What question did this study set out to answer?

To develop and evaluate a hybrid deep learning model that differentiates between benign and malignant breast tumors using multi-view ultrasound images.

May 1, 2026

A Hybrid CNN-Transformer Deep Learning Model for Differentiating Benign and Malignant Breast Tumors Using Multi-View Ultrasound Images.

View Full Paper

Ask AI

Bookmark

Authors

QZQi ZhangGuangzhou University of Chinese Medicine RLRuizhuo Li PTPan TangGuangzhou University of Chinese Medicine

Discussion

Member takes

Implication

Randomized trial demonstrates improved diagnostic accuracy of breast tumors using multi-view ultrasound images, indicating the potential for enhanced clinical decision-making.

Key Points

To develop and evaluate a hybrid deep learning model that differentiates between benign and malignant breast tumors using multi-view ultrasound images.
Introduced a computer-aided diagnosis system integrating CNN and Transformer modules.
Used pre-trained EfficientNetV2 for feature extraction and a Transformer encoder for multi-view information fusion.
Utilized datasets from 840 patients internally and 133 externally, followed by validation on a prospective dataset of 188 lesions.
Achieved an internal test accuracy of 0.960 and an AUC of 0.9788.
External dataset accuracy was 0.940 with an AUC of 0.9730.
Prospective validation confirmed an accuracy of 0.952 and an AUC of 0.9801, outperforming baseline models and human experts.

Cite This Study

Zhang et al. (2026) studied this question.

synapsesocial.com/papers/69f44390967e944ac5566bdc https://doi.org/10.1177/15330338261447344

View Full Paper

Ask AI

Bookmark

Implication

Randomized trial demonstrates improved diagnostic accuracy of breast tumors using multi-view ultrasound images, indicating the potential for enhanced clinical decision-making.

A Hybrid CNN-Transformer Deep Learning Model for Differentiating Benign and Malignant Breast Tumors Using Multi-View Ultrasound Images.

Discussion

Member takes

Implication

Key Points

Cite This Study

A Hybrid CNN-Transformer Deep Learning Model for Differentiating Benign and Malignant Breast Tumors Using Multi-View Ultrasound Images.

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider