Publications by authors named "Zeshuo Jiang"
Deep Learning-Based DCE-MRI Automatic Segmentation in Predicting Lesion Nature in BI-RADS Category 4.
J Imaging Inform Med
November 2024
Article Synopsis
- A study explored the effectiveness of a deep learning (DL) algorithm for automatic segmentation of DCE-MRI images in differentiating BI-RADS 4 breast lesions, comparing it to manual segmentation.
- Involving 197 patients, the DL model demonstrated high consistency with manual methods (Dice coefficient of 0.84) and outperformed professional radiologists in predictive accuracy with an area under the curve (AUC) of 0.85.
- The DL model not only provided better diagnostic results but also significantly increased efficiency, reducing the time needed for segmentation by 83.2%, indicating a promising future for clinical use in breast cancer diagnosis.