AI Enhances Mammography for Improved Breast Cancer Risk Prediction

Artificial intelligence improves mammography-based risk prediction

The future of breast cancer screening and risk-reducing strategies is being shaped by artificial intelligence (AI), according to a review article published by Cell Press on December 12 in the journal Trends in Cancer. "We discuss recent advances in AI-assisted breast cancer risk prediction, what this means for the future of breast cancer screening and prevention, and the key research needed to progress mammographic features from research into clinical practice," says senior study author Erik Thompson of the Queensland University of Technology in Brisbane, Australia. Breast tissue that appears white on a mammogram is radiologically dense, while breast tissue that appears dark is considered non-dense. It is widely accepted that women with higher mammographic density for their age and body-mass index have a greater risk of breast cancer. In addition, higher density makes breast cancer harder to detect by mammography, known as the "masking effect." Advocacy movements across the world are demanding that women be notified of their mammographic density, with policy changes in the U.S., Canada, and Australia. Mammographic density is guiding the use of supplemental imaging technologies in some places, with ultrasound and magnetic resonance imaging (MRI) providing increased cancer detection rates in clinical studies of women with extremely dense breasts. Yet scientists and clinicians continue to struggle with the complexity arising from the masking effect, the breast cancer risk associated with mammographic density, and how to optimally implement changes in clinical practice. To predict a future breast cancer diagnosis, advanced computational approaches such as deep learning are now being used to analyze mammographic images. In particular, AI methods are uncovering mammographic features that have potential to be stronger predictors of breast cancer risk than any other known risk factor. These features might explain a large proportion of the association between mammographic density and breast cancer risk. The discovery of the risk-predicting AI-generated mammographic features is providing new opportunities to identify women at most risk of developing breast cancer in the future and separating them from those women most at risk of having a breast cancer missed due to the masking effect. "A woman with mammographic features associated with a high risk of breast cancer detection could benefit from more frequent screening or risk-reducing medication," Thompson says. "On the other hand, a longer interval between screens could be provided to a woman with a low chance of breast cancer diagnosis in the next five years. Additionally, a woman with high mammographic density without high-risk mammographic features might benefit from supplementary imaging such as MRI or ultrasound." Research suggests that some AI-generated mammographic features are indicative of early malignancy that is undetectable by radiologist-read mammography, while others may be benign conditions associated with an increased risk of breast cancer. The identity of AI-generated mammographic features that are not identified as cancer or a benign condition remains unclear. "Critically, we need to identify the pathobiology associated with mammographic features and the underlying mechanisms that link them with breast cancer oncogenesis," Thompson says. "This will be essential in establishing their relevance to short- and long-term breast cancer risk, as well as future efforts to reduce that risk."

