Search Results

This is a digital mobile diagnostic x-ray system intended for use by a qualified/trained doctor or technician on both adult and pediatric subjects for taking diagnostic radiographic exposures of the skull, spinal column, chest, abdomen, extremities, and other body parts. Applications can be performed with the patient sitting, standing, or lying in the prone or supine position. Not for mammography.

This is a modified version of our previous predicate mobile PHOENIX. The predicate PHOENIX mobile is interfaced with Konica – Minolta Digital X-ray panels and CS-7 or Ultra software image acquisition. PHOENIX mobile systems will be marketed in the USA by KONICA MINOLTA. Models with the CS-7 Software will be marketed as AeroDR TX m01. Models with the Ultra software will be marketed as mKDR Xpress. The modification adds two new models of compatible Konica-Minolta digital panels, the AeroDR P-65 and AeroDR P-75, cleared in K210619. These newly compatible models are capable of a mode called DDR, Dynamic Digital Radiography wherein a series of radiographic exposures can be rapidly acquired, up to 15 frames per seconds maximum (300 frames).

The provided text describes a 510(k) premarket notification for a mobile x-ray system. The document focuses on demonstrating substantial equivalence to a legally marketed predicate device rather than presenting a study to prove the device meets specific performance-based acceptance criteria for an AI/algorithm.

Therefore, many of the requested details, such as specific acceptance criteria for algorithm performance, sample sizes for test sets, expert ground truth establishment, MRMC studies, or standalone algorithm performance, are not applicable or not present in this type of submission.

The essence of this submission is that the entire mobile x-ray system, including its components (generator, panels, software), is deemed safe and effective because it is substantially equivalent to a previously cleared device, with only minor modifications (adding two new compatible digital panels and enabling a DDR function in the software, which is stated to be "unchanged firmware" and "moderate level of concern").

Here's an attempt to address your questions based on the provided text, while acknowledging that many of them pertain to AI/algorithm performance studies, which are not the focus of this 510(k):

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

The document does not specify performance-based acceptance criteria for an AI/algorithm. Instead, it demonstrates substantial equivalence to a predicate device by comparing technical specifications. The "acceptance criteria" in this context are implicitly met if the new device's specifications (kW rating, kV range, mA range, collimator, power source, panel interface, image area sizes, pixel sizes, resolutions, MTF, DQE) are equivalent to or improve upon the predicate, and it remains compliant with relevant international standards.

2. Sample size used for the test set and the data provenance

No specific test set or data provenance (country, retrospective/prospective) is mentioned for AI/algorithm performance. The "testing" involved "bench and non-clinical tests" to verify proper system operation and safety, and that the modified combination of components produced diagnostic quality images "as good as our predicate generator/panel combination." This implies physical testing of the device rather than a dataset for algorithm evaluation.

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

Not applicable. There was no specific test set requiring expert-established ground truth for an AI/algorithm evaluation. The determination of "diagnostic quality images" likely involved internal assessment by qualified personnel within the manufacturer's testing process.

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

Not applicable. No adjudication method is described as there was no formal expert-read test set for algorithm performance.

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

No. An MRMC study was not conducted as this submission is not about an AI-assisted diagnostic tool designed to improve human reader performance. It is for a mobile x-ray system.

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

No. This submission is for a medical device (mobile x-ray system), not a standalone AI algorithm. The software components (CS-7 and Ultra) are part of the image acquisition process, and the only software "modification" mentioned is enabling the DDR function, which is a feature of the new panels, not an AI for diagnosis.

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

Not applicable. The substantial equivalence argument relies on comparing technical specifications and demonstrating that the physical device produces images of "diagnostic quality" equivalent to the predicate, rather than an AI producing diagnostic outputs against a specific ground truth.

8. The sample size for the training set

Not applicable. This is not an AI/ML algorithm submission requiring a training set. The software components are for image acquisition and processing, not for AI model training.

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

Not applicable, as no training set for an AI/ML algorithm was used or mentioned.

Ask a specific question about this device