(87 days)
The SPECTRALIS is a non-contact ophthalmic diagnostic imaging device. It is intended for:
- · viewing the posterior segment of the eye, including two- and three-dimensional imaging
- · cross-sectional imaging (SPECTRALIS HRA+OCT and SPECTRALIS OCT)
- · fundus imaging
· fluorescence imaging (fluorescein angiography, indocyanine green angiography; SPECTRALIS HRA+OCT, SPECTRALIS HRA)
- · autofluorescence imaging (SPECTRALIS HRA+OCT, SPECTRALIS HRA and SPECTRALIS OCT with BluePeak)
- · performing measurements of ocular anatomy and ocular lesions.
The device is indicated as an aid in the detection and management of various ocular diseases, including:
- age-related macular degeneration
- macular edema
- · diabetic retinopathy
- · retinal and choroidal vascular diseases
- glaucoma
The device is indicated for viewing geographic atrophy.
The SPECTRALIS OCT Angiography Module is indicated as an aid in the visualization of vascular structures of the retina and choroid.
The SPECTRALIS HRA+OCT and SPECTRALIS OCT include the following reference databases:
· a retinal nerve fiber layer thickness reference database, which is used to quantitatively compare the retinal nerve fiber layer in the human retina to values of Caucasian normal subjects – the classification result being valid only for Caucasian subjects
· a reference database for retinal nerve fiber layer thickness and optic nerve head neuroretinal rim parameter measurements, which is used to quantitatively compare the retinal nerve fiber layer and neuroretinal rim in the human retina to values of normal subjects of different races and ethnicities representing the population mix of the USA (Glaucoma Module Premium Edition)
The Heidelberg Engineering SPECTRALIS HRA+OCT is a non-contact ophthalmic device used to image the anterior and posterior segments of the human eye. The SPECTRALIS HRA+OCT is a combination of a confocal laser-scanning ophthalmoscope (cSLO, the HRA portion) and a spectral-domain optical coherence tomographer (SD-OCT). The confocal laser-scanning part of the device allows for acquisition of reflectance images (with blue, green or infrared light), conventional angiography images (using fluorescein or indocyanine green dye) and autofluorescence images. The different imaging modes can be used either alone or simultaneously. The SD-OCT part of the device acquires cross-sectional and volume images, together with an HRA cSLO image.
A blue laser is used for fluorescein angiography, autofluorescence imaging, and blue reflectance imaging, and two infrared lasers are used for indocyanine green angiography and infrared reflectance imaging. A green laser is used for MultiColor imaging ("composite color images"). MultiColor imaging is the simultaneous acquisition of infrared, green and blue reflectance images that can be viewed separately or as a composite color image. For SD-OCT imaging, an infrared superluminescent diode and a spectral interferometer are used to create the cross-sectional images.
The purpose of this premarket notification [510(k)] is to add OCT Angiography (OCTA) as an optional, add-on functionality to the SPECTRALIS HRA+OCT. Based on OCT imaging, OCTA is a non-invasive method to visualize the retinal and choroidal vasculature without the need for injection of intravenous dyes. The device generates OCTA images from analysis of temporal changes in the intensity of reflected light caused by moving particles, such as erythrocytes flowing through vessels. These changes in the OCT signal are detected by repeatedly capturing OCT images on the retina to allow for the creation of an image contrast between the perfused vessels and the static surrounding tissues.
OCTA volume scans allow for the visualization of en-face images generated from predefined or user-customizable slabs. OCT Angiography is intended for qualitative use only. No quantitative measurements are provided by the OCTA functionality.
Besides the addition of OCT Angiography, the SPECTRALIS device is unchanged. The unmodified device is the predicate device for all functions besides OCT Angiography.
The provided document details the 510(k) submission for the SPECTRALIS HRA+OCT and variants with OCT Angiography Module (K181594). The primary predicate device for the OCT Angiography module is the RTVue XR OCT Avanti with AngioVue™ Software (K153080).
Here's a breakdown of the acceptance criteria and the study proving the device meets them:
1. A table of acceptance criteria and the reported device performance
The acceptance criteria are not explicitly stated in numerical thresholds in the provided document. Instead, the study aims to demonstrate comparable performance to the predicate device (RTVue XR) in terms of image quality and agreement in identifying vascular pathologies. The reported device performance is presented in various tables comparing the SPECTRALIS with the RTVue XR.
Below is a summary table aggregating some of the reported performance metrics that serve as de facto acceptance criteria based on the comparison to the predicate.
