AI can act as your second reader for breast cancer screening programmes

By Pieter Kroese on October 02, 2025

AI can act as your second reader for breast cancer screening programmes
4:56

Breast cancer screening programmes across Europe and across the globe have proven their value time and again by reducing the impact of breast cancer on women, yielding health care savings, and effectively improving the health of populations worldwide.

Yet these programmes face challenges.

Traditionally, global programmes have employed a double reader model, in which two separate radiologists review breast imaging for each patient separately. If their opinions differ, they discuss further and come to a consensus. In this model, two pairs of eyes are better than one.

However, a scarcity of breast imaging specialists combined with an increase in imaging volumes may mean the double reader standard of care is in jeopardy. Two pairs of eyes can be hard to come by. Despite the tireless work of screening staff, the constant demand leads to burnout for radiologists and the potential for missed cancer diagnoses for women.

ScreenPoint Medical sees a new way forward: harnessing artificial intelligence to make the most of human intelligence.

Combined intelligence is a collective solution

While artificial intelligence (AI) isn’t new, its impact continues to grow, especially in healthcare. It promises a new paradigm: by learning from vast amounts of data – millions of data points across diverse training sets – it can better tailor treatment for each person. Learning from everyone allows us to care for anyone.

In the field of radiology and breast screening in particular, artificial intelligence offers a lifeline. Recently, a new study added to a growing body of literature that attests to a simple fact: radiologists working together with artificial intelligence are more efficient and more effective than the traditional double reader model.

Across a study population of more than 42,000 women who attended the Dutch population-based breast cancer screening, researchers explored the use of ScreenPoint Medical’s Transpara Detection, our flagship Breast AI, in evaluating screening mammograms program-wide.

Transpara is your key to a concurrent reading model

The findings were unequivocal: using Transpara to support a single radiologist in detecting “breast cancer in population-based mammography screening is comparable with double human reading.” While Transpara may miss “some breast cancers that are recalled by human-assessment, [it] detects a similar number of breast cancers otherwise missed by the interpreting radiologists.” As the second pair of eyes in a screening workflow, Transpara can unlock your radiology resources while improving the overall impact of your programme: your breast imaging teams will be able to get more done, in less time, and with a greater chance to catch cancer early.

Importantly, we advocate for a risk-based concurrent reading workflow: Transpara can be used as a concurrent second reader for Low Risk cases – substituting a second reader – and as a third reader alongside two radiologists for higher risk cases. This allows for a reduction in consensus meetings, which bog down workflow, and the ability to allocate resources effectively and consistently.

This framework has been proven across other large, substantive studies, including the landmark MASAI randomized controlled trial in Sweden which found a 44% reduction in workload and a 29% increase in cancer detection when adopting Transpara Detection across a study of over 100,000 women. Additional large-scale studies from Denmark, Norway, Turkey, and Spain attest to the global impact of Transpara.

Progress, in practice

Furthermore, these studies attest to change that is happening in current clinical workflows: Transpara Detection has been deployed as a second reader in Denmark since 2021 in a real live setting, making it the first live implementation of such a technology across a national screening programme. At a national scope, over 60% of breast cancer screening in Denmark and roughly 25% in Sweden currently uses Transpara Detection as a second reader. Not only is this clinically advantageous, but it is cost effective: an independent health economic study from Sweden published in July 2025 shows that implementing Transpara is cost-positive and results in nearly 11 additional quality-adjusted life years per 1,000 women screened.

While other solutions in the market may require a qualifier - “more research is needed” - for Transpara, the results are clear: consistently and cumulatively, Transpara is proven as a new standard of care for your programme’s workflow and your population’s well-being. Transpara is the Breast AI trusted by leading healthcare organizations and programmes across the globe.

10 million mammograms in, Transpara is just getting started. What are you waiting for?

Click play below to learn more:

 

Full publication available from The Lancet Digital Health.

For more on the study:
●  From AuntMinnie
●  From The Imaging Wire
●  From Health Imaging
●  From RTL
●  From IO+
●  From Radboud University Medical Center
●  From Bioengineer.org