AI and the Breast Health Journey, Part 1

AI and the Breast Health Journey

In this 5-part Breast Health Series, we explore the challenges related to breast cancer screening and the role AI can play in early detection. In our previous breast health interview series, Dr. Amy Patel discussed the importance of prioritizing breast care and her hope that AI becomes the mainstay of every breast imagining practice across the country.

The Breast Density Dilemma

Breast density is one of the most significant obstacles to early cancer detection. Dense breast tissue makes it more difficult for radiologists to detect cancer on mammograms and women with dense breast tissue are at increased risk for developing breast cancer. AI could play a significant role in optimizing care for patients with dense breast tissue.

What does it mean to have dense breasts? In simple terms, it means that there is a higher amount of fibrous and glandular tissue relative to fatty tissue. About half of women 40 years and older have what the CDC would define as dense breasts. Mammograms identify the composition of the breasts, but one of the main issues with dense tissue is the difficulty of distinguishing normal tissue from cancerous tissue. The American Cancer Society stresses the importance of women with dense breasts continuing to get routine mammograms, as most breast cancer can be detected through mammography.

AI solutions regarding breast density take a couple of different approaches. The first approach is to improve the identification of breast density through mammograms. As an example, Densitas offers AI solutions for standardizing breast density classification at the point of care. The goal is to eliminate as much subjectivity in the readings as possible and streamline patient care. In Italy, there is an AI tool being produced with a similar goal in mind, achieving 89% accuracy in discerning between low and high-density breast tissue. This is particularly beneficial as countries around the world have different classifications and communication standards for breast healthcare.

The second approach is to increase accessibility to supplemental screening methods to support the timely and accurate diagnosis of breast cancer. Common secondary screening methods include breast ultrasound, MRI, nuclear imaging, and perhaps most notable digital breast tomosynthesis (DBT). DBT looks at multiple layers of the breast in a 3D image, providing the ability to catch positives not picked up on 2D mammograms. An issue with 3D mammography is that not all screening centers currently have access to this technology.

Can AI support radiologists in managing patients with breast dense tissue? Will the use of AI enable a cost-efficient, second read on all mammograms? There is a growing body of evidence that tells us AI will be game-changing for patients and radiologists. Artificial Intelligence Advances Breast Cancer Detection highlights the use of AI in reading breast MRIs. Improving Breast Cancer Imaging with Artificial Intelligence outlines the role of AI in helping radiologists read breast ultrasound exams. A team from MIT has developed a predictive analytics model that uses mammograms to forecast breast cancer risk. These are just a few of the inroads AI is making along the breast health journey.

That’s a wrap for part 1 of the Breast Health Series – The Breast Density Dilemma. Join us over the next few weeks as we explore 2D and 3D mammography, the confusion around breast screening guidelines, and the role of AI in breast health.

Picture of Alex uy

Alex uy

Alex is a recent UC San Diego graduate with a degree in economics and communications. His focus is digital marketing, and he has a passion for technology driven healthcare solutions.

Contact Us

CASE STUDY

ARA Health Specialists

Use the button below to download your free case study and learn how our approach to validation has improved the number of clinically significant findings in AI software.

CASE STUDY

Sutter Health

Use the button below to download your free case study and learn how our approach to validation has improved the number of clinically significant findings in AI software.