To establish a non-invasive imaging-based biomarker for predicting preoperative Ki-67 expression status in breast cancer, addressing the clinical challenge of accurate assessment.
Key Findings:
Clinics_Habitat_Radiomics model achieved AUCs of 0.877 (95% CI: 0.826–0.929) in training and 0.830 (95% CI: not provided) in validation cohorts.
Sensitivity of 60.3% and specificity of 91.7% in validation cohort.
Calibration curves showed close agreement between predicted probabilities and observed outcomes.
Interpretation:
The integration of habitat analysis with ultrasound radiomics enhances the precision of tumor biology assessment and provides actionable insights for treatment optimization, potentially improving patient outcomes.
Limitations:
Retrospective design may introduce selection bias, potentially affecting the validity of the findings.
Single-center study limits generalizability.
Conclusion:
The developed nomogram offers a promising non-invasive predictive tool for Ki-67 expression, bridging the gap between radiomic diagnostics and personalized oncology care, emphasizing the importance of non-invasive methods in clinical practice.
