Breast Neoplasm, Female
Conditions
Keywords
breast neoplasm, ultrasound
Brief summary
The CadAI-B (Computer Aided Design Artificial Intelligence-Breast) system is a real-time AI diagnostic tool for breast ultrasound. It integrates with ultrasound devices to detect suspicious lesions, providing size, BI-RADS, and malignancy probability. After installation and user training, the system displays real-time breast conditions and automatically analyzes lesions when a freeze frame is set, showing results in seconds. This study will assess CadAI-B's accuracy and reliability by comparing its findings with biopsy results.
Interventions
DIAGNOSTIC_TESTCadAI-B
The ultrasound is performed following the usual process. If there are any lesions or areas of concern identified by the patient, a more detailed analysis is conducted on the affected area. When a lesion is confirmed, the examiner verifies the results through the frozen image, while also securing the results displayed on CadAI-B. The examiner uses the analysis from CadAI-B as a reference to make the final BI-RADS classification. Static images and cine clips are captured from the most suspicious areas.
Sponsors
Kyungpook National University Chilgok Hospital
BeamWorks Inc.
JeeYeon Lee
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
FEMALE
Age
20 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Women aged 20 years or older * Individuals scheduled to undergo breast ultrasound for diagnostic evaluation * Patients whose Ground Truth can be determined through biopsy, surgery, or follow-up
Exclusion criteria
* Images containing artifacts that may affect the interpretation * Presence of breast implants * Significant breast trauma or mastitis at the time of the breast ultrasound that could cause distortion in image interpretation * History of surgery or chemotherapy
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Lesion and Per-patient Diagnostic Performance | At the end of the trial, up to 6 months | AUC (Area Under the Curve), Sensitivity, Specificity |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Lesion and Per-patient Detection Performance | At the end of the trial, up to 6 months | Sensitivity, Specificity |
Other
| Measure | Time frame | Description |
|---|---|---|
| BeamWorks Usability Evaluation Scale (Score of BeamWorks System Usability Scale (SUS)) | At the end of the trial, up to 6 months | 12\ 30 : Unacceptable, 30\ 40 :Marginal, >40:Acceptable |
Countries
South Korea
Outcome results
None listed