(91 days)
M-Vu Breast Density is a software application intended for use with digital mammography systems. M-Vu Breast Density calculates breast density as a ratio of fibroglandular tissue and total breast area estimates. M-Vu Breast Density provides these numerical values for each breast as well as a density category to aid radiologists in the assessment of breast tissue composition. M-Vu Breast Density produces adjunctive information. It is not an interpretive or diagnostic aid.
M-Vu Breast Density automatically analyzes "for processing" digital mammograms and calculates the dense tissue area of each breast. The measured dense tissue area is then used to provide a Calibrated Density Category which maps the percentage of breast density to a BI-RADS category number (1 - 4).
M-Vu Breast Density is a stand-alone software application designed to interoperate with all digital radiography (DR) and computed radiography (CR) mammography systems. M-Vu Breast Density is displayed in the form of a DICOM mammography structured report or secondary capture and reports the following for output:
- Breast Area (cm2) for cach breast .
- Dense Area (cm2) for each breast .
- Percent Breast Density for each breast .
- . Breast Density Category for each case
The results of M-Vu Breast Density are designed to display on a mammography workstation, high resolution monitor. or in a printed case report. M-Vu Breast Density is designed to process approximately 60-120 cases per hour.
M-Vu Breast Density Version 1.0.0.0 has been built and tested on the M-Vu CAD Station system (K061160).
Here's an analysis of the acceptance criteria and study details for the M-Vu Breast Density device, based on the provided FDA 510(k) summary:
M-Vu® Breast Density - Acceptance Criteria and Study Details
1. Table of Acceptance Criteria and Reported Device Performance
The provided document does not explicitly list acceptance criteria as quantitative targets for the device's performance. Instead, it describes verification and validation activities designed to demonstrate the device's functionality and alignment with known biological phenomena. The "reported device performance" is thus interpreted as the successful completion of these validation tests.
| Acceptance Criteria (Implied from Validation) | Reported Device Performance (Successful Validation) |
|---|---|
| Agreement with Expert Radiologist BI-RADS Scores: | A weighted Kappa statistic was used to compare M-Vu Breast Density's calibrated density categories with BI-RADS scores obtained from 13 expert radiologists. The study implies satisfactory agreement, as it states the comparison was performed, and the overall conclusion is that the device is "safe and effective" and "substantially equivalent." However, no specific threshold for Kappa or actual Kappa value is provided. |
| Accuracy of Fibroglandular Tissue Estimates: | M-Vu percent breast density (PBD) and breast area measurements were verified. The verification of fibroglandular area measurement was confirmed "by implication." No specific metrics or thresholds for accuracy are provided. |
| Correlation with Age-Related Density Decrease: | The device's results were compared with the expected and known decrease in breast density with age using the Spearman Rank Correlation test. This implies a significant negative correlation was observed. No specific correlation coefficient or p-value is provided. |
| Consistency between Left and Right Breast Density: | Percent Breast Density measurements from the left and right breasts of the same patient were compared using Pearson's Correlation Coefficient. This implies a high positive correlation was observed. No specific correlation coefficient or p-value is provided. |
| Consistency of PBD over Time (Up to 2 years): | Percent Breast Density measurements were made on patient images and corresponding prior images (maximum two years apart) and compared using Pearson's Correlation Coefficient. This implies a high positive correlation was observed, indicating stability of measurements over time. No specific correlation coefficient or p-value is provided. |
| Successful Clinical Network Integration: | M-Vu CAD Station (on which M-Vu Breast Density is deployed) was successfully tested for clinical network integration. This is a functional and technical acceptance, not a performance metric for density assessment. |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size for Expert BI-RADS Comparison: Not explicitly stated. The document mentions "a set of x-ray images for which a BI-RADS scores were obtained from 13 expert radiologists," but the number of cases or images in this set is not provided.
- Sample Size for Age Correlation: Not explicitly stated. The document mentions "a data set where the women's age and results were compared," but the number of cases is not provided.
- Sample Size for Left/Right Breast Comparison: Not explicitly stated.
- Sample Size for Prior Image Comparison: Not explicitly stated.
- Data Provenance: Not specified. It is likely retrospective, given the nature of image analysis and comparison with existing data (e.g., prior images, expert BI-RADS scores). The country of origin of the data is not mentioned.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
- Number of Experts: 13 expert radiologists were used to provide BI-RADS scores.
- Qualifications of Experts: Stated as "expert radiologists." No specific details on their years of experience or subspecialty are provided.
4. Adjudication Method for the Test Set
For the primary comparison (BI-RADS scores), the document states that expert BI-RADS scores were "obtained from 13 expert radiologists." It does not specify an adjudication method (e.g., 2+1, 3+1, none) to establish a consensus ground truth from these 13 experts. It's possible each expert's score was treated individually, or a simple majority/median was used, but this detail is missing.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- Was an MRMC study done? No. The document explicitly states: "This submission contains no information from clinical studies." The studies conducted are validation studies for the algorithm's performance against established measures or trends, not studies comparing human reader performance with and without AI assistance.
- Effect size of human reader improvement: Not applicable, as no MRMC study was performed.
6. Standalone Performance Study
- Was a standalone study done? Yes. All described "non-clinical performance data" are standalone studies of the M-Vu Breast Density algorithm. The device "was run over a set of x-ray images" and its outputs were then compared to various metrics (expert BI-RADS, age trends, intra-patient consistency). This demonstrates the algorithm's performance without human intervention in the density calculation process.
7. Type of Ground Truth Used
The ground truth used for the validation studies includes:
- Expert Consensus/Opinion: BI-RADS scores obtained from 13 expert radiologists.
- Known Biological/Clinical Trends: The expected decrease in breast density with age.
- Internal Consistency: Comparison of left vs. right breast density and density over time for the same patient.
No pathology or outcomes data were used as ground truth.
8. Sample Size for the Training Set
The document does not provide any information regarding the sample size used for the training set of the M-Vu Breast Density algorithm.
9. How the Ground Truth for the Training Set Was Established
The document does not provide any information on how the ground truth for the training set was established. Given the lack of training set details, this information is entirely absent.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).