What Is a Bilateral Mammogram With Tomosynthesis?

A bilateral mammogram with tomosynthesis is a breast cancer screening exam that images both breasts using three-dimensional X-ray technology. “Bilateral” simply means both sides, and “tomosynthesis” refers to a technique where the X-ray source sweeps in a short arc around each breast, capturing a series of low-dose images at different angles. Those images are then reconstructed into thin slices, giving radiologists a layered view of the tissue rather than the single flat picture produced by a conventional mammogram. The approach has become the dominant form of screening mammography in the United States because it finds more cancers and leads to fewer false alarms, though the details of how much better it performs and who benefits most are worth understanding.

How the Technology Works

During a standard two-dimensional mammogram, the breast is compressed between two plates and a single X-ray exposure produces one flat image per view. Tomosynthesis uses the same compression setup, so the physical experience is similar, but the X-ray tube moves along an arc while firing a series of exposures at slightly different angles. Early systems captured about nine equally spaced projections over a 40-degree sweep, a range chosen to balance image quality against radiation dose.¹ Modern machines may vary in the number of projections and the width of the arc, but the principle is the same: a computer reconstructs those angled shots into a stack of thin slices, each about one millimeter thick.

The key advantage is reducing what radiologists call tissue overlap. In a flat 2D image, all the glandular tissue, fat, and any abnormalities are compressed into a single plane, so one structure can hide behind another. When you scroll through individual slices, dense tissue that might mask a small tumor on a flat image is separated out layer by layer. This was the central insight when the technology was first tested on breast phantoms and cadaveric specimens: three-dimensional reconstruction reduced the tissue overlap that obscures lesions in conventional mammography.¹

Synthetic 2D Images and Why They Matter

In the early days of tomosynthesis, a full 2D mammogram was taken alongside the 3D sweep, which meant double the exposures and roughly double the radiation. Manufacturers solved this by developing “synthetic” 2D images: a computer-generated flat picture assembled from the same 3D data, eliminating the need for a separate 2D exposure. These synthetic images have been validated extensively. A recent study found that combining synthetic 2D images with the 3D slices produced sensitivity around 99 percent and specificity around 85 percent, with no meaningful difference from using a traditional 2D mammogram alongside the 3D data.² Synthetic 2D mammography has also been shown to be comparable with traditional digital mammography for assessing breast density, which matters because density classifications drive decisions about supplemental screening.³

One area where synthetic images still trail slightly is the detection of microcalcifications, tiny calcium deposits that can be an early marker of ductal carcinoma in situ. Studies show that synthetic images actually display better contrast for microcalcifications in fatty and moderately dense backgrounds, with improved visibility in roughly 90 percent of cases.⁴ However, when it comes to detecting clusters of microcalcifications, direct 3D slices still outperform the synthetic flat images.⁵ In practice, radiologists look at both the synthetic 2D and the individual 3D slices together, so the two views complement each other.

Cancer Detection Rates

The main reason tomosynthesis has been widely adopted is that it finds cancers that conventional mammography misses. A large randomized trial involving over 114,000 women observed a roughly 50 percent increase in cancer detection when tomosynthesis was used compared with standard digital mammography alone. The improvement was particularly strong in women over 50 and, interestingly, in women undergoing their very first screening regardless of age.⁶

A population-based trial in Norway found that cancer detection rates with a single reader using tomosynthesis reached about 8.2 to 8.4 per 1,000 screens, compared with 6.3 per 1,000 for double-reading of conventional mammography. That difference was statistically significant, and it came with single-reader tomosynthesis rather than two readers, which has implications for workflow efficiency in screening programs.⁷

That said, the US Preventive Services Task Force has noted that trials reporting on multiple consecutive rounds of screening generally found no statistically significant difference in cancer detection or tumor characteristics between tomosynthesis and standard mammography over time.⁸ The Task Force considers both modalities effective for screening. The initial round of tomosynthesis screening tends to catch more cancers, likely picking up tumors that were present but hidden on prior 2D exams. Whether that advantage persists over many screening rounds is still being studied.