AI enhances mammography for better breast cancer risk prediction

The future of breast cancer screening and risk-reducing strategies is being shaped by artificial intelligence (AI), according to a review article published on December 12 in the journal Trends in Cancer. "We discuss recent advances in AI-assisted breast cancer risk prediction, what this means for the future of breast cancer screening and prevention, and the key research needed to progress mammographic features from research into clinical practice," says senior study author Erik Thompson of the Queensland University of Technology in Brisbane, Australia. Breast tissue that appears white on a mammogram is radiologically dense, while breast tissue that appears dark is considered non-dense. It is widely accepted that women with higher mammographic density for their age and body-mass index have a greater risk of breast cancer. In addition, higher density makes breast cancer harder to detect by mammography, known as the "masking effect." Advocacy movements across the world are demanding that women be notified of their mammographic density, with policy changes in the U.S., Canada, and Australia. Mammographic density is guiding the use of supplemental imaging technologies in some places, with ultrasound and magnetic resonance imaging (MRI) providing increased cancer detection rates in clinical studies of women with extremely dense breasts. Yet scientists and clinicians continue to struggle with the complexity arising from the masking effect, the breast cancer risk associated with mammographic density, and how to optimally implement changes in clinical practice. To predict a future breast cancer diagnosis, advanced computational approaches such as deep learning are now being used to analyze mammographic images. In particular, AI methods are uncovering mammographic features that have the potential to be stronger predictors of breast cancer risk than any other known risk factor. These features might explain a large proportion of the association between mammographic density and breast cancer risk. The discovery of the risk-predicting AI-generated mammographic features is providing new opportunities to identify women at most risk of developing breast cancer in the future and separating them from those women most at risk of having a breast cancer missed due to the masking effect. "A woman with mammographic features associated with a high risk of breast cancer detection could benefit from more frequent screening or risk-reducing medication," Thompson says. "On the other hand, a longer interval between screens could be provided to a woman with a low chance of breast cancer diagnosis in the next five years. Additionally, a woman with high mammographic density without high-risk mammographic features might benefit from supplementary imaging such as MRI or ultrasound." Research suggests that some AI-generated mammographic features are indicative of early malignancy that is undetectable by radiologist-read mammography, while others may be benign conditions associated with an increased risk of breast cancer. The identity of AI-generated mammographic features that are not identified as cancer or a benign condition remains unclear. "Critically, we need to identify the pathobiology associated with mammographic features and the underlying mechanisms that link them with breast cancer oncogenesis," Thompson says. "This will be essential in establishing their relevance to short- and long-term breast cancer risk, as well as future efforts to reduce that risk."

Harnessing the power of AI to optimize breast cancer management

Cell PressDec 13 2024 The future of breast cancer screening and risk-reducing strategies is being shaped by artificial intelligence (AI), according to a review article published by Cell Press on December 12 in the journal Trends in Cancer. "We discuss recent advances in AI-assisted breast cancer risk prediction, what this means for the future of breast cancer screening and prevention, and the key research needed to progress mammographic features from research into clinical practice," says senior study author Erik Thompson of the Queensland University of Technology in Brisbane, Australia. Breast tissue that appears white on a mammogram is radiologically dense, while breast tissue that appears dark is considered non-dense. It is widely accepted that women with higher mammographic density for their age and body-mass index have a greater risk of breast cancer. In addition, higher density makes breast cancer harder to detect by mammography, known as the "masking effect." Advocacy movements across the world are demanding that women be notified of their mammographic density, with policy changes in the U.S., Canada, and Australia. Mammographic density is guiding the use of supplemental imaging technologies in some places, with ultrasound and magnetic resonance imaging (MRI) providing increased cancer detection rates in clinical studies of women with extremely dense breasts. Yet scientists and clinicians continue to struggle with the complexity arising from the masking effect, the breast cancer risk associated with mammographic density, and how to optimally implement changes in clinical practice. To predict a future breast cancer diagnosis, advanced computational approaches such as deep learning are now being used to analyze mammographic images. In particular, AI methods are uncovering mammographic features that have potential to be stronger predictors of breast cancer risk than any other known risk factor. These features might explain a large proportion of the association between mammographic density and breast cancer risk. The discovery of the risk-predicting AI-generated mammographic features is providing new opportunities to identify women at most risk of developing breast cancer in the future and separating them from those women most at risk of having a breast cancer missed due to the masking effect. "A woman with mammographic features associated with a high risk of breast cancer detection could benefit from more frequent screening or risk-reducing medication," Thompson says. On the other hand, a longer interval between screens could be provided to a woman with a low chance of breast cancer diagnosis in the next five years. Additionally, a woman with high mammographic density without high-risk mammographic features might benefit from supplementary imaging such as MRI or ultrasound." Erik Thompson, Queensland University of Technology Research suggests that some AI-generated mammographic features are indicative of early malignancy that is undetectable by radiologist-read mammography, while others may be benign conditions associated with an increased risk of breast cancer. The identity of AI-generated mammographic features that are not identified as cancer or a benign condition remains unclear. "Critically, we need to identify the pathobiology associated with mammographic features and the underlying mechanisms that link them with breast cancer oncogenesis," Thompson says. "This will be essential in establishing their relevance to short- and long-term breast cancer risk, as well as future efforts to reduce that risk." Cell Press Journal reference: Ingman, W. V., et al. (2024) Artificial intelligence improves mammography-based breast cancer risk prediction. Trends in Cancer. doi.org/10.1016/j.trecan.2024.10.007.