| Acceptance Criteria (Implicit - Comparison to RTVue XR) | Reported Device Performance (SPECTRALIS HRA+OCT) |
|---|---|
| Image Quality (Score ≥ 1, Average or Good) | |
| Percentage of eyes with SPECTRALIS image quality score ≥ RTVue XR (Normal Subjects, Scan Pattern 1) | 75.0% (95% CI: 56.6, 88.5) |
| Percentage of eyes with SPECTRALIS image quality score ≥ RTVue XR (Normal Subjects, Scan Pattern 2) | 71.9% (95% CI: 53.3, 86.3) |
| Percentage of eyes with SPECTRALIS image quality score ≥ RTVue XR (Pathology Subjects, Scan Pattern 1) | 93.6% (95% CI: 82.5, 98.7) |
| Percentage of eyes with SPECTRALIS image quality score ≥ RTVue XR (Pathology Subjects, Scan Pattern 2) | 89.4% (95% CI: 76.9, 96.5) |
| Anatomical Structure Visualization (Score ≥ RTVue XR) | |
| Foveal Avascular Zone (FAZ) visualization (All Subjects, Scan Pattern 1) | 95.6% (95% CI: 87.6, 99.1) |
| Large Vessels (LV) visualization (All Subjects, Scan Pattern 1) | 97.5% (95% CI: 91.2, 99.7) |
| Small Vessels (SV) visualization (All Subjects, Scan Pattern 1) | 94.9% (95% CI: 87.5, 98.6) |
| End Capillaries (EC) visualization (All Subjects, Scan Pattern 1) | 89.9% (95% CI: 81.0, 95.5) |
| Agreement in Vascular Pathology Identification (SPECTRALIS vs. RTVue XR) | |
| Agreement in Microaneurysms identification (All Subjects, Scan Pattern 1) | 98.7% (95% CI: 93.1, 100.0) |
| Agreement in Retinal Vascular Tortuosity identification (All Subjects, Scan Pattern 1) | 94.9% (95% CI: 87.5, 98.6) |
| Agreement in Choroidal Neovascularization identification (All Subjects, Scan Pattern 1) | 96.2% (95% CI: 89.3, 99.2) |
| Agreement in Retinal Ischemia identification (All Subjects, Scan Pattern 1) | 91.1% (95% CI: 82.6, 96.4) |
| Agreement in Retinal Neovascularization identification (All Subjects, Scan Pattern 1) | 100.0% (95% CI: 95.4, 100.0) |
| Agreement in Primary Pathology of Interest (PPOI) identification (All Subjects, Scan Pattern 1) | 96.2% (95% CI: 89.3, 99.2) |
Note: Scan Pattern 1 refers to SPECTRALIS 10° x 10° and RTVue 3mm x 3mm; Scan Pattern 2 refers to SPECTRALIS 20° x 20° and RTVue 6mm x 6mm.
2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)
- Sample Size: A total of 95 participants were enrolled, with 37 in the "Normal" group and 58 in the "Pathology" group. Of these, 79 participants (32 "Normal" and 47 "Pathology") completed the study and were included in the primary analyses.
- Data Provenance: The study was conducted at a single clinical site in the United States. It was a prospective, case series study.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience)
The ground truth for the test set was established by three qualified, masked graders for OCTA image grading and another three qualified, masked graders for FA/ICGA grading. The document does not specify their exact qualifications (e.g., "radiologist with 10 years of experience"), only that they were "qualified."
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set
The document mentions "consensus OCTA image quality score" and "consensus overall anatomical structure score," implying an adjudication method where graders reached a consensus. For image quality scoring, one table shows "total inter-grader agreement on image quality scoring (where the same image quality score was determined by all three graders)." This suggests that when all three agreed, that was the score. For cases where agreement was not total (e.g., "same score from 2 graders"), the document does not explicitly state the adjudication rule (e.g., majority rule or a tie-breaker). For agreement in identification of vascular pathology, the percentages are also based on "consensus identification."
Therefore, it appears a consensus method (likely implicitly 3/3 or majority for 2/x scenarios) was used for grading both image quality and pathology identification, with some tables explicitly reporting instances where all three graders agreed.
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
Yes, a comparative effectiveness study was done, comparing the SPECTRALIS device (with its new OCTA module) to a legally marketed predicate device, the RTVue XR. However, this was not an MRMC study comparing human readers with AI assistance versus without AI assistance.
Instead, this study compares the performance of two different medical imaging devices (SPECTRALIS OCTA vs. RTVue XR OCTA) in visualizing vascular structures and pathologies. The AI component, if any, is embedded within the OCTA module's acquisition and processing, but the study design is not to assess the improvement of human readers with AI assistance. It evaluates the device's ability to produce images comparable to an existing device and align with a clinical reference standard (FA/ICGA).
Therefore, no effect size for human reader improvement with AI assistance is provided as that was not the objective of this comparative study.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
The study does not describe a standalone algorithm-only performance evaluation. The data presented is based on the comparison of images generated by the SPECTRALIS OCTA module (which includes its proprietary algorithm for OCTA) with images from the predicate RTVue XR OCTA, and then these images were evaluated by human experts. The focus is on the device's output, not an autonomous diagnostic algorithm.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
The primary ground truth for the test set was Fluorescein Angiography (FA) and Indocyanine Green Angiography (ICGA) images, which are considered a defined reference standard for visualizing retinal and choroidal vasculature. This was augmented by expert consensus (three qualified, masked graders) for evaluating image quality and ease of identifying vascular structures and pathologies on both OCTA devices, in comparison to the FA/ICGA reference.
8. The sample size for the training set
The document does not provide information on the sample size for the training set. The study described is a clinical performance study for device clearance, focusing on demonstrating substantial equivalence to a predicate device, not on the development or training of algorithms.
9. How the ground truth for the training set was established
Since no information on a training set is provided, how its ground truth was established is not detailed in this document. The study focuses on evaluating the performance of the already developed OCTA module in the SPECTRALIS device.
§ 886.1570 Ophthalmoscope.
(a)
Identification. An ophthalmoscope is an AC-powered or battery-powered device containing illumination and viewing optics intended to examine the media (cornea, aqueous, lens, and vitreous) and the retina of the eye.(b)
Classification. Class II (special controls). The device, when it is an AC-powered opthalmoscope, a battery-powered opthalmoscope, or a hand-held ophthalmoscope replacement battery, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 886.9.