Fewer Callbacks and False Alarms

Getting called back after a screening mammogram for additional imaging is stressful, and most callbacks turn out to be nothing. Tomosynthesis reduces these false alarms because the 3D slices let radiologists see through overlapping tissue that can mimic an abnormality on a flat image. Studies show that using tomosynthesis in screening decreases recall and false-positive rates, improving the overall effectiveness of breast cancer screening programs.⁹

Research from the Metro Chicago Breast Cancer Registry, which analyzed over a million screening mammograms, found that optimal cancer detection while minimizing unnecessary biopsies fell within a recall rate range of 7 to 9 percent for both 2D digital mammography and 3D tomosynthesis.¹⁰ The practical difference for you is that tomosynthesis makes it easier for a facility to stay in that sweet spot, calling back fewer healthy patients without missing cancers.

Dense Breasts

If you have been told you have dense breast tissue, tomosynthesis is especially relevant. Dense tissue appears white on a mammogram, and so do many tumors, which makes conventional screening in dense breasts a bit like looking for a snowball in a snowstorm. By separating tissue into individual slices, tomosynthesis cuts through this problem. A systematic review looking specifically at women with dense breasts found that when tomosynthesis was combined with mammography, sensitivity ranged from about 83 to 93 percent in symptomatic women, compared with roughly 57 to 81 percent for mammography alone.¹¹

Many states in the US now require that women be notified if they have dense breasts, and some mandate insurance coverage for supplemental screening. Tomosynthesis does not replace ultrasound or MRI for very high-risk patients, but it represents a meaningful step up from conventional mammography for the roughly half of screening-age women who have heterogeneously dense or extremely dense tissue.

Radiation Dose

A reasonable concern is whether the extra images mean extra radiation. The answer depends on how the exam is set up. When tomosynthesis alone was compared with conventional full-field digital mammography, radiation doses ranged from lower to roughly equivalent. Two-view tomosynthesis delivered dose ratios of 0.68 to 1.17 compared with standard digital mammography, meaning it could actually deliver less radiation in some configurations.¹² The old approach of taking a full 2D mammogram plus the 3D sweep roughly doubled the dose, but this has been largely replaced by synthetic 2D imaging, which eliminates that second exposure.

To put the numbers in context, one study measured an average dose of about 2.1 milligray for a standard 2D mammogram and about 2.5 milligray for tomosynthesis at the same breast thickness.¹³ Both figures are well within the regulatory limits set for mammography, and the modest increase from tomosynthesis is widely considered acceptable given the improved cancer detection. For comparison, the radiation from a single bilateral mammogram, whether 2D or 3D, is roughly equivalent to a few weeks of normal background radiation from the environment.

What the Exam Feels Like

The patient experience during tomosynthesis is almost identical to a standard mammogram. Your breast is compressed between two plates, and you hold still for a few seconds while the X-ray tube sweeps. The exam takes slightly longer per view because the tube needs time to move through its arc, but you are talking about seconds, not minutes. The compression is the same kind used in standard mammography, and it remains the part most people find uncomfortable.

Research has explored whether less compression could be used during tomosynthesis, since the 3D slicing reduces the need to flatten tissue as aggressively. One study found that reducing compression significantly lowered both pain and anxiety scores during the exam.¹⁴ This is still an active area of study and is not yet standard practice, but it suggests that tomosynthesis may eventually enable a more comfortable screening experience.

What Happens If Something Shows Up

One practical consequence of tomosynthesis finding more abnormalities is that it also finds more things that need a closer look. A category worth knowing about is architectural distortion, which is a subtle pulling or distortion of the breast tissue visible on imaging. Conventional mammography misses many of these. After facilities switched to tomosynthesis-guided biopsy, the rate of architectural distortion biopsies jumped from about 2 percent to nearly 18 percent of all biopsies performed.¹⁵ Many of these turned out to be benign radial sclerosing lesions rather than cancers, which means more biopsies with benign results. The tradeoff is that some of those distortions do turn out to be cancer and would have been missed entirely on 2D imaging.

