{"doi":"10.1002/jmri.70059","title":"Construction of Phantoms for <scp>MR</scp> Electrical Properties Tomography (From Structure to Composition): A Guideline From the <scp>ISMRM</scp> Electro‐Magnetic Tissue Properties Study Group","abstract":"MR-based Electrical Properties (EPs) Tomography (MR-EPT) denotes non-invasive electrical conductivity (σe) and permittivity (εr) mapping methods using MR measurements. The lack of standardized tissue-mimicking phantoms hinders reproducibility studies and method comparisons. For example, the NIST/ISMRM MRI system phantom has contributed significantly to standardization efforts in relaxometry, underscoring the importance of having similar benchmarks for MR-EPT. The guidelines presented herein outline the importance of MR-EPT phantom design and construction, focusing on structure, composition, and reliability. The structure should mimic the shape and size of the anatomy of interest. For example, the head can be modeled using a simple spherical geometry or with anatomically realistic designs such as the one presented in [1]. The abdomen can be represented using large elliptic cylindrical phantoms or specialized holders that conform to human anatomy and are commonly used for training purposes [2, 3]. This allows for testing compatibility with standard receive coils and allows the use of established methods to mitigate B0 inhomogeneity artifacts and spurious phase contributions. Phantoms that are too small to sufficiently load the coil could detune the coil elements or couple opposite receivers. Moreover, phantoms with translational symmetry along the longitudinal axis and significantly larger geometry in the Frankfort horizontal orientation (B0 direction) than the anterior-to-posterior and right-to-left (e.g., cylinders with length ≥ 4× radius) orientations are ideal for 2D MR-EPT methods since the derivatives along z of B1 in the midplane of a birdcage coil are approximately null. However, they do not provide a good validity test for human anatomies. To discriminate between different tissue EPs, heterogeneous phantoms should be used. The phantom's compartments can be separated with plastic boundaries (e.g., for liquid-based compounds) or kept in direct contact (gel-based compounds). Liquid-based compounds may be suboptimal due to the risk of flow artifacts in images, which are avoided in gel-based compounds. Gel-based compounds may be put in direct contact [4], but electrolyte diffusion between compartments may alter the conductivity and internal geometry near the interfaces, as highlighted in [5-7]. To avoid these issues, solid structures can be used to separate different compartments. However, due to the significant mismatch in EP between the plastic dividers and adjacent tissue-mimicking materials, these interfaces introduce localized artifacts [1]. In particular, at least one voxel in the magnitude transmit magnetic field is consistently corrupted by noise near the boundary. Figure 1 illustrates this boundary effect in a two-compartment cylindrical phantom, demonstrating how these image artifacts around the plastic dividers limit EP mapping assessment in small structures. Precise quantification of EP reconstruction errors related to plastic boundary thickness and material composition has not yet been established in MR-EPT literature. Phantoms should use deionized water as the solvent. Sodium chloride (NaCl) can be added to raise conductivity [5, 9-11]. Permittivity can be reduced using low-permittivity materials such as polyvinylpyrrolidone (PVP) and sucrose. Although more costly, PVP is preferred, as sucrose can lead to stability issues over time, degradation in highly concentrated solutions, caramelization when exposed to high temperatures, and T2* shortening that may lower the SNR [12]. Ethanol and other organic solvents can also be used for permittivity reduction, but they have either poor water solubility, are volatile, or require light-proof containers [5]. Glycerol can also be used, but it can induce chemical shift artifacts [11] and its high viscosity limits its useful concentration to around 30% and 50% for gel-based and liquid-based phantoms, respectively. Barium titanate (BaTiO3) or calcium titanate (CaTiO3) can inc","journal":"Journal of Magnetic Resonance Imaging","year":2025,"id":534498,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":2,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9261,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1417092,"name":"Alessandro Arduino","orcid":"0000-0002-4829-5130","position":1,"is_corresponding":false},{"id":1406997,"name":"Cornelis A. T. van den Berg","orcid":"0000-0002-5565-6889","position":2,"is_corresponding":false},{"id":1406998,"name":"Zhongzheng He","orcid":"0009-0001-9680-7671","position":3,"is_corresponding":false},{"id":1407000,"name":"Kyu‐Jin Jung","orcid":"0000-0003-2842-1707","position":4,"is_corresponding":false},{"id":1407002,"name":"Dong‐Hyun Kim","orcid":"0000-0002-6717-7770","position":5,"is_corresponding":false},{"id":439963,"name":"Riccardo Lattanzi","orcid":"0000-0002-8240-5903","position":6,"is_corresponding":false},{"id":449703,"name":"Jessica Martinez","orcid":"0000-0002-3274-566X","position":7,"is_corresponding":false},{"id":1417093,"name":"Thierry Meerbothe","orcid":"0009-0009-5736-1038","position":8,"is_corresponding":false},{"id":1417094,"name":"Freddy Odille","orcid":"0000-0001-5260-8905","position":9,"is_corresponding":false},{"id":1417095,"name":"Adriano Troia","orcid":"0000-0002-4199-8220","position":10,"is_corresponding":false},{"id":1417096,"name":"Luca Zilberti","orcid":"0000-0002-2382-4710","position":11,"is_corresponding":false},{"id":1406996,"name":"Stefano Mandija","orcid":"0000-0002-4612-5509","position":12,"is_corresponding":false},{"id":580449,"name":"Ilias I. Giannakopoulos","orcid":"0000-0003-2180-5898","position":0,"is_corresponding":true}],"reference_count":21,"raw_metadata":null,"created_at":"2026-07-19T02:51:47.434742Z","pmid":"40831322","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}