As AI adoption in breast cancer screening grows, RadNet’s Enhanced Breast Cancer Detection (EBCD) program addresses clinical and accessibility challenges. In this AXIS Imaging News exclusive, Timothy M. Merchant, RadNet’s national director of screening networks and population health strategy, and Karen Morad, senior vice president of marketing and communications, discuss efforts to overcome barriers, improve affordability, and validate AI’s role in care.
AXIS Imaging News: How is RadNet encouraging enrollment in its self-pay AI breast screening program and addressing adoption barriers?
Timothy Merchant: RadNet, Inc., the largest outpatient imaging service provider in the U.S., launched the Enhanced Breast Cancer Detection (EBCD) service, which complements a patient’s annual breast screening regimen. EBCD includes the application of DeepHealth’s AI-powered breast cancer solution. According to the study, 36% of women are now electing this program, and this number continues to grow. The rate of cancer detection remains substantially higher among these women.
To build trust in AI and demonstrate the clinical effectiveness of its self-pay, AI-enhanced breast cancer screening program, RadNet has implemented several initiatives. A key approach involves showcasing the program’s impressive clinical outcomes. Additionally, outreach efforts target socioeconomically disadvantaged areas, while accessible locations—such as MammogramNow centers in Walmart superstores—and AI algorithms proven effective across patient subgroups, including those with dense breast tissue, aim to expand access to high-quality care.
AXIS: Are there efforts to advocate for insurance coverage of AI in mammography?
Karen Morad: RadNet, the outpatient imaging services provider behind the Enhanced Breast Cancer Detection (EBCD) program, is actively engaging with several health insurers to establish reimbursement for this innovative service. The hope is that the history of digital breast tomosynthesis (DBT) will repeat itself, where initial out-of-pocket costs eventually led to payer coverage. While the EBCD program currently requires an out-of-pocket payment, high adoption rates and positive patient experiences indicate significant interest and potential for broader reimbursement.
AXIS: What types of cancers are detected more often with the AI program, and which subtypes or stages show the greatest impact?
Merchant: At RSNA 2024, data presented indicated that the AI-enhanced breast cancer screening program has shown promising results, particularly in its ability to detect dangerous cancers at rates comparable to routine screening. Women who enrolled in the self-pay option were, on average, 21% more likely to have cancer detected compared to those who did not participate.
AXIS: How does RadNet ensure affordability and accessibility for women interested in the self-pay AI-enhanced screening program, especially in underserved populations?
Morad: RadNet is committed to ensuring affordability and accessibility for women interested in the self-pay, AI-enhanced screening program, particularly in underserved populations. RadNet has observed high adoption rates at its imaging centers in socioeconomically disadvantaged areas, as well as at MammogramNow centers located in Walmart superstores.
These locations help bridge access gaps for women who might otherwise face barriers to receiving quality screening services. Furthermore, the use of DeepHealth’s AI-powered breast cancer detection solution has been shown to raise generalists’ detection rates to levels comparable to those of specialists, expanding access to expert-level mammography.
AXIS: How do radiologists integrate AI as a “second set of eyes” in their workflow, and what feedback have they provided about its effectiveness and usability?
Merchant: Radiologists integrate AI as a “second set of eyes” in their workflow by utilizing AI-powered detection and triaging tools that provide automatic lesion localization, degree of suspicion, and tissue density categorization as decision support during the mammography screening process. The process includes a safeguard review by an expert breast radiologist when discordance arises between the initial reading and the AI analysis. This method ensures that suspicious findings receive thorough scrutiny, enhancing the overall accuracy and reliability of screenings.
Feedback from radiologists about the effectiveness and usability of AI-enhanced workflows has been positive. Many radiologists value AI’s ability to provide valuable insights, increasing their confidence in diagnostic decisions. AI serves as a supportive tool, helping radiologists identify more cancers while streamlining the decision-making process. By acting as an additional layer of review, AI alleviates some of the workload on radiologists, allowing them to focus on more complex cases.
The collaborative use of AI in radiology is thus seen as a beneficial enhancement to traditional methods, facilitating early detection and improving patient outcomes. According to the study, DeepHealth’s solution has led to a 21% increase in cancer detection rates—a significant advancement in breast cancer diagnosis.
AXIS: What specific randomized controlled trial designs are being considered to validate these findings further, and when can results from these trials be expected?
Morad: Similar to the introduction of digital breast tomosynthesis and as is common with medical devices, the evidence supporting the Enhanced Breast Cancer Detection initiative and safeguard review is already so strong that DeepHealth believes this innovation should be made available to as many women as possible without delay. The company is actively engaging with regulators and biostatisticians regarding prospective validation studies.
