Search Results

The MOZART SUPRA Specimen Tomosynthesis System is a Cabinet x-ray system that is specifically designed to provide high detail radiographic imaging of surgically excised medical specimens from various anatomical regions, i.e. breast, both in 2-dimensional and 3-dimensional tomosynthesis views.

The MOZART SUPRA(XPERT84) Specimen Radiography System is a Cabinet X-ray System specifically designed to provide 3-D high detail radiographic imaging of surgically excised medical specimens utilizing tomosynthesis. The MOZART SUPRA(XPERT84) is a fully self-contained and shielded cabinet system equipped with a 90kVp micro-focus x-ray source and a 10'' x 12" 49.5 micron high resolution CMOS Digital Detector. It is the only cabinet specimen imaging system to utilize 3-D Tomosynthesis technology. Creates images in 1mm digital slices of the specimen, allowing physicians to evaluate the specimen layer by layer.

The provided document is a 510(k) premarket notification for a medical device, the Kubtec MOZART SUPRA (XPERT 84) Radiography System. It primarily focuses on demonstrating substantial equivalence to predicate devices rather than providing detailed acceptance criteria and a study demonstrating performance against specific metrics for AI-powered device functionality.

The document describes a stationary X-ray system designed for high detail radiographic imaging of surgically excised medical specimens, including breast specimens, in both 2D and 3D tomosynthesis views. It highlights the device's capability to provide more anatomical information, precisely identify lesion locations, exclude overlying tissue, and identify surgical margins. There is no mention of an Artificial Intelligence (AI) component within this device or its functionality. The "software" mentioned (KUBTEC DIGICOM) relates to image acquisition and processing, not AI for diagnosis or analysis.

Therefore, many of the requested elements regarding acceptance criteria and study design for an AI-powered device cannot be extracted from this document. The document describes a traditional medical imaging device, not an AI/ML product.

However, I can extract information related to the device's performance based on the general information provided for its intended use as a radiography system.

Here's an attempt to address the request based on the available information, noting the absence of AI-specific details:

Acceptance Criteria and Device Performance (as a Radiography System, NOT AI)

The document primarily focuses on demonstrating substantial equivalence to predicate devices and adherence to regulatory standards (electrical safety, EMC, software V&V for function, not AI performance). There are no explicit quantitative acceptance criteria (e.g., sensitivity, specificity, AUC) for diagnostic performance laid out in a table, as one would expect for an AI algorithm. The closest to "performance" is the device's technical specifications and the claim that it "produces diagnostic quality images."

Table of Acceptance Criteria and Reported Device Performance (Proxy)

Since this is not an AI device, I will create a table based on the device's technical specifications and claimed imaging capabilities, which implicitly serve as performance indicators for a radiography system, rather than diagnostic accuracy metrics.

Study Details (as a Diagnostic Imaging Device - No AI)

1. Sample size used for the test set and data provenance:

   • The document states: "Images attained by the system have been read and verified by a Board Certified Radiologist to product diagnostic quality images." This implies some form of review or test set, but it does not specify a sample size for this "test set" (if it was a formal one) nor any details on data provenance (e.g., country of origin, retrospective/prospective). This sounds more like an internal verification than a formal clinical study with a specified sample size.

2. Number of experts used to establish the ground truth for the test set and qualifications of those experts:

   • One expert is explicitly mentioned: "a Board Certified Radiologist." No further details on experience level (e.g., 10 years of experience) are provided.

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

   • No formal adjudication method is described. The statement "read and verified by a Board Certified Radiologist" suggests a single reader's assessment.

4. 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:

   • No, an MRMC study was NOT done. This document pertains to a traditional X-ray imaging system, not an AI-assisted device. Therefore, a study comparing human readers with and without AI assistance is irrelevant and not performed.

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

   • No, this is not applicable. The device is a physical X-ray system that produces images for human interpretation. There is no AI algorithm to evaluate in a "standalone" fashion.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • The "ground truth" implicitly refers to the assessment by the "Board Certified Radiologist" that the images are of "diagnostic quality." This points to expert assessment as the primary form of validation, likely against a gold standard like pathology where applicable (specimen imaging). However, the document does not explicitly state pathology as the ground truth for image quality. It implies the system accurately portrays the pathology of the specimen without directly stating pathology reports were used as ground truth for a quantitative study.

7. The sample size for the training set:

   • Not applicable for this device. This is a hardware imaging system, not an AI/ML algorithm that requires a training set. The software mentioned (DIGICOM) is for image acquisition and processing, not model training.

8. How the ground truth for the training set was established:

   • Not applicable. As there is no AI/ML algorithm with a training set, no ground truth establishment for a training set is relevant.

Ask a specific question about this device