AI shaping the future of breast cancer risk predic | Newswise

A new publication by a national collective of researchers has highlighted the potential for the use of artificial intelligence (AI) in identifying women with increased breast cancer risk. The piece, published in Trends in Cancer, explores how AI can help clinicians to better identify features on a mammogram that indicate a high risk of developing breast cancer. The University of Adelaide's Associate Professor Wendy Ingman, part of the Robinson Research Institute and based at The Queen Elizabeth Hospital, was lead author on the publication, which also featured experts from QUT, University of Melbourne, Peter MacCallum Cancer Centre and University of Western Australia. "Artificial intelligence is enabling us to delve deeply into the information inherent in a mammogram and identify novel features associated with higher risk of a future breast cancer diagnosis," said Associate Professor Ingman. The patterns of white and dark on a mammogram have long been studied as mammographic breast density, which is a known risk factor for breast cancer. It's within these patterns of mammographic density that AI is now finding new mammographic features that can be used to identify those women most at risk of a future breast cancer diagnosis. "AI methods are now uncovering mammographic features that are stronger predictors of breast cancer risk than any other known risk factor," said Associate Professor Ingman. Professor Rik Thompson, Professor of Breast Cancer Research and Domain Leader, Centre for Genomics and Personalised Health and School of Biomedical Sciences, QUT, was senior author of the article. "There are a growing number of studies from Australia and internationally suggesting that AI-generated mammographic features are indicative of early malignancy, undetectable by radiologists, but may also represent benign conditions like atypical ductal hyperplasia, which is associated with an increased risk of breast cancer," said Professor Rik Thompson. "Certain mammographic features could be areas of high oncogenic activity that increases the chance of cancer developing." "Critically, we need to identify the pathobiology associated with mammographic features and the underlying mechanisms that link them with breast cancer oncogenesis. It is this common goal that brings us together." Associate Professor Helen Frazer, a breast radiologist leading research studies that investigate use of AI-generated risk-scores within the BreastScreen Victoria program, said research in this space could create new opportunities to improve breast cancer screening, tailored to suit individual needs. "Use of AI could help us identify those women at increased risk of developing breast cancer in the future and be a step forward in personalising screening to best suit the individual and improve outcomes," said Associate Professor Frazer. Gerda Evans, breast cancer survivor and Co-Chair of the Australian Breast Density Consumer Advisory Council, has been working side-by-side with researchers exploring how AI can help refine mammography-based risk prediction. "This is a great advance in predicting breast cancer risk, with potentially huge benefits for the community," said Mrs Evans. Associate Professor Ingman said mammographic density is still a valuable measure of risk at the time of a mammogram. "AI is enabling us to refine mammographic density as a risk factor, and hone in on particular features in a mammogram that are stronger risk predictors, however high mammographic density remains a significant breast cancer risk factor," said Associate Professor Ingman. "More information about mammographic breast density can be found on the InforMD website that our research team developed to help de-mystify this breast cancer risk factor." Tragically, one of the scientists involved in this research passed away before the work was published. Professor John Hopper from the University of Melbourne was passionate about the potential for AI-generated mammographic features to shape the future of breast cancer screening. "With this work, we intend to continue John's legacy," said Professor Thompson.

