BMC medical imaging · 2026
Multimodal radiomics for early prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer: integration of amide proton transfer weighted imaging in radiomics.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Xu Mingzhe, Wang Ailing, Shan Dongqiu, Chen Xuejun, Zhang Renzhi, Xu Chunmiao, Yuan Junhui, Li Jing, An Guangguang, Shen Zhiwei, Wu Yue, Chu Funing, Yang Guang, Qu Jinrong
Аннотация
BACKGROUND: Early prediction of pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer remains challenging. Traditional imaging lacks molecular sensitivity, while single-modality radiomics may miss tumor microenvironment dynamics. Integrating amide proton transfer-weighted imaging (APTWI)-quantifying protein levels linked to chemo-response-into multimodal radiomics may enable precise early pCR prediction, guiding personalized therapy. This study aims to develop and validate a multimodal radiomics model, integrating amide proton transfer-weighted imaging (APTWI), diffusion-weighted imaging (DWI), and early-phase contrast-enhanced T1WI, for preoperative prediction of pathological complete response (pCR) in breast cancer patients undergoing neoadjuvant chemotherapy (NAC).
METHODS: This retrospective study included 109 women (mean age 50 ± 10 years) with untreated breast cancer between May 2023 and August 2024, underwent NAC and pretreatment MRI scanning, including APTWI, DWI, and dynamic contrast-enhanced T1WI. Three-dimensional tumor segmentation was performed using ITK-SNAP. Radiomics features were extracted from APT, ADC maps, and enhanced subtraction images (30s/90s) using PyRadiomics, adhering to IBSI guidelines. LASSO regression selected predictive features, followed by Support Vector Machine and Logistic Regression classifiers with five-fold cross-validation. Performance was evaluated via ROC analysis, calibration curves, decision curve analysis (DCA), and SHAP interpretability.
RESULTS: Patients were randomly divided into training (n = 77) and testing (n = 32) cohorts. The integrated model combining clinical (e.g. HER2, Ki-67) and radiomics features achieved superior predictive performance (AUC = 0.888, 95% CI: 0.768-1.000) versus single-modality radiomics (ADC: AUC = 0.708), standalone clinical (AUC = 0.815), or radiomics models (AUC = 0.763). Calibration showed strong prediction-reality agreement (Brier score = 0.157). DCA confirmed clinical utility at 10%-65% thresholds. SHAP identified HER2 status and ADC-derived texture features (e.g. Cluster Shade) as top predictors.
CONCLUSION: Multimodal integration of APTWI, DWI, contrast-enhanced T1WI, and clinicopathological biomarkers enables accurate preoperative pCR prediction in breast cancer, supporting personalized NAC management.
ADVANCES IN KNOWLEDGE: Compared to single-modal imaging, our multimodal radiomics model provides more comprehensive biomarkers, thereby enabling personalized therapeutic decisions and optimizing NAC regimens to reduce overtreatment.
METHODS: This retrospective study included 109 women (mean age 50 ± 10 years) with untreated breast cancer between May 2023 and August 2024, underwent NAC and pretreatment MRI scanning, including APTWI, DWI, and dynamic contrast-enhanced T1WI. Three-dimensional tumor segmentation was performed using ITK-SNAP. Radiomics features were extracted from APT, ADC maps, and enhanced subtraction images (30s/90s) using PyRadiomics, adhering to IBSI guidelines. LASSO regression selected predictive features, followed by Support Vector Machine and Logistic Regression classifiers with five-fold cross-validation. Performance was evaluated via ROC analysis, calibration curves, decision curve analysis (DCA), and SHAP interpretability.
RESULTS: Patients were randomly divided into training (n = 77) and testing (n = 32) cohorts. The integrated model combining clinical (e.g. HER2, Ki-67) and radiomics features achieved superior predictive performance (AUC = 0.888, 95% CI: 0.768-1.000) versus single-modality radiomics (ADC: AUC = 0.708), standalone clinical (AUC = 0.815), or radiomics models (AUC = 0.763). Calibration showed strong prediction-reality agreement (Brier score = 0.157). DCA confirmed clinical utility at 10%-65% thresholds. SHAP identified HER2 status and ADC-derived texture features (e.g. Cluster Shade) as top predictors.
CONCLUSION: Multimodal integration of APTWI, DWI, contrast-enhanced T1WI, and clinicopathological biomarkers enables accurate preoperative pCR prediction in breast cancer, supporting personalized NAC management.
ADVANCES IN KNOWLEDGE: Compared to single-modal imaging, our multimodal radiomics model provides more comprehensive biomarkers, thereby enabling personalized therapeutic decisions and optimizing NAC regimens to reduce overtreatment.