Development of an photoacoustic-based radiomics nomogram to preoperatively predict Ki-67 expression level in patients with breast cancer

Abstract

Objective This study aimed to develop and validate a radiomic nomogram utilizing photoacoustic imaging to predict Ki-67 status in breast cancer patients. Methods A retrospective analysis included 223 breast cancer patients diagnosed between October 2022 and October 2023. Patients underwent multimodal photoacoustic/ultrasound imaging and Ki-67 detection. Random allocation into training (n = 178) and test sets (n = 45) followed an 8:2 ratio. Tumor regions were outlined, and radiomic features were extracted from both photoacoustic and ultrasound images. Feature screening involved independent samples t-tests and the least absolute shrinkage with selection operator (LASSO). Rad-Score was computed for each radiomic score, and logistic regression integrated Rad-Score with clinical risk factors to construct the nomogram. Comparative analysis between nomogram models of the two images was performed. Model performance was assessed using receiver operating characteristic (ROC) curves, decision curve analysis (DCA), and calibration curves. Results In both cohorts, the nomogram model outperformed clinical and radiomic models. In the test cohort, the area under the curve (AUC) for photoacoustic and ultrasound-based nomogram models were 0.87 (95% CI: 0.69–0.89) and 0.84 (95% CI: 0.67–0.86), respectively, indicating superior performance of the photoacoustic-based nomogram in predicting Ki-67 expression. DCA further demonstrated the clinical utility of the model. Conclusions The nomogram model based on photoacoustic radiomics shows promise as a potential tool for predicting Ki-67 levels in breast cancer.