How can we improve early detection of breast cancer and better identify women at risk of advanced or second breast cancer who need additional screening?
This is the mission of a national research team co-led by a UC Davis professor and chair of the Department of Biostatistics Diana Miglioretti. Thanks to the renewal of a $15 million 5-year grant from National Cancer Institute (NCI), the team will use Artificial Intelligence (AI) to make it breast cancer Screening and monitoring are more accurate and equitable.
Breast cancer early screening
Despite great strides in breast cancer diagnosis and treatment, the disease remains The second leading cause of death from cancer for women in the United States. Disease burden varies, with racial and ethnic disparities in breast cancer prognosis, second breast cancer incidence, and even mortality rates.
We are studying whether AI algorithms can improve cancer detection, especially for women from marginalized communities who may not have access to highly experienced breast imaging specialists.” —Diana Miglioretti
A mammogram is diagnosing breast cancer early — when it’s most treatable. However, even with regular screening, some women are diagnosed with advanced cancer. These women may have benefited from extensive or thorough examination.
“US Preventive Services Task Force He recommends screening every two years, which is sufficient for most women. But Miglioretti said, “Some women can benefit from screening every year or with complementary imaging.” She is Professor Dean and Head of the Biostatistics Department at Department of Public Health Sciences at the University of California, Davis and researcher in University of California, Davis Comprehensive Cancer Center. “However, we need to be very careful about the impact of additional screening on women.”
The examination comes with possible damages to False positive results and overdiagnosis, which occurs more frequently with annual screening and examination versus every two years with complementary imaging, such as ultrasound and MRI. The new grant will allow the Miglioretti Research Program to assess whether improvements in the quality of mammograms and regular screening can lead to more equitable health outcomes for women.
“Screening will be most effective and equitable when all women have access to high-quality risk assessment and mammography, and when strategies are targeted to achieve clinically meaningful outcomes,” said the co-programme leader. Anna Tostison. Tostison is the James J. Carroll Professor at the Geisel School of Medicine at Dartmouth.
Artificial intelligence to model fairer breast cancer risk
Miglioretti and her team began studying and promoting safer, more personalized breast cancer screening in 2011. Their program has advanced the science of risk-based screening and surveillance in several ways.
The team has already refined their models based on patient factors, such as age and breast density. Now, researchers are looking to incorporate imaging features (such as calcifications) and artificial intelligence algorithms to make it better at predicting women’s breast cancer risk.
“We are at a point where we have developed risk models for women with or without breast cancer, and now we want to be able to use these models to best select those who need to undergo more intensive screening or monitoring,” Miglioretti said. . “The exciting thing about renewing this grant is the integration of artificial intelligence into these models to identify women who are at high risk of advanced cancer despite regular screening or are at risk for a second cancer missed by their annual mammogram.”
The grant will fund three new projects.
Project 1: More equitable models for breast cancer risk
The first project will use artificial intelligence to predict which women with no history of breast cancer are at high risk of developing advanced cancer. The team will work to develop advanced breast cancer risk models that include imaging features and assessment of FDA-approved AI scores from five vendors. They will compare the benefits and harms mammogram Frequency of breast cancer mortality screening based on advanced cancer risk in women.
Project 2: Using AI to identify factors that contribute to inequalities in breast cancer screening
The second project seeks to identify the factors that lead to inequalities in breast cancer screening. It will explore whether the use of AI detection scores and other facility-wide interventions (such as mobile mammography software) can improve outcomes, with particular attention to health equity.
“We are studying whether AI algorithms can improve cancer detection, especially for women from marginalized communities who may not have access to highly experienced mammography specialists. In fact, many mammograms of women from disadvantaged communities are read. by general radiologists.
The team will evaluate whether the use of artificial intelligence algorithms can improve breast cancer detection and reduce disparities.
“We expect this program to help show how AI can improve breast cancer detection and predict the risk of advanced breast cancer,” said Karla Kirlikowski, professor of medicine, epidemiology and biostatistics at UCSF and co-leader of the program. “This in turn will allow for the development of new, more equitable screening strategies that maximize benefits while minimizing harms.”
Project 3: Reducing Monitoring Failure
Women who receive treatment for breast cancer and who receive full medical certification are monitored. They are required to have annual mammograms to help detect cancer recurrence or a new cancer. Some of these women are diagnosed with a second breast cancer because of the symptoms that occur between the two screens. When this diagnosis occurs before it is time to return to the next screen, this is considered a monitoring failure.
Project 3 seeks to develop a risk-based approach to identify the women most at risk of monitoring failure. You will study the many factors that may be associated with these failures and possible ways to prevent them.
“So, for this project, we ask: Based on the women’s primary cancer stage and their characteristics and personal factors such as age and breast density, what is their chance of missing a second breast cancer by mammography? The answer to this will help us identify those who might benefit from screening Intensive using complementary magnetic resonance imaging.
What is the Breast Cancer Surveillance Consortium?
This research program takes advantage of Breast Cancer Surveillance Consortium (BCSC).
BCSC is a nationwide research network with a robust set of societal data from geographically, socially and demographically diverse settings. It has a long history of evaluating the benefits and harms of different screening methods.
The research team will use the BCSC database to help improve screening and monitoring. Their findings may contribute to public health efforts to promote more balanced risk-based screening and reduce disparities in breast cancer.
“We will use the powerful BCSC database to inform population-level simulations in collaboration with CISNETTostson said. “This simulation will present the long-term impact of integrating risk-based screening into clinical care on mortality.”
CISNET is a consortium of investigators sponsored by NCI. They use simulation modeling to understand cancer interventions and their impact on prevention, screening, and treatment.