Mammography AI Predicts Cancer Before It’s Detected

A new study highlights the predictive power of AI for mammography screening – before cancers are even detected. Researchers in a study JAMA Network Open found that risk scores generated by Lunit’s Insight MMG algorithm predicted which women would develop breast cancer – years before radiologists found it on mammograms. 

Mammography image analysis has always been one of the most promising use cases for AI – even dating back to the days of computer-aided detection in the early 2000s. 

  • Most mammography AI developers have focused on helping radiologists identify suspicious lesions on mammograms, or triage low-risk studies so they don’t require extra review.

But a funny thing has happened during clinical use of these algorithms – radiologists found that AI-generated risk scores appeared to predict future breast cancers before they could be seen on mammograms. 

  • Insight MMG marks areas of concern and generates a risk score of 0-100 for the presence of breast cancer (higher numbers are worse). 

Researchers decided to investigate the risk scores’ predictive power by applying Insight MMG to screening mammography exams acquired in the BreastScreen Norway program over three biennial rounds of screening from 2004 to 2018. 

  • They then correlated AI risk scores to clinical outcomes in exams for 116k women for up to six years after the initial screening round.

Major findings of the study included … 

  • AI risk scores were higher for women who later developed cancer, 4-6 years before the cancer was detected.
  • The difference in risk scores increased over three screening rounds, from 21 points in the first round to 79 points in the third round.
  • Risk scores had very high accuracy by the third round (AUC=0.93).
  • AI scores were more accurate than existing risk tools like the Tyrer-Cuzick model.

How could AI risk scores be used in clinical practice? 

  • Women without detectable cancer but with high scores could be directed to shorter screening intervals or screening with supplemental modalities like ultrasound or MRI.

The Takeaway
It’s hard to overstate the significance of the new results. While AI for direct mammography image interpretation still seems to be having trouble catching on (just like CAD did), risk prediction is a use case that could direct more effective breast screening. The study is also a major coup for Lunit, continuing a string of impressive clinical results with the company’s technology.

Breast Cancer Mortality Falls Again

New data from the American Cancer Society highlight the remarkable strides that have been made against breast cancer, with the U.S. death rate falling 44% over the last 33 years – saving over half a million lives. But the statistics also underscore the work that remains to be done, particularly with minority women. 

The fight against breast cancer has been one of public health’s major success stories.

  • High mammography screening uptake has led to early detection of cancers that can then be treated with revolutionary new therapies. 

Much of the credit for this success goes to the women’s health movement, which has conducted effective advocacy campaigns that have led to …

But breast cancer remains the third most common killer of women after heart disease and lung cancer, and there have been disturbing trends even as the overall death rate falls. 

  • Breast cancer incidence has been rising especially in younger women, and major disparities continue to be seen, particularly with survival in Black women.

The American Cancer Society’s new report represents the group’s biennial review of breast cancer statistics, finding … 

  • In 2024 there will be 311k new cases of invasive breast cancer, 56.5k cases of DCIS, and 42.3k deaths. 
  • The breast cancer mortality rate has fallen 44% from 1989 to 2022, from 33 deaths per 100k women to 19 deaths.
  • Some 518k breast cancer deaths have been averted.
  • The mortality rate ranges from 39% higher than average for Black women to 38% lower for Asian American Pacific Islander women. 
  • The mortality rate is slightly higher than average (0.5%) for White women.
  • The average breast cancer incidence rate is 132 per 100k women, but ranges from 5% higher for White women to 21% lower for Hispanic women.
  • Women 50 years and older will account for most invasive cases (84%) and deaths (91%).

The Takeaway

As Breast Cancer Awareness Month begins, women’s health advocates should be heartened by the progress that’s been made overall. But battles remain, from eliminating patient out-of-pocket payments for follow-up studies to addressing race-based disparities in breast cancer mortality. In many ways, the fight is just beginning. 

The Cost of Extra Cancer Detection

It’s well known that using additional screening modalities beyond traditional 2D mammography can detect more cancers in women with dense breast tissue. But at what cost? A new study in Clinical Breast Cancer documents both the clinical value and the economic cost of supplemental breast imaging technologies. 

