The National Breast Cancer Foundation estimates that one in eight women in the U.S. will develop breast cancer in her lifetime, and studies show that women with dense and fatty breast tissue have a four- to six-fold risk of cancer. With the advancement of technology and research, we know a lot more about mammography reports and risk factors than ever before. And now, a new AI tool could help locate breast cancer in women with dense tissue that standard images don’t always detect.

RELATED: Scientists May Have Cracked the Code for Predicting Breast Cancer in Dense Tissue.

A new AI tool may more accurately locate cancerous breast tumors.

In a study published in the journal Radiology, researchers unveiled a new AI detection model that can accurately locate cancerous tumors on breast images. The tool isn’t a replacement for magnetic resonance imaging (MRI), but could be used to “enhance the accuracy and efficacy of screening breast MRI,” per the authors.

“AI-assisted MRI could potentially detect cancers that humans wouldn’t find otherwise,” Felipe Oviedo, PhD, lead study author and a senior research analyst at Microsoft’s AI for Good Lab, said in a press release.

But while MRIs are more comprehensive than mammograms, they’re more costly and have a higher false-positive rate. Researchers believe the new AI detection model could bridge the gap and further the progression of women’s health.

Most importantly, perhaps, the AI model could be a life-saving tool for women with dense breast tissue, a factor that increases breast cancer risk, according to the Breast Cancer Research Foundation (BCRF).

RELATED: Doctors Say Most Women Over 65 Don’t Need Pap Smears—Cancer Rates Say Otherwise.

The test could be life-saving for women with dense breast tissue.

As Best Life previously explained, “Dense breast tissue refers to breasts made up of less fatty tissue and more fibrous and glandular tissue—in other words, it’s how your breasts appear on a mammogram. You can not feel dense breast tissue, and it doesn’t cause any pain or changes to the breast.”

On a mammogram, both dense breast tissue and tumors appear white, which (it goes without saying) makes it harder for radiologists to decipher what’s potentially cancerous. It can also be harder to see tumors through dense tissue, adds Cleveland Clinic.

Breast cancer starts in your fibroglandular tissue, which is naturally dense. Unfortunately, “the more fibroglandular tissue that you have in your breast, the greater the chance you’ll develop breast cancer,” says the clinic.

There are four types of dense breast tissue:

Mostly fatty tissue

Scattered fibroglandular breast tissue

Heterogeneously dense breast tissue

Extremely dense breast tissue

Of the four, scattered fibroglandular and heterogeneously dense breast tissue are most common (40 percent of women).

“Dense breast tissue is more common in women under 40, those with a low body mass index (BMI), those with a family history of dense tissue, women who are pregnant or breastfeeding, and those taking hormone replacement therapy (HRT),” reported Best Life.

RELATED: 85% of Unvaccinated Women Will Likely Get This Virus—And New Research Links It to Heart Disease.

How does the tool work, and when will it be available?

The AI anomaly detection model is the brainchild of Olviedo’s research team and clinical investigators at the University of Washington’s Department of Radiology. It’s designed to distinguish between normal and abnormal data and flag anomalies for further assessment. As a result, breast MRI images are given an “estimated anomaly score.”

So, how did they achieve this? The researchers trained the AI model “using data from nearly 10,000 consecutive contrast-enhanced breast MRI exams,” of which included 42.9 percent with heterogeneously dense breasts and 11.6 percent with extremely dense breasts.

Oviedo called the AI tool “a promising solution.”

“Unlike traditional binary classification models, our anomaly detection model learned a robust representation of benign cases to better identify abnormal malignancies,” he said.

Additionally, researchers said the AI tool can produce “a spatially resolved heatmap” for MRIs. With this feature, the heatmap highlights or “colors” regions in the breast image that it detects are abnormal.

This feature was tested on 171 women who received an MRI for either screening or pre-operative evaluation for a known cancer, and 221 women with invasive breast cancer. According to the findings, “the abnormal regions identified by the model matched areas of biopsy-proven malignancy annotated by a radiologist, largely surpassing the performance of benchmark models.”

The researchers say the AI model succeeded in both high prevalence of cancer and low cancer prevalence cases. Oviedo believes that integrating the tool into radiology workflow can “improve reading efficiency.”

“Our model provides an understandable, pixel-level explanation of what’s abnormal in a breast,” Oviedo said. “These anomaly heatmaps could highlight areas of potential concern, allowing radiologists to focus on those exams that are more likely to be cancer.”

Oviedo said the AI tool is still undergoing assessment before it’s ready for clinical application.

We offer the most up-to-date information from top experts, new research, and health agencies, but our content is not meant to be a substitute for professional guidance. When it comes to the medication you're taking or any other health questions you have, always consult your healthcare provider directly.