(30 days)
Volpara is a software application intended for use with digital mammography systems. Volpara calculates volumetric breast density as a ratio of fibroglandular tissue and total breast volume estimates. Volpara provides these numerical values for each image to aid radiologists in the assessment of breast tissue composition. Volpara produces adjunctive information. It is not an interpretive or diagnostic aid. Volpara is a software application which runs on Windows or Linux based computers.
Volpara™ analyzes raw ("for processing") digital mammograms in a fully automated, volumetric fashion and produces a quantitative assessment of breast composition, namely volume of fibroglandular tissue in cubic centimeters (cm³) volume of breast tissue in cm³ and their ratio, volumetric breast density. Volpara v1.3 handles DICOM files as input. Volpara v1.3has been built and tested on Windows XP and Linux. Volpara software is a component which accepts as input digital mammography images along with associated calibration data. The software processes the image according to proprietary algorithms. It provides measures of: volume of fibroglandular tissue, volume of breast, breast density. The software does not perform image display but outputs to the console.
Here's an analysis of the acceptance criteria and study information for the Volpara Imaging Software, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
The provided text describes several verification and validation tests, implying that acceptance criteria were met for each, but it does not explicitly list numerical acceptance criteria. Instead, it states that "All verification and validation testing was successful in that established acceptance criteria was met for all of the tests conducted."
| Acceptance Criteria (Implicit) | Reported Device Performance (Implied) |
|---|---|
| Verification Bench Testing: | |
| 1. Volpara measurements compared to known values of standardized and calibrated breast phantoms. | Test successful; Volpara measurements met established acceptance criteria when compared to known values from phantoms. |
| 2. Volpara results compared with BI-RADS scores from MQSA qualified radiologists for X-ray images. | Test successful; Volpara results showed agreement with BI-RADS scores provided by MQSA qualified radiologists, meeting acceptance criteria for this comparison. |
| 3. Volpara estimates of fibroglandular tissue compared with 3D breast MRI data for X-ray images. | Test successful; Volpara's fibroglandular tissue estimates showed acceptable correlation or agreement with 3D breast MRI data. |
| 4. Volpara breast density results compared with expected decrease in breast density with age in substantial datasets. | Test successful; Volpara's results aligned with the known physiological decrease in breast density with age in large datasets. |
| 5. Volpara results for left and right breasts and CC and MLO views compared to confirm similarity. | Test successful; Volpara consistently produced similar results across different views (CC, MLO) and between left and right breasts, indicating robustness and consistency. |
| 6. Volpara results compared for the same woman imaged on GE and Hologic systems one year apart, to confirm similarity. | Test successful; Volpara provided similar results (within acceptance criteria) for the same individual when imaged on different mammography systems (GE and Hologic) over a one-year period, demonstrating inter-system consistency over time. |
| Clinical Validation Testing: | |
| 1. Beta site testing to assess the ability of physicians to successfully integrate the software into existing systems and assess usability for target users. | Test successful; Physicians were able to successfully integrate and use the software in existing systems, indicating good usability and integration capabilities. |
| 2. Beta site testing to collect minimum, average, and maximum Volpara breast densities and compare these to other existing databases. | Test successful; Volpara's breast density measurements (min, avg, max) were within acceptable ranges or demonstrated comparison with existing databases, meeting established acceptance criteria. |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size: The text mentions "substantial datasets" for several tests (e.g., comparison with age, left/right and CC/MLO views, GE/Hologic system comparison). However, it does not provide specific numerical sample sizes for any of the test sets.
- Data Provenance: The images used for Verification and Validation testing were acquired from detectors manufactured by both GE and Hologic. The country of origin of the data is not specified. The studies appear to be retrospective as they involve existing images and data (e.g., images with existing BI-RADS scores, 3D breast MRI data, and data where women were imaged one year apart).
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
- For the comparison with BI-RADS scores, ground truth was established by "a MQSA qualified radiologist." It does not specify how many radiologists were involved, only stating "a radiologist" in the singular.
- No specific qualifications beyond "MQSA qualified radiologist" are provided (e.g., years of experience).
4. Adjudication Method for the Test Set
- For the comparison with BI-RADS scores, it just states "a MQSA qualified radiologist" provided the score, implying no adjudication for this specific ground truth.
- For other tests (e.g., phantoms, MRI, age correlation, left/right breast comparison), the "ground truth" seems to be objective measurements (phantoms, 3D MRI) or established medical knowledge (density change with age) rather than expert consensus requiring adjudication.
- The text does not mention any expert adjudication methods (e.g., 2+1, 3+1) for establishing ground truth on any of the test sets.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance
- No, a multi-reader multi-case (MRMC) comparative effectiveness study was not explicitly mentioned as performed in the provided text.
- The device is stated to "aid radiologists in the assessment of breast tissue composition" and "produces adjunctive information," but the studies described focus on the device's accuracy and consistency in calculating volumetric breast density rather than its impact on human reader performance, either with or without AI assistance. Therefore, no effect size for human reader improvement is provided.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
- Yes, a standalone performance assessment was conducted through the "Verification Bench testing." These tests evaluated Volpara's outputs against objective measures (phantoms, 3D MRI), established clinical knowledge (density and age), and consistency checks (left/right, CC/MLO views, different systems over time). The software's output is numerical values, and these tests directly assess the algorithm's accuracy in calculating these values without a human-in-the-loop for interpretation of the device's output.
7. The Type of Ground Truth Used
The ground truth types varied depending on the specific test:
- Known values from standardized and calibrated breast phantoms.
- BI-RADS scores from an MQSA qualified radiologist.
- 3D breast MRI data (for fibroglandular tissue estimates).
- Expected and known decrease in breast density with age (established medical knowledge).
- Consistency across different views and breasts (internal consistency checks).
8. The Sample Size for the Training Set
- The provided text does not mention any specific sample size for a training set. The descriptions focus on the testing of the software.
9. How the Ground Truth for the Training Set Was Established
- Since no information about a training set or its sample size is provided, there is also no information on how its ground truth was established. The document describes verification and validation of the developed software.
§ 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).