Executive Summary:

Diagnosing breast cancer remains a subjective process, in which radiologists visually assess mammograms to determine whether a patient needs a more detailed MRI. Because this judgment does not follow quantitative criteria, medical care can vary considerably based on the radiologist, hospital policy, affluence of the hospital, and insurance coverage. For the millions of women who are screened for breast cancer each year, we want to quantify their risk assessment to enable early, accurate diagnoses. Quantifying breast cancer risk will assist radiologists performing the screenings, standardize hospital procedure, provide more concrete evidence for insurance coverage, and reduce expenses of treating more progressed cancers. More broadly, machine learning in medical research is often restricted to a single hospital or institution’s data. Such models do not necessarily generalize well when only trained on single-institution data. MammoDL is a software tool that uses convolutional neural networks to quantitatively assess breast tissue density from mammograms to standardize clinicians’ decisions on prescribing MRIs to patients. ​​It incorporates a federated learning pipeline to enable secure training on datasets across multiple institutions. MammoDL’s user-friendly interface will allow for the tool to be configured to individual clinician in order to guide them effectively.

Built With

Share this project:

Updates