AI Reads Multiple Mammograms to Help Predict Breast Cancer Risk

FRIDAY, Dec. 6, 2024 (HealthDay News) -- A new AI can help identify women at higher risk for developing breast cancer by tracking changes in breast tissue, a new study shows. The AI compares women's own mammograms over time, looking for early signs of breast cancer that are tough to see even by a well-trained specialist, researchers said. "Our new method is able to detect subtle changes over time in repeated mammogram images that are not visible to the eye," said lead researcher Shu (Joy) Jiang, an associate professor of surgery at Washington University School of Medicine in St. Louis. The AI identified women at high risk of breast cancer 2.3 times more accurately than standard screening methods, researchers reported Dec. 5 in the journal JCO Clinical Cancer Informatics. "We are seeking ways to improve early detection, since that increases the chances of successful treatment," said senior researcher Dr. Graham Colditz, associate director of the Siteman Cancer Center at Barnes-Jewish Hospital with Washington University in St. Louis. "This improved prediction of risk also may help research surrounding prevention, so that we can find better ways for women who fall into the high-risk category to lower their five-year risk of developing breast cancer," Colditz added in a university news release. For the study, researchers built an AI that discerns subtle differences in mammograms, including changes in density, texture, calcification and asymmetry within the breasts. The team trained the AI on mammograms taken of more than 100,000 women who received breast cancer screening at Siteman Cancer Center between 2008 and 2012. Those women were followed through 2020, and nearly 500 subsequently developed breast cancer. The researchers then tested the AI on a separate set of more than 18,000 women who got mammograms at Emory University in Atlanta between 2013 and 2020. The group included 332 women diagnosed with breast cancer. Women judged to be high-risk by the AI were 21 times more likely to be diagnosed with breast cancer within five years, compared to those at lowest risk, researchers found. About 53 out of every 1,000 women in the high-risk group developed breast cancer, compared to fewer than 3 out of every 1,000 women in the low-risk group, results showed. Researchers now are testing the AI in women of diverse racial and ethnic backgrounds, to make sure it is equally accurate for everyone. The research team has a patent pending on their AI. More information The U.S. Centers for Disease Control and Prevention has more on breast cancer screening.

