Breast

Breast Fibroglandular Tissue Segmentation

Dense, fibroglandular breast tissue is an established risk factor for breast cancer in women. However, visual analysis is often limited by subjectivity and inter-reader variability. We have developed AI tools capable of automatically segmenting fibroglandular breast tissue from the surrounding breast tissue. This will allow for clinicians to help better evaluate and stratify a patient’s risk to allow for enhanced surveillance of possible breast cancer.

Breast Lesion Detection

Over 300,000 new cases of breast cancer and over 41,000 breast cancer related-deaths are expected to occur in the United States during 2019. Early detection of cancer is crucial to improving survival. We have developed an automated 3D/2D convolutional neural network in order to enhance detection of breast lesions on MRIs. This will allow for faster, automated detection of suspicious lesions on Breast MRI.