing metallic implants. The large difference in magnetic susceptibility between metal and the surrounding tissue causes significant local magnetic field distortion, leading to the familiar image artifacts around implants. Even with advanced techniques such as WARP and SEMAC, image quality on conventional 1.5T and 3T MRI systems remains limited, making accurate diagnosis difficult. At a lower field strength of 0.55T, the systems of the MAGNETOM Free. Platform have the advantage of producing significantly fewer susceptibility artifacts than high-field systems. With far fewer distortions and artifacts near implants, these lower-field scanners enable better visualization of pathologies, despite the lower resolution and signal-to-noise ratio. As in high-field imaging, these examinations use turbo spin echo (TSE) sequences with optimized acquisition parameters that are activated via the WARP option. Increased excitation and readout bandwidths are crucial here, especially in combination with STIR technology as a robust alternative to conventional spectral fat saturation for fat suppression. In addition to established advanced acceleration techniques (parallel imaging and Simultaneous Multi-Slice acquisition), deep-learning-based image reconstruction methods offer enormous potential also at lower field strengths. Enabling high-quality diagnostic MR images in very acceptable acquisition times on 0.55T systems.