Search Results

Volpara Imaging Software is a software application intended for use with the raw data from digital breast x-ray systems, including tomosynthesis. Volpara Imaging Software calculates and quantifies a density map and from that determines volumetric breast density as a ratio of fibroglandular tissue and total breast volume estimates. Volpara Imaging Software provides these numerical values along with a BI-RADS breast density 4th or 5th Edition category and quality assurance metrics (i.e., dose and pressure) to aid healthcare professionals in the assessment of breast composition. Volpara Imaging Software is not a diagnostic aid and should be used only as adjunctive information when the final assessment of breast density category is made by an MQSA-qualified interpreting physician.

Volpara Imaging Software 3.2 analyzes raw ("For Processing") digital mammograms or raw DBT projections in a fully automated, volumetric fashion. It produces a quantitative assessment of breast composition in the form of a Volpara density map, wherein the value at each pixel represents the thickness of fibroglandular tissue between that pixel and the x-ray source.

From the density maps various quantitative density-map based metrics are computed as follows:

• volume of fibroglandular tissue in cm3
• volume of breast in cm³
• the volumetric breast density (the percentage of fibroglandular tissue in breast)
• average thickness of dense tissue in cm
• maximum thickness of dense tissue in cm
• maximum volume of dense tissue above any 1 cm² square region (and location)
• image quality assessment metrics

Volpara Imaging Software typically receives four standard views from the digital x-ray system following a mammographic or digital breast tomosynthesis screening examination (i.e., left CC, right CC, left MLO, right MLO). The generated results are then displayed via either of the following:

• A Volpara scorecard (a DICOM Secondary Capture Image)
• A DICOM Mammography CAD Structured Report

The device outputs those metrics along with the density maps themselves marked with the location of the various maxima. Using the volumetric breast density, a Bl-RADS 4th Edition and 5th Edition breast density category is generated by applying thresholds set by the software.

Volpara Imaging Software operates on a Microsoft Azure-connected, off-the-shelf virtual appliance, which provides a secure, pre-configured virtual host. Volpara Imaging Software does not contact the patient, nor does it control any life-sustaining devices.

The provided text describes a Special 510(k) submission for Volpara Imaging Software 3.2, which is an updated version of a previously cleared device. While it mentions verification and validation, it does not provide specific acceptance criteria or detailed results of a study (such as a clinical performance study with defined endpoints, sample sizes, and ground truth methodologies) that proves the device meets typical acceptance criteria for medical AI/software.

The document focuses on demonstrating substantial equivalence to a predicate device (Volpara Imaging Software 1.5.6) by highlighting minor changes and stating that "complete verification and validation data testing conducted for the predicate was repeated to ensure integration and backwards compatibility." This implies that the performance shown for the predicate device is considered sufficient for the updated device, given the minor nature of the changes.

Therefore, I cannot fulfill your request for a table of acceptance criteria and reported device performance based solely on the provided text, nor can I provide detailed information about a "study that proves the device meets the acceptance criteria" in the way you've outlined for clinical performance. The submission type (Special 510(k)) and the content provided indicate a focus on demonstrating that the changes do not alter the substantial equivalence to the predicate, rather than a de novo clinical performance study for this specific version.

However, I can extract information related to the device and the nature of its submission:

Based on the provided text, a detailed clinical study demonstrating the specific acceptance criteria and performance of Volpara Imaging Software 3.2 (as defined in your prompt) was not described. The submission focuses on demonstrating substantial equivalence to a predicate device through verification and validation of minor software changes.

Here's what can be inferred or stated from the text, addressing your points where possible, and stating "Not provided" where the information is absent:

1. A table of acceptance criteria and the reported device performance

• Not provided in this document. The document discusses "verification and validation" and "compliant to the following standards: ISO 14971:2012, ISO 62304:2006, DICOM 2016." These are general standards for medical device software development and risk management, not specific acceptance criteria or performance metrics for breast density assessment.

2. Sample sized used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Not provided. The document states "complete verification and validation data testing conducted for the predicate was repeated," but does not mention specific test set sizes or data provenance.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

• Not provided. Given that the device calculates volumetric breast density and BI-RADS categories but is "not a diagnostic aid" and "should be used only as adjunctive information when the final assessment... is made by an MQSA-qualified interpreting physician," the ground truth for such a device would typically relate to the accuracy of its calculations against established physical models or potentially expert consensus on BI-RADS categories. However, no details on how "ground truth" for validation was established are given in this document.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

• Not provided.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

• Not provided. This type of study is not described. The device is a quantitative imaging software, not explicitly an AI-assisted diagnostic tool in the typical sense of MRMC studies (though it uses "automated" analysis). Its primary function is to provide numerical values and BI-RADS categories, which are then used by a physician.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

• Implied, but details not provided. The description of the device's function ("calculates and quantifies a density map and from that determines volumetric breast density... provides these numerical values along with a BI-RADS breast density... category") describes a standalone algorithmic calculation. However, no performance metrics from such a standalone evaluation are presented in this document.

7. The type of ground truth used (expert concensus, pathology, outcomes data, etc)

• Not provided. For a volumetric density calculation, ground truth could be established via phantoms, physical measurements, or potentially a highly detailed, independently verified volumetric analysis of breast tissue. No specific method is mentioned.

8. The sample size for the training set

• Not provided. This document describes a Special 510(k) for software updates to an already cleared device, implying the original device's training (if any neural networks were involved, though the description sounds more rule-based/analytical) would have occurred for the predicate. Information for this submission focuses on verification/validation of changes, not a new development cycle.

9. How the ground truth for the training set was established

• Not provided. (See point 8).

Ask a specific question about this device