New AI method enhances breast cancer risk prediction from mammograms

Washington University School of MedicineDec 6 2024 A new study from Washington University School of Medicine in St. Louis describes an innovative method of analyzing mammograms that significantly improves the accuracy of predicting the risk of breast cancer development over the following five years. Using up to three years of previous mammograms, the new method identified individuals at high risk of developing breast cancer 2.3 times more accurately than the standard method, which is based on questionnaires assessing clinical risk factors alone, such as age, race and family history of breast cancer. The study is published Dec. 5 in JCO Clinical Cancer Informatics. We are seeking ways to improve early detection, since that increases the chances of successful treatment. This improved prediction of risk also may help research surrounding prevention, so that we can find better ways for women who fall into the high-risk category to lower their five-year risk of developing breast cancer." Graham A. Colditz, MD, DrPH, senior author, associate director of Siteman Cancer Center, based at Barnes-Jewish Hospital and WashU Medicine, and the Niess-Gain Professor of Surgery This risk-prediction method builds on past research led by Colditz and lead author Shu (Joy) Jiang, PhD, a statistician, data scientist and associate professor of surgery in the Division of Public Health Sciences at WashU Medicine. The researchers showed that prior mammograms hold a wealth of information on early signs of breast cancer development that can't be perceived even by a well-trained human eye. This information includes subtle changes over time in breast density, which is a measure of the relative amounts of fibrous versus fatty tissue in the breasts. For the new study, the team built an algorithm based on artificial intelligence that can discern subtle differences in mammograms and help identify those women at highest risk of developing a new breast tumor over a specific timeframe. In addition to breast density, their machine-learning tool considers changes in other patterns in the images, including in texture, calcification and asymmetry within the breasts. "Our new method is able to detect subtle changes over time in repeated mammogram images that are not visible to the eye," said Jiang, yet these changes hold rich information that can help identify high-risk individuals. At the moment, risk-reduction options are limited and can include drugs such as tamoxifen that lower risk but may have unwanted side effects. Most of the time, women at high risk are offered more frequent screening or the option of adding another imaging method, such as an MRI, to try to identify cancer as early as possible. "Today, we don't have a way to know who is likely to develop breast cancer in the future based on their mammogram images," said co-author Debbie L. Bennett, MD, an associate professor of radiology and chief of breast imaging for the Mallinckrodt Institute of Radiology at WashU Medicine. "What's so exciting about this research is that it indicates that it is possible to glean this information from current and prior mammograms using this algorithm. The prediction is never going to be perfect, but this study suggests the new algorithm is much better than our current methods." AI improves prediction of breast cancer development The researchers trained their machine-learning algorithm on the mammograms of more than 10,000 women who received breast cancer screenings through Siteman Cancer Center from 2008 - 2012. These individuals were followed through 2020, and in that time 478 were diagnosed with breast cancer. The researchers then applied their method to predict breast cancer risk in a separate set of patients -; more than 18,000 women who received mammograms through Emory University in the Atlanta area from 2013 - 2020. Subsequently, 332 women were diagnosed with breast cancer during the follow-up period, which ended in 2020. According to the new prediction model, women in the high-risk group were 21 times more likely to be diagnosed with breast cancer over the following five years than were those in the lowest-risk group. In the high-risk group, 53 out of every 1,000 women screened developed breast cancer over the next five years. In contrast, in the low-risk group, 2.6 women per 1,000 screened developed breast cancer over the following five years. Under the old questionnaire-based methods, only 23 women per 1,000 screened were correctly classified in the high-risk group, providing evidence that the old method, in this case, missed 30 breast cancer cases that the new method found. The mammograms were conducted at academic medical centers and community clinics, demonstrating that the accuracy of the method holds up in diverse settings. Importantly, the algorithm was built with robust representation of Black women, who are usually underrepresented in development of breast cancer risk models. The accuracy for predicting risk held up across racial groups. Of the women screened through Siteman, most were white, and 27% were Black. Of those screened through Emory, 42% were Black. In ongoing work, the researchers are testing the algorithm in women of diverse racial and ethnic backgrounds, including those of Asian, southeast Asian and Native American descent, to help ensure that the method is equally accurate for everyone. The researchers are working with WashU's Office of Technology Management toward patents and licensing on the new method with the goal of making it broadly available anywhere screening mammograms are provided. Colditz and Jiang also are working toward founding a start-up company around this technology. Jiang S, Bennett DL, Rosner BA, Tamimi RM, Colditz GA. Development and validation of a dynamic 5-year breast cancer risk model using repeated mammograms. JCO Clinical Cancer Informatics. Dec. 5, 2024. This work was supported by Washington University School of Medicine in St. Louis. Jiang and Colditz have patents pending related to this work, predicting disease risk using radiomic images. Washington University School of Medicine Journal reference: Jiang, S., et al. (2024). Development and Validation of Dynamic 5-Year Breast Cancer Risk Model Using Repeated Mammograms. JCO Clinical Cancer Informatics. doi.org/10.1200/cci-24-00200.

