FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The ANTERION is a diagnostic imaging device for the eye. The technology is based on swept-source optical coherence tomography (SS-OCT) technology. The device itself has two basic component groups:

. ANTERION Hardware (Imager/Basis) with integrated forehead/ chin rest: The hardware includes imaging hardware (e.g., laser, LEDs, optics, detectors, hardware for spatial encoding) as well as a touch screen.
ANTERION Software (PC): The ANTERION Software includes the main user . interface. The software allows for device control, such as selection of examination(s) and imaging parameter(s). The PC software additionally stores, receives, and displays the acquired data.

The ANTERION hardware is separated in three parts: the Basis (bottom part), the Imager (top part), and the Head Rest (forehead/chin rest).

For examinations, the patient places his/her head in the forehead/chin rest. The Head Rest is mechanically and electronically connected to the Basis and controlled via a ioystick. Placed within the stand are a stepper motor with additional mechanical parts and a controller board, allowing the operator to move the motorized chin rest up or down for optimally positioning the patients' eye. An external fixation light is mounted at the forehead rest.

The Basis mainly contains the power supply and PC connection of the device. In the Imager, the components for scanning, signal generation, and signal processing are contained.

The operator directly accesses two software modules, which are named AQM (acquisition module) and VWM (viewing module). The AQM allows selecting between examinations. The VWM shows acquired images, parameters, and reports.

The ANTERION device contains two imaging modalities, a scanning optical coherence tomography (OCT) modality and an infrared (IR) camera. The OCT modality allows for cross-sectional imaging, while the IR camera allows for en-face imaging of a patient's eye.

The Imaging App, the subject of this submission, is the foundation of the ANTERION platform, and focuses on the high-resolution visualization of the entire anterior segment, from the anterior surface of the cornea to the posterior surface of the lens.

The Imaging App delivers swept-source OCT images that allow visualization of anterior segment pathologies and evidence of surgical interventions, e.g. keratoplasty, LASIK, implanted IOLs, and phakic lenses.

AI/ML Overview

Here's a breakdown of the acceptance criteria and study details for the ANTERION device, based on the provided text:

Acceptance Criteria and Device Performance

The provided document focuses on establishing substantial equivalence rather than explicit quantitative acceptance criteria with pass/fail thresholds. However, the study aims to demonstrate that the ANTERION's performance is similar to or better than the predicate and reference devices in key areas.

Here's a table summarizing the implicit acceptance criteria (demonstrated similarity or superiority) and the reported device performance:

Acceptance Criteria (Implied)	Reported Device Performance
1. Image Quality of Anterior Segment Structures	Overall, the ANTERION shows superiority in image quality compared to each of the three CIRRUS scans. Specifics shown in Figure 1 indicate a higher percentage of "Good" image quality ratings for ANTERION across all evaluated structures (Cornea/Conjunctiva, Anterior Lens Surface, Iris, Angle, Scleral Spur) compared to CIRRUS scan types (CCN, CAC, CAG). For example, for Cornea/Conjunctiva, ANT showed 83.3% "Good" vs. 52.6% for CCN. This demonstrates ANTERION is at least similar to, and often superior, in image quality.
2. Ability to Identify Abnormality in Anterior Segment (compared to slit lamp reference)	Overall, the ANTERION shows superiority compared to each of the three CIRRUS scans in identifying the abnormality in reference with the slit lamp examination. Figure 2 graphically represents this, showing higher percentages of "Match" (agreement with slit-lamp) for ANTERION (ANT) across various predefined abnormalities (e.g., cornea pterygium, corneal inlay, corneal transplant, trabeculectomy bleb, peripheral iridotomy, glaucoma drainage device) compared to CIRRUS scan types (CAC, CAG/CCN). This indicates ANTERION offers an equal or superior ability in identifying each pre-specified abnormality compared to CIRRUS.
3. Visualization of Key Anatomical Structures	The ANTERION's ability to visualize key anatomical structures was similar to or better than CIRRUS. (This is a conclusion drawn from the image quality assessment where individual structures' visibility was graded).

Study Details:

Sample Size Used for the Test Set and Data Provenance:
- Sample Size: 87 participants (34 in "Normal Anterior Segment group", 53 in "Abnormal Anterior Segment group"). One eye per subject was selected.
- Data Provenance: Prospective, single clinical site in the United States.
Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications:
- Number of Experts: Three independent graders were used for image quality and abnormality identification.
- Qualifications: "Qualified graders" are mentioned. No specific experience (e.g., "radiologist with 10 years of experience") is provided, but they were described as "masked" and "independent."
Adjudication Method for the Test Set:
- Method: Majority rule was applied to get consensus from the three independent graders.
Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- Was it done? Yes, effectively. The study compared the ANTERION device to the CIRRUS HD-OCT 5000 device using multiple graders and multiple cases, assessing both image quality and abnormality identification.
- Effect size of human readers improve with AI vs. without AI assistance: This study does not directly measure the improvement of human readers with AI assistance. Instead, it compares the device's performance (ANTERION) against a comparator device (CIRRUS) for diagnostic imaging purposes. Human graders (experts) were used to evaluate the output of both devices and establish ground truth, not to assess AI assistance to human readers.
Standalone (Algorithm Only) Performance:
- The study evaluates the ANTERION imaging app as a diagnostic device, which produces images and potentially analyses thereof. The assessment of image quality and abnormality identification was performed by human graders viewing the images produced by the device. It is not explicitly stated whether the ANTERION has an "algorithm only" mode that provides automated diagnoses without human review, but the clinical evaluation method suggests a focus on the quality of the images provided for human interpretation rather than fully automated standalone performance.
Type of Ground Truth Used:
- For Abnormality Identification: The slit lamp examination performed by the investigator was used as the reference (ground truth) for identifying abnormalities.
- For Image Quality and Visibility of Structures: Expert consensus (majority rule of three masked graders) was used to grade image quality and visibility of structures on the OCT images from both devices.
Sample Size for the Training Set:
- The document does not provide information on the sample size used for the training set for the ANTERION device's algorithms or software. The clinical study described is a test set evaluation.
How the Ground Truth for the Training Set Was Established:
- The document does not provide this information. The clinical study details focus solely on the evaluation of the device's performance using a test set.

Summary

Regulation Number and Section

§ 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.