Summary from the 2025 International Society for Magnetic Resonance in Medicine workshop on body MRI: Unsolved problems and unmet needs.

To complement the growing interest in body MRI across different B 0 magnetic field strengths and a significant number of abstracts highlighting work at 0.55 T, Clarissa Cooley, PhD (United States), explored the potential of low‐field body MRI. Limited access and the cost of installation, facility, and operation of high‐field MRI units are key reasons why the MR community has been focusing on low‐field systems. Low‐field MRIs are also less limited by the common constraints seen with high‐field scanners, such as susceptibility artifacts (i.e., implants) and safety constraints (i.e., RF heating). Some other advantages of low‐field MRI include lighter weight, smaller footprint, lower power consumption, less acoustic noise generation, better B 1 + homogeneity, shorter T 1 relaxation times, and slightly longer T 2 (and T 2 *) relaxation times. She emphasized the critical dependence of SNR on B 0 field strength. She argues that AI will be especially advantageous for low‐field MRI as it will be able to enhance image quality and improve SNR, and she posits that with appropriate training data, ˜0.05 T body MRI is possible. Although there is significant ongoing research across the ultra‐low‐field (˜0.01 T) to low‐field spectrum (0.1 T), to date, body MRI applications in adults have been successfully demonstrated at 0.55 T in the chest and abdomen and are gaining clinical acceptance and exhibiting diagnostic value. She discussed single‐sided MRI as a possible direction of future development for dedicated low‐field body applications such as prostate, lower spine, and breast, where the magnet is placed on one side of the body, effectively having no bore. Additionally, portable low‐field brain and body MRI may address an unmet need for accessibility and bedside MRI for neonates in the intensive critical unit. 72 , 73 , 74 , 75 , 76 Tom Scheenen, PhD (The Netherlands), demonstrated with examples in oncology where staging for primary tumors such as prostate cancer and tracking of lymph nodes metastases can significantly benefit from higher field strengths, because of the higher SNR, finer spatial resolution, and greater sensitivity to iron oxide nanoparticle‐enhanced T 2 *‐weighted imaging. He reminded the audience that at 7 T, there is no RF body coil surrounding the system gantry and that all imaging coils are transmit/receive. RF inhomogeneity, RF power requirements, and patient safety are important points of attention. The community has exploited individual B 1 + shimming and multi‐transmit technology as potential solutions. The strength of MRI of the body at ultra‐high fields will be in magnetization‐prepared pulse sequences with fast low flip angle readouts. He further reviewed developments in free‐breathing radial stack‐of‐stars acquisitions with respiratory navigators and water–fat separation for high spatial resolution 7 T imaging of organs and lymph nodes in the upper abdomen. Prof. Scheenen concluded his talk by reminding the audience that 7 T should be exploited for X‐nuclei imaging, including 31P and 2H for assessing dynamic tumor metabolism. 77 , 78 , 79 , 80 , 81

National Institutes of Health, Grant/Award Number: 1R13EB037422‐01.

A highlight of the workshop was a candid roundtable discussion between participants and MRI vendor representatives on how the study group members can more effectively collaborate, engage in team science and industry partnerships, and efficiently translate new imaging technologies into clinical practice. Hersh Chandarana, MD, MBA, and Li Feng, PhD (United States), shared the origin story of the golden‐angle radial sparse parallel (GRASP) technique and discussed their team's decade‐long successful journey of bringing the methodology from benchside into clinical practice for use in a free‐breathing motion‐robust technique for DCE‐MRI. Although GRASP and its variants have been used in over 200 000 cases, the team continues to critically assess and study the limitations of the technique, such that further refinements and optimizations can be made. 82 , 83 Richard Ehman, MD (United States), shared inspirational key lessons he learned about advancing MRE from the laboratory bench to standard‐of‐care clinical practice. He conveyed that it is critical to ensure that a new technology or invention is addressing a real‐world problem. Being proactive and engaging with clinical colleagues was emphasized, as was seeking clinical validation and continuously informing stakeholders of developments. He asserted that exclusivity of a technique to a single MRI vendor can limit the widespread dissemination of technology and that it is critical to standardize implementation to enhance generalizability. He further encouraged the audience to investigate partnerships with industry and take advantage of institutional resources that can assist with technology management and transfer. He stressed that collaborations with medical organizations, regulatory agencies, and industry partners can accelerate the advancement of technology into practice and that these bodies should be engaged early in the process. 84 , 85 , 86

