Related Experiment Video
Updated: Nov 20, 2025

13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
43.3K
Can artificial intelligence reduce the interval cancer rate in mammography screening?
Kristina Lång1,2, Solveig Hofvind3,4, Alejandro Rodríguez-Ruiz5
1Diagnostic Radiology, Department of Translational Medicine, Lund University, Inga Maria Nilssons gata 47, SE-20502, Malmö, Sweden. kristina.lang@med.lu.se.
European Radiology
|January 24, 2021
Summary
Artificial intelligence (AI) shows promise in reducing interval cancers detected after mammography screening. This AI system could potentially identify nearly 20% of interval cancers missed in prior screenings.
Area of Science:
- Radiology
- Artificial Intelligence
- Oncology
Background:
- Interval cancers represent a significant challenge in mammography screening programs.
- Early detection of these cancers is crucial for improving patient outcomes.
Purpose of the Study:
- To evaluate the potential of a deep learning-based artificial intelligence (AI) system to reduce interval cancer rates in mammography screening.
- To assess if AI can identify cancers missed in preceding screening examinations.
Main Methods:
- A deep learning AI system was used to analyze prior screening mammograms of women diagnosed with interval cancer.
- The AI system assigned a risk score, and radiologists reviewed cases for AI accuracy in localization.
- Potential reduction in interval cancer was calculated at various AI risk score thresholds.
Main Results:
- A statistically significant correlation was found between AI risk scores and interval cancer classification.
- AI identified 19.3% of interval cancers that showed minimal signs on prior mammograms.
- The AI system also showed potential in reducing particularly aggressive interval cancers.
Conclusions:
- Artificial intelligence (AI) in mammography screening has the potential to decrease interval cancer rates.
- AI can help identify cancers missed in previous screenings without requiring additional imaging modalities.
- The study highlights AI's role in improving the efficacy of breast cancer screening.

