Distinguishing Ki67 stratification expression in ER-positive/HER2-negative breast cancers: comparison of advanced MRI diffusion models

Motivation: Advanced diffusion models can assess the Ki67 expression (≤ 5% and ≥ 30%), aiding in the clinical treatment decision for ER+/HER2- breast cancers. Goal(s): This study investigated the first-order features extracted from conventional DWI, diffusion kurtosis imaging (DKI), fractional order calculus (FROC), and continuous-time random walk (CTRW) in discriminating Ki67 expression status. Approach: Retrospective study analyzed nine diffusion parameters derived from DWI, DKI, FROC and CTRW in 121 patients divided into low, medium and high Ki67 groups; utilized AUC and logistic regression for status prediction. Results: The FROC model could help identify the low Ki67expression (≤ 5%) and prevent unnecessary chemotherapy. Impact: This research reveals the utilization of FROC model improved differentiation of the low Ki67 expression group (Ki67≤ 5%) in ER+/HER2- breast cancer, who frequently exhibit a favorable prognosis and do not require excessive adjuvant chemotherapy.