The Body MRI Study Group of the International Society for Magnetic Resonance in Medicine (ISMRM) was established in 2023 and aims to facilitate research, promote education, and provides a forum for dialogue between academic and industry partners to foster the development and translation of new MRI technology for abdominal and pelvic imaging that will improve and impact patient care. This inaugural ISMRM‐sponsored Study Group workshop ( https://www.ismrm.org/workshops/2025/Body/ ) was held at the Children's Hospital of Philadelphia's Hub for Clinical Collaboration in Pennsylvania, USA. In addition to invited speakers, 62 proffered abstracts (12 oral, 25 power pitches, and 25 online posters) were accepted. Major themes of the abstracts were liver and kidney function, relaxometry and fat quantification, advanced and robust diffusion‐weighted imaging (DWI), low‐field body MRI, MR fingerprinting (MRF), quantitative imaging biomarkers, artificial intelligence (AI), non‐Cartesian imaging techniques, and the translation of technology into clinical product. Secret audience judges scored and selected the top prizes from the abstract pool (see Table  1 ). Oral, power pitch, and poster presentation awardees and travel stipend recipients. a 1st prize Sherya Ramachandran 2nd prize Jinjia Chen 3rd prize Gastao Cruz 1st prize Jonathan Stelter 2nd prize Tom Griesler 3rd prize Nada Kamona 1st prize Anika Knupfer 2nd prize Corina Margain 3rd prize Elizabeth Huaroc Moquillaza Abbreviations: DCE, dynamic contrast‐enhanced; DWI, diffusion‐weighted imaging; MRF, magnetic resonance fingerprinting; PDFF, proton density fat fraction; TGSE, turbo gradient spin echo. Thank you to the judges (Ryan Brunsing, MD, PhD, Hero Hussain, MD, Sila Kurugol, PhD, Thomas Küstner, PhD, Rina Neeman, MD, Jürgen Machann, PhD, Mark Pagel, PhD, James Pipe, PhD, Kristina Ringe, MD, PhD, Amita Shuka‐Dave, PhD, and Holden Wu, PhD) for their time and effort. Throughout the meeting, attendees were asked to identify and prioritize challenges, opportunities, unsolved problems, and unmet needs in body MRI for future directions of study group activities. This article provides a summary of the event by grouping speaker contributions into several overarching and interconnected themes (see Table  2 ). Consensus points from roundtable dialogue with vendor panelists are also included. Several clinical review talks are additionally summarized in the attached Supporting Information, along with extensive references provided by each speaker. List of unsolved problems, unmet needs, and future directions for the ISMRM Body MRI Study Group. Theme 1: Toward robust, reproducible, and standardized body MRI to promote consistent image quality. Variability and lack of consistency in body MRI protocols and in image quality across facilities still exists and remains a key concern for clinical care and translational research. Research effort and payer reimbursement frameworks for targeted (i.e., problem‐focused, abbreviated) protocols are needed to improve efficiency and patient access. Variability and lack of consistency in body MRI protocols and in image quality across facilities still exists and remains a key concern for clinical care and translational research. Research effort and payer reimbursement frameworks for targeted (i.e., problem‐focused, abbreviated) protocols are needed to improve efficiency and patient access. Theme 2: Dedicated focus on pediatric populations and increased accessibility. One size does not fit all. Body MRI solutions that are designed specifically for the pediatric population are needed. Patient access to body MRI remains challenging worldwide, particularly in lower‐income and under‐resourced areas, and especially for pediatric patients. One size does not fit all. Body MRI solutions that are designed specifically for the pediatric population are needed. Patient access to body MRI remains challenging worldwide, particularly in lower‐income and under‐resourced areas, and especially for pediatric patients. Theme 3: Quantitative imaging biomarkers. Multi‐tissue‐contrast sequences have the potential to improve efficiency for body MRI applications and permit wider adoption of quantitative imaging biomarkers. Repeatable and reproducible quantitative imaging in the body remains challenging, and efforts are needed to harmonize metrics and standards. Recent works in metrology are a step in the right direction. More education is warranted to facilitate clinical translation and vendor involvement is critical. Multi‐tissue‐contrast sequences have the potential to improve efficiency for body MRI applications and permit wider adoption of quantitative imaging biomarkers. Repeatable and reproducible quantitative imaging in the body remains challenging, and efforts are needed to harmonize metrics and standards. Recent works in metrology are a step in the right direction. More education is warranted to facilitate clinical translation and vendor involvement is critical. Theme 4: Artificial Intelligence. High‐quality data training is crucial to the success of AI. Real‐world data is superior to simulated data but is logistically difficult to obtain. Optimal AI‐based solutions may, in some cases, require self‐supervised or unsupervised approaches. AI can play a central role in ensuring consistent image quality and reducing variability across patients and facilities, for example, by improving SNR in signal‐poor sequences, facilitating quantitative imaging accuracy and precision, and reducing the dose or removing the need for exogenous contrast agents. AI integration into clinical practice is in its infancy. There is a critical need to develop systematic mechanisms for quality control and quality assurance from the beginning to avoid unintended pitfalls and maximize clinical adoption. High‐quality data training is crucial to the success of AI. Real‐world data is superior to simulated data but is logistically difficult to obtain. Optimal AI‐based solutions may, in some cases, require self‐supervised or unsupervised approaches. AI can play a central role in ensuring consistent image quality and reducing variability across patients and facilities, for example, by improving SNR in signal‐poor sequences, facilitating quantitative imaging accuracy and precision, and reducing the dose or removing the need for exogenous contrast agents. AI integration into clinical practice is in its infancy. There is a critical need to develop systematic mechanisms for quality control and quality assurance from the beginning to avoid unintended pitfalls and maximize clinical adoption. Theme 5: Low‐ and high‐field MRI. Low‐field systems can increase patient access and reduce cost. Research effort is needed to validate these systems in comparison to high‐field systems and establish specific use cases. High and ultra‐high field platforms can offer SNR advantages with reasonable safety considerations. Effort is needed to also validate these systems in comparison to status‐quo and establish use cases. AI can assist both low‐ and high‐field systems for body MRI. Low‐field systems can increase patient access and reduce cost. Research effort is needed to validate these systems in comparison to high‐field systems and establish specific use cases. High and ultra‐high field platforms can offer SNR advantages with reasonable safety considerations. Effort is needed to also validate these systems in comparison to status‐quo and establish use cases. AI can assist both low‐ and high‐field systems for body MRI. Abbreviation: AI, artificial intelligence.

Data S1: Supporting Information.