Deep Learning Model Enhances Identification of Calcifications on Mammograms

A new deep-learning model accurately detects, segments, and quantifies breast arterial calcifications on mammography, aiding both cancer and cardiovascular risk screening.

Key Details

• The DL model uses a modified U-Net architecture tailored for BAC segmentation and quantification.
• In testing with mammograms from 369 women, the model achieved Dice scores of 0.90 (training) and 0.89 (validation) for segmentation.
• Classification accuracy for BAC detection was high, with F1 scores of 0.97 (validation) and 0.93 (training).
• Bland-Altman analysis validated the model's reliability for calcium quantification, with a mean difference of -0.98 mg in the training set.
• Manual annotation by seven radiologists enabled categorization into six BAC mass levels.
• The approach could transform mammography into a dual-purpose screening tool for breast cancer and cardiovascular disease risk.

Why It Matters

Accurate, automated BAC quantification may enable earlier, simultaneous screening for cardiovascular risk during routine mammography, streamlining workflows and providing broader patient benefit. Adoption of such AI tools could advance integrated care in women's imaging.

Read more