Algorithm analyzes multiple mammograms to improve breast cancer risk prediction

by Julia Evangelou Strait, Washington University School of Medicine A new study from Washington University School of Medicine in St. Louis describes an innovative method of analyzing mammograms that significantly improves the accuracy of predicting the risk of breast cancer development over the following five years. Using up to three years of previous mammograms, the new method identified individuals at high risk of developing breast cancer 2.3 times more accurately than the standard method, which is based on questionnaires assessing clinical risk factors alone, such as age, race and family history of breast cancer. The study is published Dec. 5 in JCO Clinical Cancer Informatics. "We are seeking ways to improve early detection, since that increases the chances of successful treatment," said senior author Graham A. Colditz, MD, DrPH, associate director of Siteman Cancer Center, based at Barnes-Jewish Hospital and WashU Medicine, and the Niess-Gain Professor of Surgery. "This improved prediction of risk also may help research surrounding prevention, so that we can find better ways for women who fall into the high-risk category to lower their five-year risk of developing breast cancer." This risk-prediction method builds on past research led by Colditz and lead author Shu (Joy) Jiang, Ph.D., a statistician, data scientist and associate professor of surgery in the Division of Public Health Sciences at WashU Medicine. The researchers showed that prior mammograms hold a wealth of information on early signs of breast cancer development that can't be perceived even by a well-trained human eye. This information includes subtle changes over time in breast density, which is a measure of the relative amounts of fibrous versus fatty tissue in the breasts. For the new study, the team built an algorithm based on artificial intelligence that can discern subtle differences in mammograms and help identify those women at highest risk of developing a new breast tumor over a specific timeframe. In addition to breast density, their machine-learning tool considers changes in other patterns in the images, including in texture, calcification and asymmetry within the breasts. "Our new method is able to detect subtle changes over time in repeated mammogram images that are not visible to the eye," said Jiang, yet these changes hold rich information that can help identify high-risk individuals. At the moment, risk-reduction options are limited and can include drugs such as tamoxifen that lower risk but may have unwanted side effects. Most of the time, women at high risk are offered more frequent screening or the option of adding another imaging method, such as an MRI, to try to identify cancer as early as possible. "Today, we don't have a way to know who is likely to develop breast cancer in the future based on their mammogram images," said co-author Debbie L. Bennett, MD, an associate professor of radiology and chief of breast imaging for the Mallinckrodt Institute of Radiology at WashU Medicine. "What's so exciting about this research is that it indicates that it is possible to glean this information from current and prior mammograms using this algorithm. The prediction is never going to be perfect, but this study suggests the new algorithm is much better than our current methods." AI improves prediction of breast cancer development The researchers trained their machine-learning algorithm on the mammograms of more than 10,000 women who received breast cancer screenings through Siteman Cancer Center from 2008 -- 2012. These individuals were followed through 2020, and in that time 478 were diagnosed with breast cancer. The researchers then applied their method to predict breast cancer risk in a separate set of patients -- more than 18,000 women who received mammograms through Emory University in the Atlanta area from 2013 -- 2020. Subsequently, 332 women were diagnosed with breast cancer during the follow-up period, which ended in 2020. According to the new prediction model, women in the high-risk group were 21 times more likely to be diagnosed with breast cancer over the following five years than were those in the lowest-risk group. In the high-risk group, 53 out of every 1,000 women screened developed breast cancer over the next five years. In contrast, in the low-risk group, 2.6 women per 1,000 screened developed breast cancer over the following five years. Under the old questionnaire-based methods, only 23 women per 1,000 screened were correctly classified in the high-risk group, providing evidence that the old method, in this case, missed 30 breast cancer cases that the new method found. The mammograms were conducted at academic medical centers and community clinics, demonstrating that the accuracy of the method holds up in diverse settings. Importantly, the algorithm was built with robust representation of Black women, who are usually underrepresented in development of breast cancer risk models. The accuracy for predicting risk held up across racial groups. Of the women screened through Siteman, most were white, and 27% were Black. Of those screened through Emory, 42% were Black. In ongoing work, the researchers are testing the algorithm in women of diverse racial and ethnic backgrounds, including those of Asian, southeast Asian and Native American descent, to help ensure that the method is equally accurate for everyone. The researchers are working with WashU's Office of Technology Management toward patents and licensing on the new method with the goal of making it broadly available anywhere screening mammograms are provided. Colditz and Jiang are also working toward founding a start-up company around this technology. Jiang and Colditz have patents pending related to this work, predicting disease risk using radiomic images.

