Breast cancer screening plays a central role in reducing mortality, yet current screening methods face several practical and clinical limitations. Mammography remains the most widely used imaging modality because of its accessibility and cost effectiveness. However, the technique relies on ionising radiation, requires breast compression that can cause discomfort and produces false positive findings in 10.2% to 14.4% of cases. Mammography may also miss between 1% and 35% of breast cancers. Magnetic resonance imaging offers a radiation-free alternative and can visualise soft tissue differences without interference from dense breast tissue. MRI screening also demonstrates low false-negative rates.
Despite these advantages, MRI screening remains constrained by high costs, limited scanner availability and the requirement for intravenous contrast administration. Screening uptake therefore remains incomplete. Ultra-low field magnetic resonance imaging provides a potential alternative imaging strategy. Imaging performed at 6.5 mT demonstrates that breast structures can be visualised without contrast agents or compression, establishing the technical feasibility of ultra-low field MRI for breast imaging applications.
Must Read: Improving MS Lesion Segmentation Across MRI Domains
Ultra-Low Field MRI Imaging Protocol and Evaluation
Breast imaging was performed using a laboratory ultra-low field MRI system operating at 6.5 mT. The imaging protocol included 11 healthy women without a history of breast cancer and three additional patients with known breast disease or a benign mass. Healthy participants underwent imaging of the left breast while positioned prone, with the breast placed in a conical radiofrequency coil located at the scanner isocentre. Imaging used a three-dimensional balanced steady-state free precession sequence with a voxel size of 3 mm × 3 mm × 8 mm, producing image sets within approximately 21 minutes.
Three board-certified breast radiologists evaluated the images. Their assessment focused on breast density patterns and the visibility of key anatomical structures including fibroglandular tissue, breast outline, nipple-areolar complex and chest wall. Breast density patterns were classified into fatty tissue, scattered fibroglandular tissue, heterogeneous fibroglandular tissue and extreme fibroglandular tissue. Agreement among the readers regarding breast tissue pattern classification was substantial, with a Fleiss’ kappa value of 0.73.
Participants tolerated the imaging procedure well. None of the images were degraded by motion and none of the participants reported discomfort during scanning. The breast rested naturally within the coil without compression. The radiologists consistently identified the breast outline and fibroglandular tissue across scans, although the visibility of the nipple-areolar complex and chest wall varied between participants.
Imaging Findings and Comparison with Mammography
Ultra-low field MRI images revealed several key anatomical structures. These included the breast outline, fibroglandular tissue patterns, the nipple-areolar complex and portions of the chest wall. Three of the healthy participants had undergone bilateral screening mammography within eight months of participating in the imaging protocol. Comparison between the ultra-low field MRI images and mammography confirmed that both modalities demonstrated similar fibroglandular tissue patterns.
Three additional patients underwent imaging because of previous breast disease or a known lesion. Two patients had a history of breast cancer and had previously undergone lumpectomy. In one patient with prior invasive ductal carcinoma and papillary carcinoma in the subareolar region of the right breast, ultra-low field MRI visualised several anatomical features including breast outline, fibroglandular tissue, retromammary fat, nipple-areolar complex and chest wall. The images also revealed a linear hypointense structure corresponding to post-surgical scar tissue at the lumpectomy site. Similar findings were visible on clinical 3 T MRI.
A notable difference emerged when comparing imaging techniques. On mammography and conventional high-field MRI, surgical clips produced susceptibility artefacts that appeared as dark distortions surrounding the clips. These artefacts were absent in the ultra-low field MRI images. The absence of susceptibility artefacts allowed unobstructed visualisation of surrounding breast tissue.
A third patient with a palpable breast mass also underwent imaging. Mammography and targeted ultrasound had previously confirmed a benign cystic lesion. Ultra-low field MRI detected the cyst clearly across several image slices located approximately one centimetre above the nipple on the medial side of the breast. Measurements obtained from the three-dimensional MRI images estimated the lesion at 33 mm × 20 mm × 18 mm. These measurements corresponded closely with ultrasound measurements of 35 mm × 26 mm × 16 mm.
Technical Characteristics and Current Limitations
Ultra-low field MRI operates under different physical conditions from conventional MRI systems operating at 1.5 T or 3 T. Reduced magnetic field strength leads to lower signal-to-noise ratio, which can influence image clarity and spatial resolution. The balanced steady-state free precession sequence used in the imaging protocol generates bright signals from both fatty tissue and fluid. This overlap can limit differentiation between these tissue types without additional imaging techniques.
Certain anatomical structures were inconsistently visualised. The nipple-areolar complex was absent in some images, which may reflect anatomical variation, slice thickness or breast positioning. Chest wall visibility also varied between participants. The radiofrequency coil design limited imaging depth to approximately 3 cm beyond the coil plate. In individuals with larger breasts, this restriction prevented full visualisation of deeper anatomical structures.
Spatial resolution represents another limitation. The voxel size used for healthy participants does not meet the spatial resolution typically required for breast cancer screening, which is approximately 1 mm × 1 mm × 3 mm for reliable detection of small tumours. Higher spatial resolution imaging was achieved in the clinical cases but required longer scan durations. Imaging also covered only one breast at a time and did not include the axillary region, an important area for detecting nodal disease.
Ultra-low field MRI performed at 6.5 mT demonstrates that in vivo breast imaging is technically feasible without contrast agents, compression or advanced image reconstruction techniques. Imaging visualised fibroglandular tissue patterns and several key anatomical structures and detected post-surgical changes and benign cystic lesions. The absence of susceptibility artefacts from surgical clips enabled unobstructed visualisation of surrounding tissue. However, current limitations include restricted spatial resolution, incomplete anatomical coverage and limited clinical data. Further technical development and evaluation will be required to determine whether ultra-low field MRI can support future clinical applications in breast imaging.
Source: Scientific Reports
Image Credit: iStock
References:
Shen S, Koonjoo N, Longarino FK et al. (2026) Breast imaging with ultra-low field MRI. Sci Rep; 16, 4518.