2D mammography is the basis for any breast cancer screening program, but the modality’s shortcomings are well known, especially in women with dense breasts. 

  • In fact, the FDA earlier this month began requiring breast imaging providers to notify women of their density status and explain how higher density is a breast cancer risk factor. 

Imaging vendors and clinicians have developed a range of technologies to supplement 2D mammography when needed, ranging from DBT to molecular breast imaging to breast MRI.

  • Each has its own advantages and disadvantages, which can leave many breast imaging providers confused about the best technology to use.

To shed some light, Matthew Covington, MD, of the University of Utah compared detection rates for various supplemental imaging modalities; he then estimated costs for each if it was the only modality used for supplemental imaging with 2D mammography in a U.S. population with 469k detectable breast cancers. 

  • The study assumed that 2D mammography would detect only 41% of cancers – leaving the majority undetected. 

Adding a supplemental modality boosted cancer detection rates, but also screening’s cost …

  • DBT detected 47% of all cancers at a cost of $933M
  • Ultrasound detected 51% at a cost of $1.84B
  • MBI detected 71% at a cost of $4.16B
  • Contrast-enhanced mammography detected 80% at a cost of $3.87B
  • MRI detected 100% at a cost of $6.36B

As the data indicate, MRI is clearly the most effective supplemental modality, but at a cost that’s almost 7X that of DBT. 

The Takeaway

The new data are a fascinating – if sobering – look at the intersection of clinical value and economic cost. They also highlight healthcare’s inconvenient truth: The resources needed to provide the highest-quality care are finite, regardless of whether you’re in a single-payor or fee-for-service system.

Why the FDA’s Density Rule Matters

The FDA’s new rules on reporting breast density to women getting mammograms went into effect on September 10. The implementation has been expected for some time, but this week’s rollout generated a wave of positive press coverage that highlights the importance both of breast density awareness and of breast screening.

The FDA in March 2023 said it would implement a national standard requiring providers to inform women of their breast density, which can obscure lesions on conventional X-ray mammography. 

  • Breast density is also a risk factor for cancer, and patient advocacy groups had been pressuring the FDA to set a standard to replace what has become a patchwork of state-by-state notification rules. 

The FDA’s rules have been incorporated into the Mammography Quality Standards Act, and require that … 

  • Mammography reports include a plain-language patient summary with “an overall assessment of breast density.” 
  • The summary must include specific language that defines breast density, explains its ramifications for detection and cancer risk, and suggests the need for additional imaging tests.

A novel aspect of the new rules is that they were mostly driven by patients – women like JoAnn Pushkin and the late Nancy Cappello who as patients discovered first-hand the shortcomings of X-ray-based mammography for women with dense breast tissue. 

What’s next? Density-awareness proponents are now turning their attention to reimbursement, which for supplemental imaging is inconsistent across the U.S.

  • A fix for the problem – the Find It Early Act – is working its way through Congress, and women’s health advocates lobbied on Capitol Hill this week to try to push the legislation through before the end of the current Congressional session. 

The new reporting landscape also creates opportunities for better software tools to detect and manage breast density and better predict risk in patients with dense breast tissue. 

  • Clinicians already realize that women with dense breasts not only need different screening modalities like MRI and ultrasound, but that they might also require more frequent screening due to their heightened cancer risk. 

The Takeaway

The FDA’s new breast density rules matter for a variety of reasons, from showing the power of patients to change their imaging experience to outlining a future in which risk plays a more prominent role in breast screening. While more work remains to be done, this is a good time to savor the triumph.

AI Detects Interval Cancer on Mammograms

In yet another demonstration of AI’s potential to improve mammography screening, a new study in Radiology shows that Lunit’s Insight MMG algorithm detected nearly a quarter of interval cancers missed by radiologists on regular breast screening exams. 

Breast screening is one of healthcare’s most challenging cancer screening exams, and for decades has been under attack by skeptics who question its life-saving benefit relative to “harms” like false-positive biopsies.  

