[Ideas] Artificial intelligence for 3D mammography

Artificial intelligence for 3D mammography

Towards more reliable and safer imaging

Arnaud Quillent, Télécom Paris PhD in artificial intelligence for medical imaging, Sept. 2025

Artificial intelligence no longer merely offers diagnoses or interpretations: it is becoming a decision-making tool, capable of self-assessment and identifying its own limitations.

Arnaud Quillent’s work illustrates how AI can not only improve the quality of medical images, but also strengthen clinicians’ confidence in these new technologies. At a time when the use of AI is booming, it is undoubtedly an asset in the fight against breast cancer.

3D mammography, or breast tomosynthesis, represents a major advance in breast cancer screening.
This technology suffers from technical limitations that can affect image quality.
It is possible to improve these images using deep learning.
“Confidence maps” generated from the results obtained can be used to estimate the errors made by the AI.

Interview with Isabelle Mauriac, in French with English subtitles