Currently, even with more than 3.8 million breast cancer survivors in the United States, the disease is still the second-leading cause of cancer-related death among women in developed countries (only lung cancer affects more women). More than 300,000 women are diagnosed in the US and 40,000 patients die from breast cancer each year, mostly due to under-screening. Black women have the highest death rate from breast cancer attributed to the fact that 1 in 5 with breast cancer have triple-negative breast cancer – more than any other racial/ethnic group. White and Asian/Pacific Islander women are more likely to be diagnosed with localized breast cancer than Black, Hispanic, and American Indian/Alaska Native women. However, patients with breast cancer do have a 99% survivability rate if the disease is detected early. An annual screening mammogram is proven to be the best way to identify breast cancer and the type of screening is instrumental to safe early detection. Breast cancer detection techniques continue to transform healthcare due to recent developments leveraging Artificial Intelligence (AI). AI is disrupting screening and diagnostics tools (such as ultrasound) to enhance breast imaging accuracy and efficiency for better patient outcomes.
Breast Cancer Screening- Mammography
Techniques in mammography include digital mammography (also known as DM, digital 2D mammography, full-field digital mammography, FFDM), synthetic mammography (SM) and digital breast tomosynthesis (DBT). Since FDA approval in 2000, DM has been the most common examination used for breast cancer screening. DBT, also referred to as 3-Dimensional mammography (3D mammography) or tomosynthesis, is superior to DM alone for cancer detection and recall rate and is emerging as the standard of care for breast imaging based on improvements in both screening and diagnostic imaging outcomes. Digital breast tomosynthesis (DBT) involves multiple projections acquired across an arc that are reconstructed into a series of stacked images. DBT is optimized with FFDM, however, concerns over increased radiation dose have prompted the development of synthetic mammography (SM), in which two-dimensional images are reconstructed from the DBT data set to replace the FFDM portion of the examination (“in combination mode”).
Advantages of Digital Breast Tomosynthesis (DBT)
The additional data set obtained from the tomosynthesis acquisition decreases the confounding effect of overlapping tissue, allowing for improved lesion detection, characterization, and localization. Research has shown that DBT improves cancer detection and reduces false-positive recalls compared to screening with digital mammography (DM) alone. The information derived from DBT allows a more efficient imaging work-up than imaging with two-dimensional full-field digital mammography alone.
- Studies show improved screening outcomes with DBT, including lower recall and higher cancer detection rates.
- The invasive cancers detected with DBT tend to be smaller, lower grade, and have a more favorable prognosis.
Disadvantages of Digital Breast Tomosynthesis (DBT)
DBT imaging, however, is associated with increased acquisition time as a result of the larger image set and requires longer reviewing. This increased professional bandwidth is likely to be more consequential as digital breast mammography increasingly becomes the standard-of-care for mammographic imaging. DBT interpretation time is almost 2x that of digital mammograms, but its use is expected to show progressive growth worldwide resulting in an increased burden for the radiologist and higher cost for screening programs.
Breast Cancer Diagnostics- Dense Breast Tissue
Given the high incidence of dense breasts in all American women, clinicians have recognized an urgent need for a national standard on breast density communication to facilitate timely supplemental imaging (such as ultrasound examination). The FDA says a National Breast Density Notification Rule is to be published in Late 2022 or Early 2023. The FDA proposed amendments to the breast density notification portion of the Mammography Quality Standards Act (MQSA) that would require that the mammography report patient-oriented summaries portion indicate low-density or high-density breasts assessment. This would include language specific to “significance of breast density” and radiologists would be also be required to identify 1 of 4 BI-RADS breast density categories (A: almost entirely fatty, B: scattered areas of fibroglandular density, C: heterogeneously dense, and D: extremely dense) for the patient in the mammography report to the referring provider. This standard is to ensure all patients receive equal information about the screening and risk implications of dense tissue and provide equal opportunity to discuss supplemental screening leading to earlier detection,” explained Joann Pushkin, the executive director of www.DenseBreast.info.org.
