Search Results

Aurora is an intra-oral sensor used by dental professionals for the purpose of acquiring x-ray images to be used for the diagnosis of diseases of the mouth and for evaluating general dental health.

Aurora is an intraoral digital x-ray system used to acquire digital x-ray images when used with an external x-ray source. The sensor is composed of a scintillation plate that converts incident x-ray light into visible light, which is coupled to a CMOS detector via a fiber optic plate (FOP) collimator. It connects to a PC via a 2-meter cable with USB 2.0 interface. A trained dental professional (e.g. Dentist, Dental Assistant, etc.) will operate the sensor by aligning the device inside the mouth using a positioning device, such as ring and bar holder, and then using an external x-ray source (i.e. dental x-ray tube head) to expose the sensor to radiation with the oral tissue of interest between. A software package will control the acquisition of the x-ray image from the sensor itself and will interpret the data to create an image on a computer screen for the dentist to use for diagnosis. Neither the software package nor the x-ray sensor controls the x-ray generating source in any fashion. The software package used by Sodium Dental is Xray Vision® from Apteryx, Inc (K983111). The Xray Vision® software is a network-based image acquisition and management software used to acquire x-ray/camera images and store them in a patient database.

This 510(k) summary report describes the Aurora intra-oral digital x-ray sensor, which is stated to be identical in hardware and uses identical software to its predicate device, the QuickRay HD (K151926). Therefore, the performance data presented primarily focuses on demonstrating this equivalence rather than establishing new performance criteria or conducting extensive effectiveness studies.

Here's an analysis based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Given the claim of an identical device to the predicate, the acceptance criteria and performance are inherently tied to those of the QuickRay HD. The submission does not explicitly list "acceptance criteria" but rather presents a comparison of specifications between the Aurora and the predicate device. The "reported device performance" is the statement that these specifications are identical.

Note: The "Differences" column in the original table states "None" or "N/A" for all technical specifications, and "Equivalent" for Indications for Use.

2. Sample Size Used for the Test Set and Data Provenance

The document does not explicitly state a specific "test set" sample size for a clinical performance evaluation in the traditional sense, as the primary argument is substantial equivalence to a predicate device.

• Clinical images: "Clinical images were provided" for review by a qualified practitioner. The specific number of images is not stated.
• Data Provenance: The general nature of "clinical images" and their examination by a practitioner in Clarkston, MI, suggests retrospective use of existing images, likely from the practitioner's local patient pool. The country of origin would be the USA based on the practioner's location.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

• Number of Experts: One (1) expert.
• Qualifications: Dr. Tiffany Danyal, D.D.S., a qualified practitioner of Clarkson Village Dental, Clarkston, MI. No further details on years of experience or specialization beyond "qualified practitioner" are provided.

4. Adjudication Method for the Test Set

The document mentions that "Clinical images were examined by Dr. Tiffany Danyal... The images were determined by Dr. Danyal to be of diagnostic quality and usefulness for evaluation of all relevant oral structures." This indicates a single-reader assessment, with no explicit adjudication method stated (e.g., 2+1, 3+1).

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done

No. The document does not describe a Multi Reader Multi Case (MRMC) comparative effectiveness study. The submission focuses on demonstrating substantial equivalence through direct comparison of hardware and software specifications with a predicate device, along with bench testing.

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

Yes, in spirit, for technical and safety aspects, but not for "diagnostic" performance as a standalone AI. The submission argues the Aurora device is identical hardware and software to the predicate. The "Performance Data" section discusses:

• Bench tests: Performed in accordance with IEC 60601-1, IEC 60601-1-2, IEC 62220-1 (which includes DQE, MTF, NNPS measurements), and IEC 60529. These are objective engineering measurements of the device's physical and imaging characteristics, which are "standalone" in nature. The graphs of DQE, MTF, and NNPS versus spatial frequency (pages 7-8) are evidence of this.
• Software Verification and Validation Testing: This also assesses the algorithm/firmware only.

However, these standalone tests are primarily to confirm the device's technical specifications and safety profile match the predicate, not to evaluate its standalone diagnostic accuracy as an AI in interpreting images. The closest to diagnostic performance is the resolution (Real ≥ 20 lp/mm) and grey levels (14 bits), which are technical specifications rather than diagnostic outcomes.

7. The Type of Ground Truth Used

• For technical specifications (e.g., resolution, DQE): The ground truth is generally derived from physical measurements and adherence to international standards like IEC 62220-1.
• For the "clinical images" review: The "ground truth" was established by the expert opinion/assessment of a single dentist, Dr. Tiffany Danyal, who determined the images were of "diagnostic quality and usefulness." This is a subjective expert assessment rather than a definitive, independently verified ground truth like pathology.

8. The Sample Size for the Training Set

Not applicable. This submission focuses on demonstrating substantial equivalence to a predicate device, which is already a cleared medical device. There is no mention of a "training set" for an AI or machine learning algorithm. The imaging software (XRayVision) is stated to be "previously cleared by the FDA". The device itself is an image acquisition sensor, not an image interpretation AI.

9. How the Ground Truth for the Training Set was Established

Not applicable. As no training set for an AI was used, no ground truth needed to be established for it.

Ask a specific question about this device