Burnside Awarded $1.3 Million Grant to Improve Breast Cancer Diagnosis
Dr. Elizabeth Burnside, associate professor of radiology at the University of Wisconsin School of Medicine and Public Health, has been awarded a new R-01 grant from the U.S. National Library of Medicine.
The grant, worth $1,289,580, goes towards Burnside's project, "Integrating Machine Learning and Physician Expertise for Breast Cancer Diagnosis."
The project will join physicians and computer scientists to create decision-support tools designed to improve the diagnosis of breast cancer by combining machine-learned rules and physician expertise in a probabilistic graphical model.
Furthermore, the proposal addresses the substantial information overload and uncertainty that physicians face in clinical practice, often resulting in errors and overuse.
"The algorithms that we will develop will leverage large data repositories, like we have in Breast Imaging at UW, to help physicians make evidence-based decisions for interventional procedures like breast biopsy," said Burnside, who is a member of the UW Carbone Cancer Center and has an affiliate appointment in the Department of Biostatistics and Medical Informatics.
"Specifically, while we test the hypothesis that these algorithms can improve care, we will simultaneously use these algorithms to discover variables that are highly predictive of breast core biopsy insufficiency."
This research will develop a system that provides individualized risk estimations for biopsy undersampling so future health care teams can engage in shared decision-making with patients as they decide whether excisional biopsy is necessary.
The U.S. National Library of Medicine, which is affiliated with the National Institutes of Health, supports research grants that advance the science of biomedical informatics.
Date Published: 09/15/2011