Analyzing multiple mammograms improves breast cancer risk prediction

"We are seeking ways to improve early detection, since that increases the chances of successful treatment," said senior author Graham A. Colditz, MD, DrPH, associate director of Siteman Cancer Center, based at Barnes-Jewish Hospital and WashU Medicine, and the Niess-Gain Professor of Surgery. "This improved prediction of risk also may help research surrounding prevention, so that we can find better ways for women who fall into the high-risk category to lower their five-year risk of developing breast cancer." This risk-prediction method builds on past research led by Colditz and lead author Shu (Joy) Jiang, PhD, a statistician, data scientist and associate professor of surgery in the Division of Public Health Sciences at WashU Medicine. The researchers showed that prior mammograms hold a wealth of information on early signs of breast cancer development that can't be perceived even by a well-trained human eye. This information includes subtle changes over time in breast density, which is a measure of the relative amounts of fibrous versus fatty tissue in the breasts. For the new study, the team built an algorithm based on artificial intelligence that can discern subtle differences in mammograms and help identify those women at highest risk of developing a new breast tumor over a specific timeframe. In addition to breast density, their machine-learning tool considers changes in other patterns in the images, including in texture, calcification and asymmetry within the breasts. "Our new method is able to detect subtle changes over time in repeated mammogram images that are not visible to the eye," said Jiang, yet these changes hold rich information that can help identify high-risk individuals. At the moment, risk-reduction options are limited and can include drugs such as tamoxifen that lower risk but may have unwanted side effects. Most of the time, women at high risk are offered more frequent screening or the option of adding another imaging method, such as an MRI, to try to identify cancer as early as possible. "Today, we don't have a way to know who is likely to develop breast cancer in the future based on their mammogram images," said co-author Debbie L. Bennett, MD, an associate professor of radiology and chief of breast imaging for the Mallinckrodt Institute of Radiology at WashU Medicine. "What's so exciting about this research is that it indicates that it is possible to glean this information from current and prior mammograms using this algorithm. The prediction is never going to be perfect, but this study suggests the new algorithm is much better than our current methods." AI improves prediction of breast cancer development The researchers trained their machine-learning algorithm on the mammograms of more than 10,000 women who received breast cancer screenings through Siteman Cancer Center from 2008 -- 2012. These individuals were followed through 2020, and in that time 478 were diagnosed with breast cancer. The researchers then applied their method to predict breast cancer risk in a separate set of patients -- more than 18,000 women who received mammograms through Emory University in the Atlanta area from 2013 -- 2020. Subsequently, 332 women were diagnosed with breast cancer during the follow-up period, which ended in 2020. According to the new prediction model, women in the high-risk group were 21 times more likely to be diagnosed with breast cancer over the following five years than were those in the lowest-risk group. In the high-risk group, 53 out of every 1,000 women screened developed breast cancer over the next five years. In contrast, in the low-risk group, 2.6 women per 1,000 screened developed breast cancer over the following five years. Under the old questionnaire-based methods, only 23 women per 1,000 screened were correctly classified in the high-risk group, providing evidence that the old method, in this case, missed 30 breast cancer cases that the new method found. The mammograms were conducted at academic medical centers and community clinics, demonstrating that the accuracy of the method holds up in diverse settings. Importantly, the algorithm was built with robust representation of Black women, who are usually underrepresented in development of breast cancer risk models. The accuracy for predicting risk held up across racial groups. Of the women screened through Siteman, most were white, and 27% were Black. Of those screened through Emory, 42% were Black. In ongoing work, the researchers are testing the algorithm in women of diverse racial and ethnic backgrounds, including those of Asian, southeast Asian and Native American descent, to help ensure that the method is equally accurate for everyone. The researchers are working with WashU's Office of Technology Management toward patents and licensing on the new method with the goal of making it broadly available anywhere screening mammograms are provided. Colditz and Jiang also are working toward founding a start-up company around this technology. Jiang S, Bennett DL, Rosner BA, Tamimi RM, Colditz GA. Development and validation of a dynamic 5-year breast cancer risk model using repeated mammograms. JCO Clinical Cancer Informatics. Dec. 5, 2024. This work was supported by Washington University School of Medicine in St. Louis. Jiang and Colditz have patents pending related to this work, predicting disease risk using radiomic images.