Breast disease
Conditions
Interventions
Index test:Breast artificial intelligence model diagnoses benign and malignant breast tumors
Sponsors
The Seventh Affiliated Hospital, Sun Yat-sen University
Eligibility
Sex/Gender
Female
Age
18 Years to 90 Years
Inclusion criteria
Inclusion criteria: 1) With a clear pathology: histological biopsy or surgical resection; 2) Provide at least a bimodal ultrasound image view (B-mode/color Doppler) and complete medical history text; 3) For patients with multiple lesions, select the highest nodule of BI-RADS class. Multiple nodules in the same category, choose the largest, ensure that there is only one nodule per breast, and a maximum of two nodules per patient; 4) The lesions were classified as BI-RADS 2~5, and the positive results were selected; 5) Select solid or mostly solid lesions (cystic component<25%).
Exclusion criteria
Exclusion criteria: 1) mental illness or major underlying medical condition (e.g. tumor); 2) Implants, pregnant or lactating women; 3) Those who have undergone surgery or chemotherapy; 4) Poor image quality; 5) Patients under 18 years of age.
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Area Under Curve;F1 Score; | — |
Secondary
| Measure | Time frame |
|---|---|
| Accuracy;Positive predictive value;Negative predictive value;Sensitivity; | — |
Countries
China
Contacts
Public ContactFengping Liang
The Seventh Affiliated Hospital, Sun Yat-sen University
Outcome results
None listed