Clinical

How can MammoScreen® 2D&3D help radiologists in their daily practice? Therapixel conducted an in-depth study* to understand and quantify the clinical benefits for both the radiologist and the practice. This study was conducted using version 1.0 of the MammoScreen® software (2D only), which has since been upgraded to 2.0 to include 3D use. This new version includes improved 2D&3D algorithms which further increase MammoScreen®’s overall performance.


Increase reader confidence

Rendering a diagnosis on your own can be very stressful. The use of MammoScreen® as a concurrent reader during the clinical evaluation increases reader confidence by as much as 53% by categorizing the level of suspicion for both low and high suspicion cases. The study results confirmed that when radiologists evaluated cases that had a high level of suspicion, MammoScreen®’s assessments helped them increase their level of suspicion 41% of the time. Conversely, when evaluating cases with a low level of suspicion, the radiologists’ level of suspicion was lowered 53% of the time.

Improve detection

MammoScreen® helps radiologists in interpreting difficult cases and detect more cancers. The tomosynthesis study** concluded that using MammoScreen® 3D reduced the false-negative rate by up to 47%, without affecting the radiologists’ specificity.

Reduce inter-reader variability

In addition to improving the individual's performance, the tomosynthesis study** showed that the use of MammoScreen® 3D decreases the inter-reader variability. It revealed that the intraclass correlation coefficient, which assessed the agreement between radiologists, significantly increased.

Improve efficiency

MammoScreen® 3D decreases reading time; the tomosynthesis study** demonstrated a reduction of up to 35% overall.
The most important decrease in reading time was for cases categorized by MammoScreen® as low suspicion (scores from 1 to 4), which can represent up to 75% of your screening population, for which radiologists felt more confident that the cases were benign.

* S. Pacilè, J. Lopez, P. Chone, T. Bertinotti, J. M. Grouin, and P. Fillard, “Improving Breast Cancer Detection Accuracy of Mammography with the Concurrent Use of an Artificial Intelligence Tool,” Radiology: Artificial Intelligence, vol. 2, no. 6, p. e190208, Nov. 2020, doi: 10.1148/ryai.2020190208.

** Data on file, pending publication


Discover our Scientific Publications

MammoScreen® has been used and mentioned in various scientific and research projects. Discover the full list of scientific disseminations featuring MammoScreen®.

Find out more

Discover our AI in practice

Review cases processed by MammoScreen® and see the clinical benefits.

Find out more

Let's talk!

Questions? Comments? Please feel free to get in touch. We’re here for you