When a biopsy is needed, tomosynthesis can also guide the needle. Compared with ultrasound-guided biopsy, tomosynthesis-guided biopsy tends to target different types of findings: asymmetries and architectural distortions rather than solid masses. The positive predictive value for malignancy is lower with tomosynthesis-guided biopsy (roughly a third of biopsied lesions turn out malignant, versus closer to half with ultrasound guidance), because the findings it detects are inherently more ambiguous.¹⁶ This is not a flaw of the technology so much as a reflection of the new territory it opens up: lesions that were invisible before are now visible and biopsy-worthy, even if many are benign.

The Overdiagnosis Question

Finding more cancers is generally good, but it also raises the question of overdiagnosis: detecting cancers that would never have caused symptoms or threatened a person’s life. This concern applies particularly to ductal carcinoma in situ, a non-invasive finding that is sometimes treated aggressively even though some cases would never progress. Modeling studies estimate that around 20 percent of screen-detected ductal carcinoma in situ cases may represent overdiagnosis.¹⁷ Because tomosynthesis detects more early-stage abnormalities, it may contribute to a higher absolute number of overdiagnosed cases. Researchers are still sorting out whether the additional cancers tomosynthesis finds are the clinically significant ones that benefit from early treatment or the slow-growing ones that might never cause harm.

Insurance Coverage and Access Gaps

Tomosynthesis is not explicitly mandated for coverage under the Affordable Care Act in the way that standard mammography is. However, several states have passed their own laws requiring private insurers to cover tomosynthesis without cost sharing.¹⁸ Medicare began covering tomosynthesis in 2015, and most major private insurers now cover it as well, though out-of-pocket costs vary. If your facility offers tomosynthesis and your insurance covers it, you will typically receive it as the default screening method. If you are unsure, it is worth asking before your appointment.

Access is not equal across populations. A large study found that Black women, Hispanic women, and Asian American women were all less likely to receive tomosynthesis compared with white women when both options were available at the same facility. Women with lower educational levels and those living in lower-income zip codes also had lower tomosynthesis uptake.¹⁹ The gap persisted even when the technology was physically available, suggesting that factors like insurance type, out-of-pocket cost differences, and patient-provider communication play a role beyond simple availability.

Artifacts and Image Quality Limitations

No imaging technology is perfect. Tomosynthesis has its own set of image artifacts that radiologists need to recognize. Motion artifacts, caused by any patient movement during the arc sweep, are particularly tricky because the inherent out-of-plane blurring in tomosynthesis makes motion blur hard to distinguish from normal image characteristics.²⁰ Holding still for those few extra seconds genuinely matters for image quality. Other artifacts can arise from dense implants, skin folds, or the reconstruction algorithm itself. Technologists are trained to recognize and correct for these, but they are worth knowing about if your radiologist ever mentions image quality as a reason for a repeat view.

Artificial Intelligence and What Comes Next

One of the biggest practical challenges with tomosynthesis is reading time. A radiologist reviewing a standard mammogram looks at four images (two views of each breast). With tomosynthesis, they may scroll through hundreds of individual slices. Artificial intelligence is being developed specifically to help with this problem. AI systems can flag suspicious areas in the 3D stack, potentially reducing reading times by an average of about 20 percent while also improving diagnostic performance and cutting recall rates.²¹

When compared head-to-head, AI algorithms designed for 3D tomosynthesis significantly outperform older computer-aided detection systems built for 2D mammography. One study found the 3D AI system achieved sensitivity of about 94 percent versus 73 percent for the 2D system, with far fewer false marks on healthy exams.²² These tools are already being deployed in some clinical settings, though broader validation is still underway.²³ The practical promise is that AI could make tomosynthesis screening faster and more accurate, helping offset the added reading burden that comes with 3D imaging and potentially enabling personalized screening intervals based on individual risk profiles extracted from the images themselves.