Artificial Intelligence Biomarker Development on Breast Cancer Histology
Our deep learning model for risk of recurrence in HR+/HER2- breast cancer identifies an area of lymphovascular invasion indicating high risk disease.
Welcome to the Howard Lab Website
Our lab’s research focuses on answering several important questions at the intersection of digital health and breast medical oncology: 1) Can artificial intelligence be used to improve prediction of response to therapy in breast cancer, and thus lead to better personalization of therapies? 2) Can deep learning use readily available pathologic and imaging data to improve upon or supplement existing genomic biomarkers in breast cancer in order to reduce cost, prevent unnecessary treatment delays, and improve access to biomarkers? and 3) Given the rapid growth of big data / artificial intelligence tools in oncology, what safeguards need to be in place to ensure these tools do not recapitulate healthcare disparities that are currently prevalent in cancer care?
News
Prediction of Breast Cancer Recurrence from Digital Histology
Gene expression-based recurrence assays are strongly recommended to guide the use of chemotherapy in hormone receptor-positive, HER2-negative breast cancer, but such testing is expensive, can contribute to delays in care, and may not be available in low-resource...
Prediction of Pathologic Complete Response with a Biophysical Simulation Model
Pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) in early breast cancer (EBC) is largely dependent on breast cancer subtype, but no clinical-grade model exists to predict response and guide selection of treatment. A biophysical simulation of...
Characterizing Histologic Batch Effect in The Cancer Genome Atlas
The Cancer Genome Atlas (TCGA) is one of the largest biorepositories of digital histology. Deep learning (DL) models have been trained on TCGA to predict numerous features directly from histology, including survival, gene expression patterns, and driver mutations....