Evaluation of TaiHao Breast Ultrasound Diagnosis Software
Study Details
Study Description
Brief Summary
The BR-USCAD DS Module is a computer-assisted detection and diagnosis software based on a deep learning algorithm. This retrospective, fully-crossed, multi-reader, multi-case (MRMC) study aims to compare the performances of readers without and with the aid of the Breast Ultrasound Image Reviewed with Assistance of Computer-Assisted Detection and Diagnosis System (BR-USCAD DS) in interpreting breast ultrasound images of lesions.
Condition or Disease | Intervention/Treatment | Phase |
---|---|---|
|
N/A |
Study Design
Arms and Interventions
Arm | Intervention/Treatment |
---|---|
Active Comparator: First session: manual review first and then review with CADx Each rater will interpret the "Dataset A" cases in different random order without any assistance of artificial assistance first, and then interpret the "Dataset B" cases in different random order with BR-USCAD DS. |
Diagnostic Test: Reader Group X
Each rater in "Reader Group X" will interpret the "dataset A" cases in different random order without BR-USCAD DS and interpret the "dataset B" cases in different random order with BR-USCAD DS.
|
Active Comparator: First session: review with CADx first and then manual review Each rater will interpret the "Dataset A" cases in different random order with BR-USCAD DS first, and then interpret the "Dataset B" cases in different random order without any assistance of artificial assistance. |
Diagnostic Test: Reader Group Y
Each rater in "Reader Group Y" will interpret the "dataset A" cases in different random order with BR-USCAD DS and interpret the "dataset B" cases in different random order without BR-USCAD DS.
|
Active Comparator: Second session: manual review first and then review with CADx At least 4 weeks after first session for memory washing out. Each rater will interpret the "Dataset A" cases in different random order without any assistance of artificial assistance first, and then interpret the "Dataset B" cases in different random order with BR-USCAD DS. |
Diagnostic Test: Reader Group Y
Each rater in "Reader Group Y" will interpret the "dataset A" cases in different random order with BR-USCAD DS and interpret the "dataset B" cases in different random order without BR-USCAD DS.
|
Active Comparator: Second session: review with CADx first and then manual review At least 4 weeks after first session for memory washing out. Each rater will interpret the "Dataset A" cases in different random order with BR-USCAD DS first, and then interpret the "Dataset B" cases in different random order without any assistance of artificial assistance. |
Diagnostic Test: Reader Group X
Each rater in "Reader Group X" will interpret the "dataset A" cases in different random order without BR-USCAD DS and interpret the "dataset B" cases in different random order with BR-USCAD DS.
|
Outcome Measures
Primary Outcome Measures
- MRMC AUC analysis [15 Day]
Area under the ROC curve. Value range: [0, 1]
Secondary Outcome Measures
- Sensitivity [2 Days]
Sensitivity = TP/P, value range: [0, 1] Comparing to the reference standard, the sensitivity of breast imaging results by readers with BR-USCAD DS will be evaluated.
- Specificity [2 Days]
Specificity = TN/N, value range: [0, 1] Comparing to the reference standard, the specificity of breast imaging results by readers with BR-USCAD DS will be evaluated.
- Inter-rater reliability [2 Days]
Calculate Kappa result to evaluate the agreement between raters and system. The Kappa value could be interpreted as follows: values ≤ 0 as indicating no agreement and 0.01-0.20 as none to slight, 0.21-0.40 as fair, 0.41- 0.60 as moderate, 0.61-0.80 as substantial, and 0.81-1.00 as almost perfect agreement.
- Reading time [2 Days]
Reading time in seconds. Measure reading time of breast imaging reporting and system lexicon value in breast imaging by readers without BR-USCAD DS for breast and also measured report time by readers with BR-USCAD DS for breast.
Eligibility Criteria
Criteria
Inclusion Criteria:
-
B-mode breast ultrasound image
-
Female, age 21 or older
-
Breast lesion images acquired before a biopsy or surgery - these images were retrospectively collected with histology report.
-
Non-biopsied benign lesions with negative follow-up for a minimum of 12 months
-
At least two orthogonal views of a lesion
-
All the cases in this study are collected from at least 3 different ultrasound apparatuses
Exclusion Criteria:
-
Breast lesion images acquired after biopsy or surgery.
-
Any breast surgeries or interventional procedures in the 12 months prior to ultrasound imaging
-
Case demonstrating administrative or technical errors
-
Multiple lesions in one 2-D ultrasound image
-
Breast ultrasound images with Doppler, elastography, or other overlays present
Contacts and Locations
Locations
Site | City | State | Country | Postal Code | |
---|---|---|---|---|---|
1 | Arlington Innovation Center: Health Research - Virginia Tech | Arlington | Virginia | United States | 22203 |
Sponsors and Collaborators
- TaiHao Medical Inc.
- Columbia University
- Taipei Veterans General Hospital, Taiwan
- Virginia Polytechnic Institute and State University
Investigators
- Study Director: Shih Chung Lo, Ph.D., Arlington Innovation Center: Health Research - Virginia Tech
Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- TaiHao BR-USCAD VT 20-249