Radiomics in breast cancer: Current advances and future directions

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Breast cancer poses significant global health challenges, with medical imaging crucial for diagnosis and treatment.
• Interpretation of medical images currently relies heavily on expert radiologists and clinicians.
• Radiomics offers a method to extract quantitative features from medical images, potentially aiding clinical decisions.

Purpose of the Study:

• To review the current applications of radiomics in predicting breast cancer clinicopathological indices and clinical outcomes.
• To highlight the integration of radiomics with multi-omics data for a comprehensive understanding of breast cancer.
• To identify limitations hindering clinical adoption and propose future research directions for radiomics in breast cancer.

Main Methods:

• Extraction of high-throughput quantitative imaging features using traditional machine learning or deep learning.
• Standardized radiomic analysis pipelines.
• Integration of radiomic data with multi-omics datasets (e.g., genomics, proteomics).

Main Results:

• Radiomic models demonstrate potential in predicting clinicopathological indices and clinical outcomes in breast cancer.
• Radio-multi-omics studies bridge the gap between phenotypic and microscopic information, offering deeper insights.
• Current deficiencies hinder widespread clinical adoption of radiomic models.

Conclusions:

• Radiomics holds significant promise for improving breast cancer diagnosis, prognosis, and treatment planning.
• Further research and validation are needed to overcome current limitations and facilitate clinical integration.
• Advancing radiomics requires addressing issues related to standardization, validation, and clinical translation.