  • But AI has the potential to change the cost-benefit equation by detecting a higher percentage of early-stage cancers and improving breast cancer survival rates. 

Indeed, 2024 has been a watershed year for mammography AI. 

U.K. researchers used Insight MMG (also used in the BreastScreen Norway trial) to analyze 2.1k screening mammograms, of which 25% were interval cancers (cancers occurring between screening rounds) and the rest normal. 

  • The AI algorithm generates risk scores from 0-100, with higher scores indicating likelihood of malignancy, and this study was set at a 96% specificity threshold, equivalent to the average 4% recall rate in the U.K. national breast screening program.

In analyzing the results, researchers found … 

  • AI flagged 24% of the interval cancers and correctly localized 77%.
  • AI localized a higher proportion of node-positive than node-negative cancers (24% vs. 16%).
  • Invasive tumors had higher median risk scores than noninvasive (62 vs. 33), with median scores of 26 for normal mammograms.

Researchers also tested AI at a lower specificity threshold of 90%. 

  • AI detected more interval cancers at this level, but in real-world practice this would bump up recall rates.  

It’s also worth noting that Insight MMG is designed for the analysis of 2D digital mammography, which is more common in Europe than DBT. 

  • For the U.S., Lunit is emphasizing its recently cleared Insight DBT algorithm, which may perform differently.  

The Takeaway

As with the MASAI and BreastScreen Norway results, the new study points to an exciting role for AI in making mammography screening more accurate with less drain on radiologist resources. But as with those studies, the new results must be interpreted against Europe’s double-reading paradigm, which differs from the single-reading protocol used in the U.S. 

DBT Detects Earlier Cancers in Swedish Tomo Study

A new analysis of a landmark DBT study from Sweden offers more support for the effectiveness of tomosynthesis mammography screening. Published in Radiology, researchers found that DBT screening seems to detect earlier cancers, most likely before they become more aggressive. 

Most U.S. mammography practices have embraced DBT since its approval in 2011, such that 48% of all certified mammography units are DBT and 90% of all facilities have at least one tomosynthesis unit. 

  • But doubts about DBT have persisted, particularly by mammography skeptics who charge that the technology was adopted without conducting randomized controlled trials to prove its value. 

But apart from RCTs, there have been plenty of observational studies in which DBT showed a benefit, one of them being the Malmö Breast Tomosynthesis Screening Trial of almost 15k women in Sweden.

  • First results from MBTST were published in 2018 and showed that single-view DBT screening had a 34% higher cancer detection rate per 1k women than digital mammography (8.7 vs. 6.5), but with a higher recall rate as well (3.6% vs. 2.5%).

In the new study, researchers wanted to see if DBT’s screening benefits persisted over two subsequent screening rounds with conventional digital mammography. 

  • Their assumption was that the cancer detection rate would be lower in subsequent rounds, and there would be fewer slow-growing, less aggressive cancers – a sign of early cancer detection. 

Their analysis found …

  • The cancer detection rate per 1k women was lower in the first (4.6) and second (5.3) rounds compared to the original MBTST
  • Recall rate was 2.1% – also lower 
  • The odds ratio of cancer detection was lower than MBTST in the first (OR=0.46) and second (OR=0.53) follow-up rounds 
  • Invasive cancers were less prevalent in the first round compared to the second round (66% vs. 83%) 

What do the results mean? The implication is that because DBT detected cancers in the initial screening round, there was lower cancer prevalence and less aggressive cancer in follow-up rounds, an effect that wore off as time went on.

The Takeaway

There may never be a randomized controlled trial of DBT due to the ethical problem of denying a live-saving technology to women in a control group. But studies like the MBTST follow-up are important in adding to the body of evidence showing that DBT actually does work.

US + Mammo vs. Mammo + AI for Dense Breasts

Artificial intelligence may represent radiology’s future, but for at least one clinical application traditional imaging seems to be the present. In a new study in Radiology, ultrasound was more effective than AI for supplemental imaging of women with dense breast tissue. 

Dense breast tissue has long presented problems for breast imaging specialists. 

