Corr-A-Net: Interpretable Attention-Based Correlated Feature Learning framework for predicting of HER2 Score in Breast Cancer from H&E Images DOI Creative Commons
Kaushik Dutta, Debojyoti Pal, Suya Li

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: April 25, 2025

Human epidermal growth factor receptor 2 (HER2) expression is a critical biomarker for assessing breast cancer (BC) severity and guiding targeted anti-HER2 therapies. The standard method measuring HER2 manual assessment of IHC slides by pathologists, which both time intensive prone to inter- intra-observer variability. To address these challenges, we developed an interpretable deep-learning pipeline with Correlational Attention Neural Network (Corr-A-Net) predict score from H&E images. Each prediction was accompanied confidence generated the surrogate estimation network trained using incentivized mechanism. shared correlated representations attention mechanism Corr-A-Net achieved best predictive accuracy 0.93 AUC-ROC 0.98. Additionally, demonstrated highest mean effective (MEC) 0.85 indicating robust level prediction. can have profound implications in facilitating status

Language: Английский

Application of deep learning on automated breast ultrasound: Current developments, challenges, and opportunities DOI Creative Commons
Ruixin Wang, Zhiyuan Wang, Yuanming Xiao

et al.

Meta-Radiology, Journal Year: 2025, Volume and Issue: unknown, P. 100138 - 100138

Published: March 1, 2025

Language: Английский

Citations

0
Corr-A-Net: Interpretable Attention-Based Correlated Feature Learning framework for predicting of HER2 Score in Breast Cancer from H&E Images DOI Creative Commons
Kaushik Dutta, Debojyoti Pal, Suya Li

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: April 25, 2025

Human epidermal growth factor receptor 2 (HER2) expression is a critical biomarker for assessing breast cancer (BC) severity and guiding targeted anti-HER2 therapies. The standard method measuring HER2 manual assessment of IHC slides by pathologists, which both time intensive prone to inter- intra-observer variability. To address these challenges, we developed an interpretable deep-learning pipeline with Correlational Attention Neural Network (Corr-A-Net) predict score from H&E images. Each prediction was accompanied confidence generated the surrogate estimation network trained using incentivized mechanism. shared correlated representations attention mechanism Corr-A-Net achieved best predictive accuracy 0.93 AUC-ROC 0.98. Additionally, demonstrated highest mean effective (MEC) 0.85 indicating robust level prediction. can have profound implications in facilitating status

Language: Английский

Citations

0