Enhancing Axillary Lymph Node Diagnosis in Breast Cancer with a Novel Photoacoustic Imaging-Based Radiomics Nomogram: A Comparative Study of Peritumoral Regions.

RATIONALE AND

OBJECTIVES

This study aims to assess the predictive ability of photoacoustic (PA) imaging-based radiomics combined with clinical characteristics for axillary lymph node (ALN) status in early-stage breast cancer patients and to compare performance in different peritumoral regions.

METHODS

This study involved 369 patients from Shenzhen People's Hospital, divided into a training set of 295 and a testing set of 74. PA imaging data were collected from all participants, and radiomics analysis was performed on intratumoral and various peritumoral regions. Features extracted from the training set were analyzed using LASSO regression to construct a model integrating radiomics features with clinical characteristics. Clinical factors were determined through multivariate logistic regression analysis. A radiomics nomogram was developed using logistic regression classifiers, combining radiomics features and clinical factors. The predictive efficacy of the model was evaluated using the areas under curves (AUC), and its clinical utility and accuracy were assessed through decision curve analysis and calibration curves, respectively.

RESULTS

The developed nomogram combines 5 mm peritumoral data with intratumoral and clinical features and shows excellent diagnostic performance, achieving an AUC of 0.972 in the training set and in the testing achieved 0.905. They both showed good calibrations. The model outperformed models based solely on clinical features or other radiomics methods, with the 5 mm surrounding tumor area proving most effective in identifying positive versus negative ALN in breast cancer patients.

CONCLUSION

The established nomogram is a prospective clinical prediction tool for non-invasive assessment of ALN status. It has the ability to enhance the accuracy of early-stage breast cancer treatment.

SUMMARY: This study highlights the effectiveness of combining photoacoustic radiomics with clinical parameters to predict axillary lymph node status in breast cancer, identifying a 5 mm peritumoral model as particularly potent. Future research should aim to enhance this model's robustness by expanding the sample size and advancing imaging technologies for broader clinical application.