AI Applications to Breast MRI: Today and Tomorrow

Abstract

In breast imaging, there is an unrelenting increase in the demand for breast imaging services, partly explained by continuous expanding imaging indications in breast diagnosis and treatment. As the human workforce providing these services is not growing at the same rate, the implementation of artificial intelligence (AI) in breast imaging has gained significant momentum to maximize workflow efficiency and increase productivity while concurrently improving diagnostic accuracy and patient outcomes. Thus far, the implementation of AI in breast imaging is at the most advanced stage with mammography and digital breast tomosynthesis techniques, followed by ultrasound, whereas the implementation of AI in breast magnetic resonance imaging (MRI) is not moving along as rapidly due to the complexity of MRI examinations and fewer available dataset. Nevertheless, there is persisting interest in AI-enhanced breast MRI applications, even as the use of and indications of breast MRI continue to expand. This review presents an overview of the basic concepts of AI imaging analysis and subsequently reviews the use cases for AI-enhanced MRI interpretation, that is, breast MRI triaging and lesion detection, lesion classification, prediction of treatment response, risk assessment, and image quality. Finally, it provides an outlook on the barriers and facilitators for the adoption of AI in breast MRI. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY: Stage 6.

Keywords: artificial intelligence; breast; deep learning; machine learning; magnetic resonance imaging.

Figure 1.

Radiomics analysis workflow using hand-crafted feature extraction along with traditional machine learning techniques and deep learning for classification and modelling. Reprinted with permission from: Bitencourt A, Daimiel Naranjo I, Lo Gullo R, Rossi Saccarelli C, Pinker K. AI-enhanced breast imaging: Where are we and where are we heading? Eur J Radiol 2021;142:109882.

Figure 2.

Flow diagram illustrating deep learning model triaging of breast magnetic resonance imaging examinations into “extremely low suspicion” and “possibly suspicious” examinations. A, Model training. B, Model testing. MIP, maximum intensity projection; BI-RADS, Breast Imaging Reporting and Data System. Reprinted with permission from: Bhowmik A, Monga N, Belen K, et al. Automated Triage of Screening Breast MRI Examinations in High-Risk Women Using an Ensemble Deep Learning Model. Invest Radiol 2023;58(10):710–719.

Figure 3.

Original CE-MRI images and corresponding color-coded sum entropy feature map as overlay of the tumor area of triple-negative (TN) and HER2-enriched (HER2) breast cancer. TN shows a clearly lower sum entropy than HER2. Reprinted under the terms of the Creative Commons license (http://creativecommons.org/licenses/by/4.0/) from: Leithner D, Horvat JV, Marino MA, et al. Radiomic signatures with contrast-enhanced magnetic resonance imaging for the assessment of breast cancer receptor status and molecular subtypes: initial results. Breast Cancer Res 2019;21(1):106.

Figure 4.

Example of semi-automatic tumor segmentation with dynamic contrast-enhanced MRI—(a) axial, (b) axial detail, (c) sagittal, (d) coronal—for radiomics analysis in a 43-year-old female with a biopsy-proven PD-L1-positive poorly differentiated triple negative breast cancer in the 10:00 axis of the left breast. Reprinted under the terms of the Creative Commons license (http://creativecommons.org/licenses/by/4.0/) from: Lo Gullo R, Wen H, Reiner JS, et al. Assessing PD-L1 Expression Status Using Radiomic Features from Contrast-Enhanced Breast MRI in Breast Cancer Patients: Initial Results. Cancers (Basel) 2021;13(24).

Figure 5.

Artificial intelligence (AI) model architecture schematic. A VGG-19 architecture was trained to classify images into four background parenchymal enhancement (BPE) categories, which were then pooled into “high BPE” and “low BPE” categories. In the maximum intensity projection (MIP) AI model, axial MIPs generated from the first subtraction phase were used as the model input. In the Slab AI model, axial slices from the first subtraction phase were pooled into three maximum intensity slabs, which were each used as an independent model input. CONV = convolutional layer; FC = fully connected layer. Reprinted with permission from: Eskreis-Winkler S, Sutton EJ, D’Alessio D, et al. Breast MRI Background Parenchymal Enhancement Categorization Using Deep Learning: Outperforming the Radiologist. J Magn Reson Imaging 2022;56(4):1068–1076.