Artificial Intelligence (AI) and Breast Imaging
Continued implementation of digital breast mammography is associated with improvements in screening outcomes, including increased cancer detection rates and improved specificity across all breast densities. The introduction of SM reconstruction technology to replace the full-field digital mammography portion of the examination can reduce radiation dose while maintaining the optimized outcomes of screening achieved with DBT. To optimize efficiency and accuracy, successful Artificial intelligence (AI)–based screening models have been introduced to detect normal, moderate-risk, and suspicious mammograms in breast cancer screening programs, and save time in the assessment of breast screening examinations. This is true for both 3D and 2D mammographic interpretation and several studies have reported a 19.3% worklist reduction, and McKinney reported a 34.8% reduction at AI testing on digital mammograms as well ultrasound.
4 Women-Led Companies Using AI in Mammography
1. iSono Health , led by CEO Maryam Ziaei, received breast imaging FDA approval for The ATUSA™ system in May 2022. ATUSA is the world’s first artificial (AI) driven, automated and portable 3D breast ultrasound scanner. The ATUSA system is designed to provide physicians, nurses, and medical assistants with hands-free access to advanced 3D visualisation enabling better diagnosis and patient monitoring. The device enables point of care personalised longitudinal monitoring for high-risk patients. ATUSA is intended to improve image acquisition speed and accessibility of breast cancer monitoring for healthcare providers and women across the world.
2. iCAD, led by CEO Stacey Stevens, inked a deal with Solis Mammography to build on iCAD’s flagship artificial intelligence (AI) solution for breast imaging, ProFound AI®, to detect calcifications within the arteries of the breast. This collaboration, announced in October 2022, will focus on using mammography in an opportunistic way to define cardiovascular risk. ProFound AI® enables quantification of the presence of breast arterial calcifications to assess the risk of cardiovascular disease (CVD) in a new application that could identify millions of women at risk for heart disease obtained from their mammogram data.
3. Volpara, led by CEO Teri Thomas, received this October 2022, the Good Design Award, in the Digital Design category, for their innovative breast cancer early detection software, Vopara Analytics. Volpara Analytics optimizes breast imaging quality for earlier detection of breast cancer and risks leveraging artificial intelligence (AI) driven measures. It’s platform includes analytics dashboard, chat for live imaging feedback and a patient hub. The software guides technologists on optimal positioning and compression, resulting in a higher-quality screening program and better personalized detection strategies in screening patients for breast cancer.
4. Hera-MI, led by founder & CEO Sylvie DAVILA , has developed Breast-SlimView®, a patented clinical decision support solution for 2D and 3D mammography based on Artificial Intelligence. The algorithm automatically detects and removes the normal physiological areas (vessels, glandular tissue, fatty tissue and mammary gland) and replaces them by artificial fat byway of it’s technical patent of “negativation” (masking of healthy areas of the breast). The Breast-SlimView® software suite includes an informative analysis tool for 2D/3D mammography imaging and innovative diagnosis reading support. Breast-SlimView® obtained its CE mark approval in 2019. Hera-MI created its subsidiary in the United States in April 2020 establishing collaborations with American imaging centers specializing in breast cancer diagnosis in consideration of the diversity of populations. Hera-MI’s Breast-SlimView software is in progress for FDA 510 (k) clearance.
Next Generation of Breast Cancer Detection
Screening programs remain one of the best tools at our disposal for catching cancer early and improving outcomes for patients. However, even the most innovative of these examinations, like DBT, still include many challenges for patients and healthcare providers. Unfortunately, with advancements in our breast cancer detecting measures, we have additional radiation exposure stemming from overdiagnosis (false positive recall rates) and increasing workloads for radiologists due to queues of image volumes. Optimistically, disruptive Artificial Intelligence technologies are utilizing pattern recognition among thousands of breast ultrasound images to aid clinicians in accurately diagnosing breast cancer. Where AI provides support of clinical decision-making and efficiency, the combination of screening and diagnostics with computer algorithms has shown to be as effective as human radiologists in spotting breast cancer from x-ray images and alleviates the pressure on healthcare systems, globally, enabling better patient outcomes.
Author
Veta Obazenu 