Artificial Intelligence Enhances Breast Cancer Detection in Screening Mammograms
Artificial intelligence shows promise in detecting one-third of interval breast cancers missed during screening, potentially reducing aggressive cancers diagnosed between exams.
A recent study published in the journal Radiology reveals that artificial intelligence (AI) can significantly improve the detection of interval breast cancers—those diagnosed between regular screening exams. The research indicates that AI algorithms have the potential to identify approximately one-third of these cases that are often missed during initial screening, potentially reducing the number of cancers diagnosed symptomatically later.
Interval breast cancers tend to have poorer outcomes because they are typically more aggressive and grow rapidly. Digital breast tomosynthesis (DBT), also known as 3D mammography, has improved visualization of breast tissue, making it easier to spot cancers obscured in dense tissue. However, despite its advanced imaging capabilities, some cancers remain undetected.
In the study, researchers retrospectively analyzed 224 interval cancers from a total of 1,376 cases. They employed a sophisticated AI algorithm (Lunit INSIGHT DBT v1.1.0.0) and found that it accurately localized 32.6% of these previously missed cancers. Notably, the cancers detected by AI were generally larger and more likely to involve lymph nodes, suggesting that AI might be especially effective at identifying more aggressive tumors.
The study highlights that AI's benefit extends beyond simple detection. It correctly localized cancer sites with high precision, emphasizing its role as a second reader in screening workflows. Dr. Manisha Bahl, a leading researcher, explained that this lesion-level accuracy offers a more realistic measure of AI's clinical performance compared to exam-level assessments.
This pioneering research suggests that integrating AI into screening processes could greatly enhance early detection, reduce interval cancer rates, and potentially save lives by catching cancers when they are more treatable. Nonetheless, further validation in diverse clinical settings is necessary to fully understand its impact on long-term outcomes.
Source: https://medicalxpress.com/news/2025-07-ai-interval-breast-cancers-screening.html
Stay Updated with Mia's Feed
Get the latest health & wellness insights delivered straight to your inbox.
Related Articles
Innovative Diagnostic Test to Predict Severity of FOXG1 Syndrome in Children
A new diagnostic workflow developed by researchers predicts the severity of FOXG1 syndrome in children, enabling early and personalized intervention strategies for this rare neurodevelopmental disorder.
Research Uncovers Role of Spinal Ligaments in Scoliosis Development
New research highlights the crucial role of spinal ligament development in understanding scoliosis formation, using chick embryo models to uncover structural changes linked to spinal deformities.
Extreme Heat Correlates with Higher Infant Mortality in Sub-Saharan Africa
A recent study reveals that rising temperatures significantly increase neonatal mortality in Sub-Saharan Africa, emphasizing the need for targeted interventions to protect vulnerable populations amid climate change.