This study aims to construct a multi-task deep learning model system to mine deep features in DBT images, so as to achieve accurate detection of breast lesions, differential diagnosis of benign and malignant (especially for the challenging BI-RADS 4A category), prediction of molecular subtypes, and evaluation of neoadjuvant chemotherapy (NAC) efficacy, providing an imaging basis for precision medicine.
Female patients aged ≥ 18 years.
Complete bilateral digital breast tomosynthesis (DBT) images available, including craniocaudal (CC) and mediolateral oblique (MLO) views.
Confirmed pathological diagnosis (core needle biopsy or surgical resection) serving as the reference standard; or benign lesions with stable findings on follow-up for more than 2 years.
(For the efficacy prediction subgroup) Patients who received complete neoadjuvant therapy and had postoperative pathological results.
Poor image quality with severe artifacts that precluded reliable analysis.
History of previous breast surgery or radiotherapy (except for the recurrence risk subgroup).
Incomplete clinical or pathological data.
Single group
To explore the value of digital breast tomosynthesis based on deep learning in the diagnosis of breast cancer
Diagnostic testThe digital breast tomosynthesis is part of the standard treatment protocol.
The accuracy of the multi-task deep learning-based intelligent diagnostic model in differentiating benign and malignant breast lesions on digital breast tomosynthesis (DBT) images.
Taking surgical or puncture histopathological results as the gold standard, the accuracy of the multi-task deep learning-based intelligent diagnostic model in differentiating benign and malignant breast lesions on digital breast tomosynthesis (DBT) images was evaluated. It focuses on challenging BI-RADS 4A lesions, covering retrospective multi-center validation sets and prospective multi-center validation sets to ensure the representativeness and rigor of the indicator.
Click on the lead sponsor to view all of their trials.
This trial is not recruiting at the moment. You can still explore other options:
