Skip to main content
Fig. 2 | Breast Cancer Research

Fig. 2

From: An integrated deep learning model for the prediction of pathological complete response to neoadjuvant chemotherapy with serial ultrasonography in breast cancer patients: a multicentre, retrospective study

Fig. 2

Evaluation of the performance of SUAS. A The ROC curves for each constituent part (clinical factors, Phase 0, Phase 1, and Phase 2) of SUAS which indicated the prediction performance for pathological response (pCR vs. NpCR) in the training cohort. B The AUC values of SUAS for pathological response prediction at each phase were illustrated with blue lines, while the incremental AUCs of SUAS when the signature of a new phase was accumulated onto signatures of previous phases were illustrated with pink lines. A significant increase in pCR prediction performance was observed in the latter (p values < 0.05). C The probability and 95% CI in predicting the pathological response for each constituent part of SUAS (the orange line for the pCR group and the steel-blue line for the NpCR group). Abbreviations: SUAS: serial ultrasonography assessment system; ROC curve: receiver operator characteristic curve; pCR: pathological complete response; NpCR: non-pathological complete response; AUC: area under the curve.

Back to article page