How to start your breast AI journey?

Analogous to reading an extremely dense mammogram, there is a lot of noise in the marketplace as it relates to AI for breast imaging. When you look beyond the hype and hyperbole, how can you know that you can trust your breast AI?

Ultimately, this is about providing accurate and consistent breast cancer screening to all women. Breast AI providers must be committed to improve accuracy and consistency, and reduce unnecessary recalls, a common cause of anxiety for women undergoing mammography.

Yet, this commitment cannot be just a fancy marketing brochure. This commitment must be demonstrated through evidence-based performance, clinically validated through reader studies and prospective and retrospective trials conducted by leading breast imaging researchers around the world.  To truly understand what your breast AI stands for, you must be able to answer the questions:

  • Can I trust this tool?
  • Is it proven?
  • Is it useful?
  • Do I trust the vision?

Proven performance

Transpara’s performance has been evaluated in more than 20 peer-reviewed studies and dozens of conference presentations. All of the data on Transpara’s performance is public, published and readily available:

  • Use of Transpara reduced false positives and radiologists’ workload by up to 62.6% in largest Danish screening program; “An Artificial Intelligence–based Mammography Screening Protocol for Breast Cancer: Outcome and Radiologist Workload,” Radiology
  • Transpara-assisted workflow helps detect more cancers while safely reducing reading workload by 44% with no change in recall rate; Mammography Screening with AI (MASAI) Study; Lancet Oncology
  • The high proportion of cancers with a Transpara score of 10 indicates the promising performance of Transpara, particularly for women with dense breasts. Results on prior mammograms with a Transpara score of 10 illustrate the potential for earlier detection of breast cancers by using AI in screen-reading; “Artificial intelligence in BreastScreen Norway: a retrospective analysis of a cancer-enriched sample including 1254 breast cancer cases”; European Radiology
  • The first time that a breast AI solution is featured in a randomized controlled trial, the study investigated use of AI in breast cancer screening as an alternative to double reading. With AI support, 90% of mammograms classified as “intermediate” or “low” risk and were read by only one reader while the remaining 10% labeled as “elevated” risk were read as usual by two readers.   

Proven utility

Transpara is FDA cleared and has European regulatory approval (CE Mark) for both abnormality detection and density assessment for use with 2D and 3D mammography from multiple manufacturers. It has been used to analyse more than 5 million mammograms – including 1 million tomosynthesis exams – at leading breast centres in more than 40 countries, delivering consistent results across regions and demographics.

Three reader studies involving 670 mammograms and 46 radiologists from both the United States and Europe show that every radiologist, regardless of their experience level, increased their accuracy.

Proven vision

ScreenPoint Medical translates cutting edge machine learning research into technology accessible by radiologists to improve screening workflow, decision confidence and breast cancer risk assessment. Transpara is trusted by radiologists globally because it has been developed by experts in machine learning and image analysis and updated with user feedback from world-renowned breast imagers.

The most clinically validated breast AI on the market, Transpara is the leading breast AI with a singular purpose of providing accurate and consistent breast cancer screening to all women.

Contributed by Sanmitra Iwanski, vice president of marketing, ScreenPoint Medical.

This news story has been sponsored by the companies concerned and does not represent the views or opinions of RAD Magazine.

You might also like

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read more