Researchers at UT Southwestern Medical Center have unveiled a groundbreaking artificial intelligence (AI) tool designed to enhance the detection of metastatic breast cancer, potentially minimizing the necessity for invasive biopsies. The innovative model, developed by a team led by Basak Dogan, M.D., leverages standard magnetic resonance imaging (MRI) in conjunction with machine learning AI to identify axillary metastasis, the presence of cancer cells in the lymph nodes beneath the arms.
Dr. Dogan, a distinguished figure serving as Professor of Radiology, Director of Breast Imaging Research, and a member of the Harold C. Simmons Comprehensive Cancer Center at UT Southwestern, emphasized the significance of accurate nodal status determination in guiding treatment decisions. Traditional imaging methods often lack the required sensitivity to definitively rule out axillary metastasis, necessitating invasive procedures such as radioisotope and dye injection followed by surgical interventions for nodal assessment.
The AI model, detailed in a study published in Radiology: Imaging Cancer, demonstrated superior performance compared to MRI or ultrasound in identifying patients with axillary metastasis. In a clinical context, the AI model exhibited the potential to prevent 51% of unnecessary surgical sentinel node biopsies while accurately detecting 95% of cases with axillary metastasis.
Dr. Dogan highlighted the significance of this advancement in mitigating the risks associated with surgical biopsies, which despite having a low probability of confirming cancer cell presence, still pose side effects and potential complications. By enhancing the ability to exclude axillary metastasis during routine MRI examinations using the AI model, the researchers aim to minimize risks and improve overall clinical outcomes.
The retrospective analysis involved dynamic contrast-enhanced breast MRI data from 350 newly diagnosed breast cancer patients at UT Southwestern and the Moody Center for Breast Health on Parkland Health’s main campus in Dallas. Leveraging a combination of imaging data and clinical parameters, the AI model was trained to enhance the detection of axillary metastasis, offering a promising avenue for more accurate and less invasive diagnostic approaches.
