A cutting-edge artificial intelligence (AI) solution assists radiologists in the interpretation of screening mammograms and allows them to diagnose breast cancer earlier and more accurately.
The MammoScreen software from Therapixel (Paris, France) helps in reading digital mammograms to detect and qualify suspicious tissue whether they are likely to be cancerous. It uses the proprietary MammoScreen Score to provide an immediate level-of-suspicion scoring of screenings; confirming certain findings, and pointing out unsure cases that demand further scrutiny. By providing AI-guided decision-making support for breast cancer screening, MammoScreen allows radiologists to validate their suspicions, identify what they might have overlooked, and focus on uncertain studies requiring more attention.
Its non-intrusive design does not interrupt the current workflow, providing instead the results on a dedicated monitor, or side-by-side with the radiologist’s reporting software. The interface presents all the required information and a synchronization with the worklist, at once, with no additional clicks.
Using deep learning technologies, the algorithm was trained to recognize cancer lesions on hundreds of thousands of cases that were previously annotated by breast radiologist experts and confirmed when needed by a biopsy. A clinical study has shown that radiologists assisted by MammoScreen have a better performance than without the software. After having received FDA clearance for commercial sales in the US in March 2020, the AI solution is now available in the European market.
"We are proud to have received the CE Mark. This brings us a step closer to our goal of reducing breast cancer burden worldwide by improving breast cancer screening," said Pierre Fillard, Therapixel's Founder and Chief Scientific Officer.
"I am confident that MammoScreen will not only increase radiologists' performance but will also improve the patient experience by reassuring women faster," added Matthieu Leclerc-Chalvet, CEO of Therapixel.