Purpose To assess the inter-reader agreement in background parenchymal enhancement (BPE) categorization on ultrafast dynamic contrast-enhanced (DCE) MRI before and after training and its relationship with breast cancer prognostic factors. Materials and methods We retrospectively analyzed the cases of 220 patients (median, age 60 years; 220 breast cancers) who underwent ultrafast DCE MRI. BPE was qualitatively classified as minimal, mild, moderate, or marked by three readers: Reader1 (R1), a radiology resident in the second year of training; Reader2 (R2), an experienced general radiologist with 28 years of experience in diagnostic radiology; and Reader3 (R3), an experienced breast radiologist with 25 years of experience in breast MRI. Inter-reader agreement was evaluated using weighted kappa (κ) values and was reassessed between R1 and R3 after intensive training. Associations between BPE levels and prognostic factors were also evaluated. Results Inter-reader agreement was substantial with weighted κ values of 0.679–0.742 for R1 versus R2, 0.612–0.665 for R1 versus R3, and 0.626–0.689 for R2 versus R3. Agreement between R1 and R3 improved slightly after training but remained substantial. No significant associations were observed between visual BPE classification and prognostic factors, including hormone receptor status, HER2 status, and proliferation markers. Conclusion Inter-reader agreement for BPE assessment on ultrafast DCE MRI was substantial and improved slightly after training. No significant relationship was observed between ultrafast BPE and breast cancer prognostic factors.
No takes yet. Share an insight, caveat, or question.
山口 et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: