Leveraging Radiomics and Artificial Intelligence with Mammography for Early Breast Cancer Detection
Description: Breast cancer (BCa) is the most common cancer among women, second only to skin cancer in the United States. Although the number of BCa deaths has decreased over time, it is still one of the most common causes of cancer-related deaths in women overall and the most common among Hispanic women. Early detection of BCa via screening programs has helped reduce mortality. Imaging modalities such as contrast enhanced mammography (CEM) and magnetic resonance imaging (MRI) have greatly improved BCa detection; however, these modalities are not suitable for routine screening due to high cost, limited availability or exposure to radiation and contrast agents. 2D mammography (2D MG) has been routinely used for BCa screening; however, it has a high false positive rate, a high false negative rate and cannot provide clinical insight until a lesion is visible on imaging, which limits early detection. Volumetric mammographic imaging of the breast using digital breast tomosynthesis (DBT) has been reported to improve BCa detection; however, DBT is associated with a higher radiation exposure to the patient. The increased availability of artificial intelligence (AI) and imaging informatics tools have made it possible to detect pre-malignancy signal and triage women with different risk profiles. In this proposed study, we will combine the strength of advanced image processing methods, AI and advanced statistical modeling, to derive an innovative biomarker: pre-malignancy index (PMI). By leveraging PMI, we can further construct a digital twin for breast health. Under this framework, we can aggregate health information including family history, lifestyle, genomic makeup, and observable physical traits to project the future trajectory of breast health. Furthermore, using digital twin framework based on PMI, we can monitor BCa risk and simulate responses from different lifestyle modifications in minimizing the risk of BCa.
Amount: $238,800
Funding type: grant
Funder scheme: Innovative, Developmental, and Exploratory Award (IDEA)
Start date: 2024-08-01
End date: 2026-07-31
Investigator: Bino Varghese
Institution: University of Southern California
Funded outputs count: 0
Funder award ID: B30IB8476
OpenAlex ID: G6489516732