Breast imaging is used to observe breast cancer and other abnormalities with the breast. Magnetic resonance imaging (MRI), in particular, can show details about the tissue in the breast that may change the course of a patient's treatment. New innovations in breast MRI could lead to even more information for physicians and oncologists.

How does breast MRI work?

MRI relies on the strong magnetic field of a super-conducting magnet to create signals from the patient's body. This is done using pulse sequences to change the magnetic field in a specific way at specific intervals. These signals are transmitted by objects called coils placed near the region of interest (the area of the body being imaged). MRI does not expose patients to ionizing radiation, which computed tomography (CT) and mammography (MM) would.

Breast MRI may be used to observe and monitor a patient with breast cancer.¹ It can be used to help measure the size of the tumor and check for additional lesions in one or both breasts. In some cases, breast MRI may detect lesions not seen on mammogram.¹ If mammogram shows potential cancer, MRI may be used to investigate further, which could prevent unneeded biopsies. It is also particularly helpful in patients with dense breast tissue.²

During an MRI of the breast, the patient lays face-down on the patient table with dedicated breast coils and an MRI contrast is injected.¹ The MRI then collects the data that is then turned into images. The images showcase multiple angles of the soft tissue of the breast. Lesions may show up on MRI as abnormalities in the soft tissue.

Advancements in breast MRI

Two additional tools that may help with the detection and monitoring of breast cancer include MRI contrast media and artificial intelligence (AI). The information provided by both of these tools can be valuable for radiologists and oncologists when observing the data.

MRI contrast agents

MRI contrast agents are typically gadolinium-based. There is little risk currently associated with gadolinium contrast. However, patients with abnormal kidney function should talk to their doctor prior to having an MRI with contrast, as these patients may have an adverse reaction. Also, the body may retain traces of gadolinium or, in rare cases, allergic reactions may occur.^3,4,5 This contrast does not contain ionizing radiation.

Once the contrast has been injected, the contrast highlights potentially cancerous breast tissue.⁶ Because cancers need an increased blood supply to grow, the contrast may become more concentrated in areas of malignancy. This allows the radiologist to determine possibly cancerous lesions. After this has happened, more tests, such as biopsy, may be needed.

Artificial intelligence in MR

Artificial intelligence, powered by deep learning, could help inform decisions regarding lesions. Deep learning for medical imaging can be used to train algorithms to detect abnormalities on magnetic resonance images. The typical number of slices used to train an MRI algorithm is at least 20,000, though the process is still being researched.

In one research study, the team trained the algorithm using 6,000 slices that had already been read.⁷ The computer then determined if there were abnormalities apparent on the scans, and what these abnormalities were. If the algorithm was wrong, the information was fed back into the program with the correct diagnosis for the system to learn what the mistake was. This algorithm was able to reliably identify breast tumors in the selected images. The team found that the algorithm may help to detect diagnostic errors with radiologists, and radiologists can help to detect errors with the algorithm. This tool is aimed to help radiologists and make the process more efficient but provides a good example of how radiologists and AI can learn together.

Both artificial intelligence (powered by deep learning) and MRI contrast could help to detect abnormalities in the breast, though artificial intelligence for radiology is still being researched. MRI contrast is used in breast MRI to highlight areas with a higher than normal concentration of blood. It can also help to highlight abnormalities in the soft tissue. Oncologists can then use the information provided through MRI to determine the best treatment options for their patients with breast cancer.

References:

Breast MRI Scans. Cancer.org. https://www.cancer.org/cancer/breast-cancer/screening-tests-and-early-detection/breast-mri-scans.html. Last accessed September 20, 2019.
New Rapid Breast MRI Screening Method Improves Cancer Detection in Dense Breast Tissue. itnonline.com. Last accessed September 25, 2019.
Information on Gadolinium-Based Contrast Agents. FDA.gov. https://www.fda.gov/drugs/postmarket-drug-safety-information-patients-and-providers/information-gadolinium-based-contrast-agents. Last accessed September 20, 2019.
FDA Drug Safety Communication: FDA warns that gadolinium-based contrast agents (GBCAs) are retained in the body; requires new class warnings. FDA.gov. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-warns-gadolinium-based-contrast-agents-gbcas-are-retained-body. Last accessed September 20, 2019.
FDA Drug Safety Communication: New warnings for using gadolinium-based contrast agents in patients with kidney dysfunction. FDA.gov. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-new-warnings-using-gadolinium-based-contrast-agents-patients-kidney. Last accessed September 20, 2019.
How Breast MRI is performed. BreastCancer.org. https://www.breastcancer.org/symptoms/testing/types/mri/how_performed. Last accessed September 20, 2019.
AI Algorithm Detects Breast Cancer in MR Images. itnonline.com. https://www.itnonline.com/article/ai-algorithm-detects-breast-cancer-mr-images-0. Last accessed September 19, 2019.