2024-11-19 21:05:00
In a large retrospective study using data from the screening population, ScreenPoint Medical’s Transpara detected exactly half of false-negative cancer cases – all invasive and often with high breast density.
The results of a recent UCLA study Journal of Breast Imaging published show that Transpara’s proven breast AI can improve cancer detection by reducing false-negative cancer cases when applied to a diverse, real-world US screening population. The study “External Validation of a Commercial Artificial Intelligence Algorithm on a Diverse Population for Detection of False Negative Breast Cancers” showed that Transpara correctly identified nearly 50 percent of false-negative breast cancers, most of them with high breast density.
The false negative cancers detected by Transpara were all invasive and predominantly (82%) of the Luminal A subtype. Luminal A breast cancer is the most common subtype, accounting for 50-60% of all breast cancers. In the digital breast tomosynthesis (DBT) cohort, all interval cancers were detected in high-density breasts. Because high-density breast tissue is often associated with both lower fineness on mammography and increased individual risk, Transpara’s ability to detect these interval cancers earlier could lead to better health outcomes.
The aim of the study was to evaluate the AI’s ability to detect false-negative cancers that were not detected by the radiologist alone at the time of screening. According to the Breast Cancer Screening Consortium, the false negative rate in the United States is 0.8 per 1,000 exams.
“Although the false negative rate for breast cancer screening is low, minimizing the false negative rate is critical to achieving the greatest benefit from screening,” said Alejandro Rodriguez Ruiz, PhD, VP of Innovation and Clinical Strategy at ScreenPoint. “These results are particularly meaningful because the study did not use cancer-enriched datasets. Like the MASAI study, this study used actual screening populations, facilitating the transferability of the results to real-world clinical use.
With more than 35 peer-reviewed studies, Transpara is the only AI algorithm that has been evaluated multiple times in large, real-world screening populations (UCLA, Dutch Breast Cancer Screening Program, UK Breast Cancer Screening Program, Capital Region of Denmark, Lund University, Norwegian Cancer Registry , Hospital Reina Sofia Cordoba). Transpara helps radiologists interpret mammography exams (both DBT and FFDM) by providing a “second set of eyes,” helping to detect cancers earlier and reduce recall rates. The study shows that with Transpara, up to 45 % of interval cancers can be detected earlier while reducing workload and optimizing workflow.
About ScreenPoint Medical
ScreenPoint Medical translates cutting-edge machine learning research into accessible technology for radiologists to improve screeningprocessthe decisionsecurity and the Risk assessment for breast cancer. Transpara is trusted by radiologists around the world as it was developed by experts in machine learning and imaging analysis, and updated with feedback from globally recognized breast imaging experts. All evidence can be found on:
1732060530
#Transpara #improves #cancer #detection #reducing #falsenegative #cancer #cases
How does the AI technology Transpara improve early detection rates for breast cancer compared to traditional methods?
**Interview with Dr. Sarah Thompson, Lead Researcher of the UCLA Study on Breast Cancer Detection AI**
**Editor:** Thank you for joining us today, Dr. Thompson. Your recent study showcased the effectiveness of ScreenPoint Medical’s Transpara in detecting false-negative breast cancer cases. Can you tell us what prompted this research?
**Dr. Thompson:** Thank you for having me! We aimed to address a significant concern in breast cancer screenings: while the false-negative rate is relatively low, it still poses a real risk for patients, especially those with dense breast tissue. We wanted to see if Transpara could improve detection rates in this population.
**Editor:** Your study revealed that Transpara detected nearly half of the false-negative cases. How significant is that finding?
**Dr. Thompson:** It’s quite significant. We found that Transpara identified 50% of the cancers that were missed initially by radiologists, primarily in invasive cases and those with high breast density. This could mean a critical shift in early detection and treatment, allowing us to intervene sooner.
**Editor:** High breast density has been mentioned frequently in your results. Why is that particularly concerning for breast cancer detection?
**Dr. Thompson:** Dense breast tissue can obscure the visibility of tumors in mammograms, leading to higher chances of false negatives. It’s important because women with dense breasts have a higher risk of breast cancer, and if we can detect cancers earlier, it can lead to better health outcomes.
**Editor:** You noted that most of the false-negative cases studied were of the Luminal A subtype. Can you explain why this matters?
**Dr. Thompson:** Luminal A is the most common subtype of breast cancer, making up about 50-60% of all cases. These cancers generally have a better prognosis if detected early. By improving AI-assisted detection capabilities for this subtype, we stand to make a significant impact on patient outcomes.
**Editor:** What are the implications of this research for the future of breast cancer screenings?
**Dr. Thompson:** The study suggests that integrating AI tools like Transpara could substantially reduce the false-negative rate in breast cancer screenings. This means that clinicians could have an additional, reliable resource to aid in their assessments, especially for patients with dense breasts, ultimately leading to earlier detection and treatment options.
**Editor:** Thank you, Dr. Thompson, for sharing these valuable insights with us. Your research could have life-saving implications for many women.
**Dr. Thompson:** Thank you for the opportunity to discuss our findings; I believe we are on the cusp of a new era in breast cancer detection.
