Search Results
Found 1 results
510(k) Data Aggregation
(109 days)
Verily Numetric Retinal Camera is a non-contact, high resolution digital imaging device which is suitable for photographing, displaying and storing images of the retina segment of the eye to be evaluated under non-mydriatic conditions.
Verily Numetric Retinal Camera is indicated for in-vivo viewing of the posterior segment of the eye. The images are intended for use as an aid to clinicians in the evaluation, diagnosis, and documentation of ocular health.
The Verily Numetric Retinal Camera (VNRC), the subject device, is a tabletop fundus camera that provides non-mydriatic, color, posterior segment images of the eye as an aid to clinicians in the evaluation and diagnosis of eye disease. It is a non-contact, high-resolution digital imaging device that is suitable for photographing, displaying, and storing images of the retina to be evaluated under non-mydriatic conditions.
The Verily Numetric Retinal Camera takes images of the fundus in the following manner:
Focusing: The patient interactively focuses an image shown on a microdisplay to the best qualitative focus by turning a knob on an external peripheral device which modulates the camera optics.
Alignment: The patient is shown a fixation target on the microdisplay and through an interactive session with position feedback, the patient aligns their pupil to the camera. The pupil location is determined using eye-tracking.
Image Capture: White light from LEDs illuminates the fundus. The light reflection from the fundus is captured by a color CMOS sensor.
The Verily Numetric Retinal Camera has the following acceptance criteria and reported device performance:
1. Table of Acceptance Criteria and Reported Device Performance
| Performance Item | Acceptance Criteria | Reported Device Performance |
|---|---|---|
| Resolving power (center) | ≥ 60 line pairs/mm at the center of the field | Pass |
| Resolving power (mid field) | ≥ 40 line pairs/mm at the mid field (r/2) | Pass |
| Resolving power (periphery) | ≥ 25 line pairs/mm at the periphery of the field (r) | Pass |
| Field of view | 45 degrees | Pass |
| Pixel pitch tolerance | < 7% | Pass |
| Alignment illumination | The alignment illumination is done via Near Infrared (NIR) LEDs during eye tracking. The intensity of the NIR LEDs is controlled by SW. | Pass |
| Flash illumination | The flash illumination is done via white LEDs during the image capture. The intensities and durations of the white LEDs are controlled by SW. | Pass |
| Range of focus | -20D to +20D (no compensating lens) | Pass |
| Minimum pupil size | 3.0mm | Pass |
| Alignment | Interactive alignment using 3D tracking. | Pass |
| Image quality | The gradability of images captured by VNRC is comparable to the predicated device on the same population. | Pass |
2. Sample Size and Data Provenance for Test Set
The document mentions that non-clinical performance testing for the Verily Numetric Retinal Camera includes analysis using raw data collected via a retrospective clinical study. It also states that the "gradability of images captured by VNRC is comparable to the predicated device on the same population."
However, the specific sample size used for the test set is not explicitly stated in the provided text. The country of origin of the data is also not specified.
3. Number and Qualifications of Experts for Ground Truth (Test Set)
The document does not provide information on the number of experts used to establish ground truth for the test set or their specific qualifications (e.g., "radiologist with 10 years of experience"). It only mentions that the images are "intended for use as an aid to clinicians in the evaluation, diagnosis, and documentation of ocular health" and that "the gradability of images captured by VNRC is comparable to the predicated device on the same population." This implies clinical evaluation but lacks details on the expert panel.
4. Adjudication Method for the Test Set
The document does not specify an adjudication method (e.g., 2+1, 3+1, none) for the test set.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
A multi-reader multi-case (MRMC) comparative effectiveness study was not explicitly described in the provided text in the typical sense of measuring human reader improvement with AI assistance. The document compares the performance of the device itself (Verily Numetric Retinal Camera) to a predicate device (NFC-700) regarding image quality and technical specifications. It notes that "the gradability of images captured by VNRC is comparable to the predicated device on the same population," implying a comparison of image utility for human readers, but no effect size on human reader improvement with AI assistance is provided as the device is a camera, not an AI diagnostic tool.
6. Standalone (Algorithm Only) Performance Study
The Verily Numetric Retinal Camera is a hardware device (ophthalmic camera) that captures images, not a standalone AI algorithm. Therefore, a standalone (algorithm only) performance study as typically understood for AI-based diagnostic software was not conducted or described. The focus is on the camera's ability to produce high-quality, gradable images.
7. Type of Ground Truth Used (for Test Set)
The document primarily refers to evaluating "image quality" and "gradability." While not explicitly stated as "expert consensus ground truth," the comparison of "gradability of images captured by VNRC... comparable to the predicated device on the same population" strongly implies that human expert evaluation of image quality was used as a form of ground truth for assessing the usability of the images for clinical purposes. There is no mention of pathology or outcomes data being used as ground truth for effectiveness.
8. Sample Size for the Training Set
The document does not provide a sample size for a training set. This is expected as the device is a camera, not an AI algorithm that typically requires a training set.
9. How Ground Truth for Training Set Was Established
As the device is a camera and not an AI algorithm, the concept of a "training set" and establishing "ground truth for the training set" does not directly apply in the context of this submission. The camera's performance is evaluated against technical specifications and a comparison to a predicate device.
Ask a specific question about this device
Page 1 of 1