They develop a machine learning model that identifies imaging biomarkers on mammograms to predict the risk of developing breast cancer.
Researchers at Massachusetts General Hospital (MGH) have developed a machine learning model that identifies imaging biomarkers on screening mammograms to predict the a patient’s risk of developing breast cancer. And to be able to do it with greater precision than traditional risk assessment tools.
The Currently available risk assessment models incorporate only a small fraction of patient data, such as family history, previous breast biopsies, and hormonal and reproductive history. Only one feature of screening mammography itself, breast density, is incorporated into traditional models.
Statistic analysis
The study included 245,753 two-dimensional digital bilateral screening mammograms consecutive tests performed on 80,818 patients between 2009 and 2016. Of the total mammograms, 210,819 exams were used in 56,831 patients for training, 25,644 exams of 7,021 patients for testing, and 9,290 exams of 3,961 patients for validation.
Through statistical analysis, the researchers compared the accuracy of the deep learning image-only model to an available risk assessment model to predict future breast cancer within five years of the index mammogram. The deep learning model achieved a predictive rate of 0.71, significantly outperforming the traditional risk model, which reached a rate of 0.61.
«Why should we limit ourselves to just breast density when there is such rich digital data embedded in every woman’s mammogram?«Says lead author Constance D. Lehman, MD, Ph.D., head of the breast imaging division at MGH. “Each woman’s mammogram is unique to her as is her fingerprint. It contains imaging biomarkers that are highly predictive of future cancer risk, but until we had the deep learning tools we couldn’t extract this information to improve patient care. ‘ Add.
–
I am an ‘off-road’ journalist, I write, record and edit videos, layout magazines … I do everything. I am passionate about the world of communication and that ‘bug’ grows every time I enter a new field or specialty related to this field. What are you going to do. We will have to live with it. ????
—-