  • Women with dense breasts are at higher risk of breast cancer, but traditional screening modalities like X-ray mammography don’t work very well (sensitivity of 30-48%), creating the need for supplemental imaging tools like ultrasound and MRI.

In the new study, researchers from South Korea tested the use of Lunit’s Insight MMG mammography AI algorithm in 5.7k women without symptoms who had breast tissue classified as heterogeneously (63%) or extremely dense (37%). 

  • AI’s performance was compared to both mammography alone as well as to mammography with ultrasound, one of the gold-standard modalities for imaging women with dense breasts. 

All in all, researchers found …

  • Mammography with AI had lower sensitivity than mammography with ultrasound but slightly better than mammography alone (61% vs. 97% vs. 58%)
  • Mammography with AI had a lower cancer detection rate per 1k women but higher than mammography alone (3.5 vs. 5.6 vs. 3.3)
  • Mammography with AI missed 12 cancers detected with mammography with ultrasound
  • Mammography with AI had the highest specificity (95% vs. 78% vs. 94%)
  • And the lowest abnormal interpretation rate (5% vs. 23% vs. 6%)

The results show that while AI can help radiologists interpret screening mammography for most women, at present it can’t compensate for mammography’s low sensitivity in women with dense breast tissue.

In an editorial, breast radiologists Gary Whitman, MD, and Stamatia Destounis, MD, observed that supplemental imaging of women with dense breasts is getting more attention as the FDA prepares to implement breast density notification rules in September. 

  • They recommended follow-up studies with other AI algorithms, more patients, and a longer follow-up period. 

The Takeaway

As with a recent study on AI and teleradiology, the current research is a good step toward real-world evaluation of AI for a specific use case. While AI in this instance didn’t improve mammography’s sensitivity in women with dense breast tissue, it could carve out a role reducing false positives for these women who get mammography and ultrasound.

US Tomo for Dense Breasts

What’s the best way to provide supplemental imaging when screening women with dense breasts? A new study this week in Radiology offers support for a newer method, whole-breast ultrasound tomography. 

It’s well-known by now that dense breast tissue presents challenges to traditional X-ray-based mammography.

  • In fact, mammography screening’s mortality reduction is far lower in women with dense breasts compared to nondense breasts (13% vs. 41%). 

A variety of alternative technologies have been developed to provide supplemental imaging for women with dense breasts, from handheld ultrasound to breast MRI to molecular breast imaging. 

  • One supplemental technology is whole-breast tomography, developed by Delphinus Medical Technologies; the firm’s SoftVue 3D system was approved by the FDA in 2021 as an adjunct to full-field digital mammography for screening women with dense breast tissue. 

With SoftVue, women lie prone on a table with the breast stabilized in a water-filled chamber that provides coupling of sound energy between the breast and a ring transducer that scans the entire breast in 2-4 minutes.

  • Unlike handheld ultrasound, the scanner provides volumetric coronal images that provide a better view of the fat-glandular interface, where many cancers are located.

SoftVue’s performance was analyzed by researchers from USC and the University of Chicago in a retrospective study funded by Delphinus. 

  • They performed SoftVue scans along with digital mammography on 140 women with dense breast tissue from 2017 to 2019; 36 of the women were eventually diagnosed with cancer. 

In all, 32 readers interpreted the scans, comparing the performance of FFDM with ultrasound tomography to FFDM alone, finding … 

  • Better performance with FFDM + ultrasound tomography (AUC=0.60 vs. 0.54)
  • An increase in sensitivity in women with mammograms graded as BI-RADS 4 (suspicious), (37% vs. 30%) 
  • No statistically significant difference in sensitivity in BI-RADS 3 cases (probably benign), (40% vs. 33%, p=0.08)
  • A mean of 3.3 more true-positive and 0.9 false-negative findings per reader with ultrasound tomography, a net gain of 2.4

The Takeaway

The findings indicate that ultrasound tomography could become a new supplementary tool for imaging women with dense breasts. They are also a shot in the arm for Delphinus, which as a smaller vendor has the challenge of competing with large multinational OEMs that also offer technologies for supplemental breast screening. 

