Breast Imaging for Computer Aided Diagnosis

Breast cancer is one of the most commonly diagnosed cancers. Breast imaging is essential for screening, diagnosing, and staging breast cancer. An important goal of breast imaging is to identify abnormalities within the breasts, localise abnormalities to facilitate further examination and characterize abnormalities to aid the clinical decision-making process for treatment planning. Researchers and clinicians tend to focus on optimising one of these objectives at a time, as current breast imaging techniques do not always achieve all these objectives. A lot of breast imaging methods map structural or morphological differences in breast abnormalities, like microcalcifications, tissue masses, and breast asymmetry. Now that the vision-language model is being integrated with the model associated with breast abnormalities, the task of detecting, segmenting, and classifying breast abnormalities is becoming appealingly complex. Further, the interpretability and explainability of deep learning-based breast image analysis are becoming increasingly important. There appears to be no single breast imaging method that can both detect and diagnose breast abnormalities with high sensitivity and specificity. Thus, breast abnormalities require new detection and classification methods that are practical, inexpensive, and less harmful to patients.

This Research Topic will focus on artificial intelligence, radiomics and deep learning applications in breast cancer screening, including image quality assessment, breast cancer risk prediction, detection and prognosis, as well as enhancing efficiency in breast cancer care. We are interested in research that explores cutting-edge techniques applied in breast imaging, a critical area in cancer research and treatment. We solicit Original Research articles, as well as Review articles addressing these topics using various imaging modalities (digital mammography, breast tomosynthesis, MRI, ultrasound etc.), with special regard to their clinical and radiological implications for breast cancer screening. We welcome the submission of papers in the following areas including, but not limited to:

• Image processing and analysis in breast imaging
• Vision-Language Model (VLM) for breast imaging
• Explainable artificial intelligence (xai) in breast imaging
• Radiomics in breast imaging
• Image quality and dose in breast imaging
• Risk stratified breast cancer screening
• Multimodality breast imaging
• New X-ray technology for breast imaging
• Virtual clinical trials (VCT) in breast imaging
• Simulation of breast imaging technology
• Anthropomorphic breast phantoms

We encourage authors to submit works related to the latest technological and machine intelligence developments and clinical experiences of novel breast imaging technologies, including digital mammography, tomosynthesis, CT, MR, ultrasound, optical, and molecular imaging. This Research Topic is designed to help advance the fields of breast imaging through the sharing of scientific discoveries, best clinical practices, and industrial innovations.

This Research Topic is developed in collaboration with the 28th UK Conference on Medical Image Understanding and Analysis (MIUA). More details can be found on the conference webpage.

Keywords: breast imaging, breast cancer, digital mammography, breast tomosynthesis, MRI, ultrasound, breast cancer screening, artificial intelligence, deep learning, VLM for breast imaging

Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.