Back to the list
Congress: ECR25
Poster Number: C-20662
Type: Poster: EPOS Radiologist (scientific)
Authorblock: A. Sorce, F. Pesapane, O. Battaglia, L. Nicosia, S. Carriero, S. Santicchia, G. Carrafiello, E. Cassano; Milan/IT
Disclosures:
Adriana Sorce: Nothing to disclose
Filippo Pesapane: Nothing to disclose
Ottavia Battaglia: Nothing to disclose
Luca Nicosia: Nothing to disclose
Serena Carriero: Nothing to disclose
Sonia Santicchia: Nothing to disclose
Gianpaolo Carrafiello: Nothing to disclose
Enrico Cassano: Nothing to disclose
Keywords: Breast, Mammography, Chemotherapy, Computer Applications-Detection, diagnosis, Observer performance, Cancer
Conclusion

This retrospective study demonstrated a significant reduction in breast tissue density following neoadjuvant chemotherapy and a strong correlation between changes in breast density and the pathological response to neoadjuvant chemotherapy. The analysis revealed moderate inter-rater agreement among radiologists and substantial agreement between the most experienced radiologist and AI tools for density evaluation. These findings suggest that changes in breast density could serve as a non-invasive imaging biomarker for predicting pathological response to NAT, and that AI software could be a valuable tool in assessing breast tissue density with high consistency and reliability. The results also underscore the potential of using AI to enhance the accuracy and reproducibility of breast density evaluations, reducing variability and supporting clinical decision-making. However, further studies are necessary to confirm these findings and to establish the optimal methods for evaluating changes in breast density.

GALLERY