Fine-Tuning AI for Breast Screening

AI has shown in research studies it can help radiologists interpret breast screening exams, but for routine clinical use many questions remain about the optimal AI parameters to catch the most cancers while generating the fewest callbacks. Fortunately, a massive new study out of Norway in Radiology: Artificial Intelligence provides some guidance. 

Recent research such as the MASAI trial has already demonstrated that AI can help reduce the number of screening mammograms radiologists have to review, and for many low-risk cases eliminate the need for double-reading, which is commonplace in Europe. 

  • But growing interest in breast screening AI is tempered by the field’s experience with computer-aided detection, which was introduced over 20 years ago but generated many false alarms that slowed radiologists down. 

Fast forward to 2024. The new generation of breast AI algorithms seems to have addressed CAD’s shortcomings, but it’s still not clear exactly how they can best be used. 

  • Researchers from Norway’s national breast screening program tested one mammography AI tool – Lunit’s Insight MMG – in a study with data obtained from 662k women screened with 2D mammography from 2004 to 2018. 

Researchers tested AI with a variety of specificity and sensitivity settings based on AI risk scores; in one scenario, 50% of the highest risk scores were classified as positive for cancer, while in another that threshold was set to 10%. The group found …

  • At the 50% cutoff, AI would correctly identify 99% of screen-detected cancers and 85% of interval cancers. 
  • At the 10% cutoff, AI would detect 92% of screen-detected cancers and 45% of interval cancers 
  • AI understandably performed better in identifying false-positive cases as negative at the 10% threshold than 50% (69% vs. 17%)
  • AI had a higher AUC than double-reading for screen-detected cancers (0.97 vs. 0.88)

How generalizable is the study? It’s worth noting that the research relied on AI of 2D mammography, which is prevalent in Europe (most mammography in the US employs DBT). In fact, Lunit is targeting the US with its recently cleared Insight DBT algorithm rather than Insight MMG. 

The Takeaway

As with MASAI, the new study offers an exciting look at AI’s potential for breast screening. Ultimately, it may turn out that there’s no single sensitivity and specificity threshold at which mammography AI should be set; instead, each breast imaging facility might choose the parameters they feel best suit the characteristics of their radiologists and patient population. 

USPSTF’s Mammography Letdown?

Last year’s relief that the USPSTF would lower its recommended starting age for breast screening to 40 gave way to frustration this week that the group did not go farther in its final decision on mammography recommendations. 

In a series of papers in JAMA journals this week, the USPSTF tackled a range of breast screening issues, from the age at which screening should start to whether modalities like ultrasound and MRI should be used to supplement conventional mammography.

That was the good news. The bad news is that breast screening advocates mostly got shut out on a variety of other issues, with the USPSTF … 

  • Advising that breast screening be conducted biennially (every two years), rather than annually as most women’s imaging advocates would prefer
  • Declining to raise the recommended upper limit for screening from 74 to 79
  • Declining to recommend supplemental screening with MRI or ultrasound for women with dense breast tissue, even as women express frustration with the lack of reimbursement for these exams

On the positive side, the USPSTF finally weighed in on DBT, stating that the 3D mammography technology is equivalent to digital mammography for breast screening. 

  • But in another disappointment, the group said it couldn’t find any studies stating that DBT was better than 2D digital mammography. 

Given the fierce battles that have been fought over screening guidelines in the last 15 years, what made the USPSTF change its mind on mammography’s starting age? 

  • One big factor is the 2% annual rise in breast cancer incidence in women in their 40s from 2015 to 2019; the higher mortality rates among Black women was another issue (see story below in The Wire).

The Takeaway

The USPSTF’s move to lower its recommended starting age for screening mammography is a welcome – if overdue – change for women, who for 15 years have borne the brunt of the group’s conservative approach to guideline formation. The question remains, is the USPSTF making the same mistake all over again when it comes to supplemental imaging and annual screening? And how long will women have to wait this time until it sees the light?

Get every issue of The Imaging Wire, delivered right to your inbox.

You might also like..

Select All

You're signed up!

It's great to have you as a reader. Check your inbox for a welcome email.

-- The Imaging Wire team

You're all set!