Predicting Pathological Characteristics of HER2-Positive Breast Cancer from Ultrasound Images: a Deep Ensemble Approach DOI

Zhihui Chen,

Hailing Zha, Qing Yao

et al.

Deleted Journal, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 26, 2024

The objective is to evaluate the feasibility of utilizing ultrasound images in identifying critical prognostic biomarkers for HER2-positive breast cancer (HER2 + BC). This study enrolled 512 female patients diagnosed with through pathological validation at our institution from January 2016 December 2021. Five distinct deep convolutional neural networks (DCNNs) and a ensemble (DE) approach were trained classify axillary lymph node involvement (ALNM), lymphovascular invasion (LVI), histological grade (HG). efficacy models was evaluated based on accuracy, sensitivity, specificity, positive predictive value (PPV), negative (NPV), receiver operating characteristic (ROC) curves, areas under ROC curve (AUCs), heat maps. DeLong test applied compare differences AUC among different models. approach, as most effective model, demonstrated AUCs accuracy 0.869 (95% CI: 0.802–0.936) 69.7% LVI, 0.973 0.949–0.998) 73.8% HG, thus providing superior classification performance context imbalanced data (p < 0.05 by test). On ALNM, 0.780 0.688–0.873) 77.5%, which comparable other single pretreatment US-based DE model could hold promise clinical guidance predicting characteristics cancer, thereby benefit facilitating timely adjustments treatment strategies.

Language: Английский

Exploring habitats-based spatial distributions: improving predictions of lymphovascular invasion in invasive breast cancer DOI
Ge Wu,

Xiaohong Fan,

Ying Zeng

et al.

Academic Radiology, Journal Year: 2024, Volume and Issue: unknown

Published: June 1, 2024

Language: Английский

Citations

6
Predicting Pathological Characteristics of HER2-Positive Breast Cancer from Ultrasound Images: a Deep Ensemble Approach DOI

Zhihui Chen,

Hailing Zha, Qing Yao

et al.

Deleted Journal, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 26, 2024

The objective is to evaluate the feasibility of utilizing ultrasound images in identifying critical prognostic biomarkers for HER2-positive breast cancer (HER2 + BC). This study enrolled 512 female patients diagnosed with through pathological validation at our institution from January 2016 December 2021. Five distinct deep convolutional neural networks (DCNNs) and a ensemble (DE) approach were trained classify axillary lymph node involvement (ALNM), lymphovascular invasion (LVI), histological grade (HG). efficacy models was evaluated based on accuracy, sensitivity, specificity, positive predictive value (PPV), negative (NPV), receiver operating characteristic (ROC) curves, areas under ROC curve (AUCs), heat maps. DeLong test applied compare differences AUC among different models. approach, as most effective model, demonstrated AUCs accuracy 0.869 (95% CI: 0.802–0.936) 69.7% LVI, 0.973 0.949–0.998) 73.8% HG, thus providing superior classification performance context imbalanced data (p < 0.05 by test). On ALNM, 0.780 0.688–0.873) 77.5%, which comparable other single pretreatment US-based DE model could hold promise clinical guidance predicting characteristics cancer, thereby benefit facilitating timely adjustments treatment strategies.

Language: Английский

Citations

1