(59 days)
The Crystaleye Spectrophotometer/Crystaleye Basic Set CE100-DC/US is intended for use in automated color shade-matching to assist in dental restoration. The targeted teeth for color analysis are the left and right central incisors, lateral incisors and canines.
The Crystaleye Device is an automated color shade-matching device, based upon a database of reference tooth shades. The Crystaleye Device is comprised of the following components: Spectrophotometer, AC Adapter, Cradle, Contact Cap (5 pcs.), AC Cable, USB Cable, Master CDs, and Instructions. A personal computer is also required to run the Master CDs. The reference color tooth shades of multiple dental restoration manufacturers are loaded into the computer. The spectrophotometer captures an image of the target tooth, and the software matches a patient's color tooth shade with the closest available shade in the computerized reference database.
Here's an analysis of the acceptance criteria and the study proving the device meets them, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance:
| Acceptance Criteria (Implicit) | Reported Device Performance |
|---|---|
| Correct identification of manufacturer shade from its database | The Crystaleye Device correctly identified the color shade for each of the 225 test evaluations, achieving 100% accuracy in matching to the pre-loaded reference shades. |
| Consistency/Repeatability of shade identification | The test was performed three times for each set of manufacturer shades, and in all instances, the device "correctly identified the same manufacturer shade." This indicates high consistency/repeatability. |
| Automation of shade matching | The device successfully automates the shade matching function, enabling medical and dental personnel to "visually evaluate quickly without manually attempting to match multiple color shades" and "substantially improves the ease and accuracy of color shade-matching." |
2. Sample Size Used for the Test Set and Data Provenance:
- Sample Size: 5 shades per set * 3 measurement points per shade * 5 sets of manufacturer shades = 75 unique measurement points/combinations. Each of these was tested 3 times, leading to a total of 225 test evaluations.
- Data Provenance: The study was a non-clinical bench test performed on the device itself. No information is provided regarding the country of origin of the reference shades or if they are representative of any specific geographic region. The "data" being evaluated is the device's ability to match to pre-loaded digital representations of these shades.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:
- The text does not explicitly state that human experts were used to establish the ground truth for the test set.
- The "ground truth" in this context appears to be the pre-defined, known shade from the reference shade guides that the device was designed to identify. The device's task was to match its spectrophotometer reading to one of these pre-loaded manufacturer shades. The test verified if the device's output matched the expected manufacturer shade, which themselves are standards.
4. Adjudication Method for the Test Set:
- None directly stated or implied. Given that the test involved the device identifying a specific, known manufacturer shade from its database, it's a binary outcome (correct/incorrect match to a pre-defined standard). There was no panel or group adjudicating discrepancies as the correct answer was already established by the shades themselves.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:
- No, an MRMC study was NOT done. The document explicitly states: "The Premarket Notification does not rely upon any clinical performance data." The study described is a non-clinical bench test of the device's accuracy in matching pre-defined shades.
- Therefore, there is no effect size reported for human readers improving with AI vs. without AI assistance. The device is intended to assist humans, but its direct impact on human reader performance was not evaluated in this submission.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done:
- Yes, a standalone test was done. The described "Bench testing" directly evaluates the "Crystaleye Device" without human intervention during the measurement and matching process. The device captures an image and then the software "matches a patient's color tooth shade with the closest available shade in the computerized reference database." The test then verifies if the device correctly identified the shade. While a human would visually evaluate the output of the device in clinical use, the test itself assesses the algorithm's performance in matching.
7. The Type of Ground Truth Used:
- The ground truth used was pre-defined manufacturer reference shades. These shades come from established commercial shade guides (e.g., VITAPAN CLASSICAL Shade Guide, VITA SYSTEM 3D-MASTER Tooth Guide). The device's database contains digital representations of these known shades, and the test verified the device's ability to correctly identify these known shades when presented with them.
8. The Sample Size for the Training Set:
- The document does not explicitly state the sample size for the training set. It mentions that "The Crystaleye Device utilizes the tooth shades from these shade guides in its shade-matching database," implying these shades form the database the device uses for matching, but it doesn't differentiate between training data for model development and the reference data used for operational matching.
9. How the Ground Truth for the Training Set Was Established:
- The "ground truth" for the device's internal database (which it matches against) was established from existing, legally marketed shade guides and shade teeth sets for manual color evaluation and matching. These are commercial standards from dental restoration manufacturers (e.g., VITAPAN CLASSICAL Shade Guide, VITA SYSTEM 3D-MASTER Tooth Guide, etc.). The device incorporates the "reference color tooth shades of multiple dental restoration manufacturers" into its computer. This suggests that the ground truth for its internal database is derived from these established physical shade guides.
§ 872.3661 Optical Impression Systems for CAD/CAM.
(a)
Identification. An optical impression system for computer assisted design and manufacturing (CAD/CAM) is a device used to record the topographical characteristics of teeth, dental impressions, or stone models by analog or digital methods for use in the computer-assisted design and manufacturing of dental restorative prosthetic devices. Such systems may consist of a camera, scanner, or equivalent type of sensor and a computer with software.(b)
Classification. Class II (Special Controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of the chapter subject to the limitations in § 872.9. The special control for these devices is the FDA guidance document entitled “Class II Special Controls Guidance Document: Optical Impression Systems for Computer Assisted Design and Manufacturing (CAD/CAM) of Dental Restorations; Guidance for Industry and FDA.” For the availability of this guidance document, see § 872.1(e).