Can I read Explainable Breast Cancer Molecular Expression Prediction Using Multi-Task Deep-Learning Based On 3D Whole Breast Ultrasound on EtoBox?

Explainable Breast Cancer Molecular Expression Prediction Using Multi-Task Deep-Learning Based On 3D Whole Breast Ultrasound by farms3287 is a document available to read on EtoBox.

What is Explainable Breast Cancer Molecular Expression Prediction Using Multi-Task Deep-Learning Based On 3D Whole Breast Ultrasound about?

This study investigates the use of multi-task deep learning to predict breast cancer biomarkers (ER, PR, HER2) from 3D whole breast ultrasound images, utilizing a dataset of 388 patients. The multi-task model outperformed the single-task model in accuracy and interpretability, achieving higher AUC scores and better focus on diseased tissue areas as visualized by Grad-CAM++. The findings suggest that the multi-task approach enhances noninvasive biomarker prediction, potentially improving targeted drug screen

Author

farms3287

Language

EN