[ad_1]
Evaluation of breast density has traditionally been based on subjective examinations and calculations, but the deep learning model – formed on tens of thousands of digital mammograms – distinguishes between different types of breast tissue, fatty tissue at extremely dense, with 90% correlation with the diagnosis of radiologists.
In comparison with traditional prediction models, the researchers used a metric called kappa score, where 1 indicates the model and human experts agree on a diagnosis each time, while lower values indicate less. ; agreements. The maximum kappa score for the existing automatic density evaluation models is about 0.6. In clinical application, the new model got a score of 0.85, which means that it makes better forecasts than previous systems.
"Breast density is an independent risk factor that determines how we communicate with women about their risk of cancer, and our motivation was to create an accurate and consistent tool that can be shared and used in all healthcare systems. health care, "says a PhD student and second author of MIT. on the model's paper, Adam Yala. "It takes less than a second per image … [and can be] easily and cheaply in all hospitals. The researchers are now exploring how the algorithm can be transferred to other hospitals and how it can be used in other healthcare applications.
Source link
