Search Results

CASIA2 is a non-contact, high-resolution tomographic and biomicroscopic device intended for the in vivo imaging and measurement of ocular structures in the anterior segment. CASIA2 measures corneal thickness, anterior chamber depth and lens thickness.

The Tomey Cornea/Anterior Segment OCT CASIA2 (CASIA2) is a non-contact, high resolution tomographic and biomicroscopic device indicated for in vivo imaging of ocular structures in the anterior segment. The Tomey Cornea/Anterior Segment OCT CASIA2 is indicated as an aid in the visualization and measurement of anterior segment findings. CASIA2 measures corneal thickness, anterior chamber depth and lens thickness.

This medical device product has functions subject to FDA premarket review (corneal thickness, curvature, anterior chamber depth and lens thickness) as well as functions that are not subject to FDA premarket review. For this application, for the (510(k) exempt functions that are not subject to FDA premarket review, FDA assessed those functions only to the extent that they either could adversely impact the safety and effectiveness of the overall device.

CASIA2 consists of several components: the main unit, AC input power source, a touch panel LCD monitor, an external hard drive (HDD), a mouse and a keyboard.

Here’s a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided FDA 510(k) clearance letter for the Tomey CASIA2.

Device: Tomey Cornea/Anterior Segment OCT (CASIA2)
Measurements evaluated: Central Corneal Thickness (CCT), Anterior Chamber Depth (ACD), and Lens Thickness (LT).

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state pre-defined quantitative acceptance criteria (e.g., "CCT agreement must be within X µm"). Instead, it focuses on demonstrating agreement and precision compared to a legally marketed reference device (LENSTAR LS900). The "acceptance criteria" can be inferred from the study's objective to show substantial equivalence through these performance metrics. The reported device performance is presented as the actual agreement (mean difference and 95% Limits of Agreement - LOA) and precision (Repeatability and Reproducibility %CV).

Therefore, the table below reflects the demonstrated performance and implicitly what was considered acceptable for substantial equivalence.

Note on "Acceptance Criteria": The document implies that meeting or exceeding the performance of the LS900 in terms of precision, and demonstrating good agreement via Bland-Altman analysis, constituted the acceptance for substantial equivalence.

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

• Sample Size for Test Set: A total of 224 subjects were enrolled and completed the study for precision and agreement testing.
  • 55 subjects in the normal group
  • 60 subjects in the cataract group
  • 109 subjects in the special eyes group (eyes without a natural lens or eyes containing artificial materials)
    N for specific analyses (e.g., Agreement Analysis for LT) varied based on acceptable scans (e.g., N=122 for LT agreement, N=138 for CCT/ACD precision, N=76 for LT precision).
• Data Provenance: The document does not explicitly state the country of origin. It indicates "The subjects of this study had no notable or unexpected/untoward assessments..." which suggests a single clinical site. However, no specific country is mentioned.
• Retrospective or Prospective: The study was a prospective clinical study, as subjects were "enrolled," "randomized," and "assigned" to configurations and sequences, and data was collected during the study.

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

• The document implies that the LENSTAR LS900 device itself served as the reference standard for establishing "ground truth" (or more accurately, the comparator for agreement) for the measured parameters.
• It states that "The clinical site had 3 device operators trained on the devices used in the study."
• Qualifications of Experts: The specific qualifications (e.g., radiologist, ophthalmologist, optometrist expertise, years of experience) of these 3 device operators are not explicitly stated in the provided text.

4. Adjudication Method for the Test Set

• The document states, "Additional scans were taken at the operator's discretion if image quality was unacceptable based on the device DFU and the Tomey CASIA2 Reference Guide and included, missing scans, truncated scans, image defocus, floaters, presence of eye blinks, eye motion, etc. Each device operator had up to 3 attempts to obtain an acceptable scan for each of the required scans."
• This suggests an operational approach to ensure data quality rather than a formal, independent adjudication process (e.g., 2+1/3+1 consensus by experts) for the measurements themselves. The "ground truth" was derived from the in-device measurements of the LS900, not a separate expert review. Therefore, there was no expert consensus-based adjudication method for the measurements.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No, a traditional MRMC comparative effectiveness study was not done.
• The study design was focused on device-to-device agreement and precision (CASIA2 vs. LENSTAR LS900) rather than evaluating how human readers' performance (e.g., diagnostic accuracy) improved with or without AI assistance.
• The CASIA2 is described as a "tomographic and biomicroscopic device intended for the in vivo imaging and measurement of ocular structures," with software providing "quantitative outputs." It does not appear to be an AI-assisted diagnostic aid for image interpretation that would typically require an MRMC study to show human reader benefit.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance)

• The study primarily assessed the measurement performance of the CASIA2 device (algorithm/system) in generating quantitative outputs (CCT, ACD, LT) and compared these directly with a reference device. It's implied that these measurements are generated automatically by the device's software.
• The role of the human operators was to acquire an "acceptable scan" based on predefined image quality criteria, not to interpret the images or provide a human "answer" for comparison with an AI-generated reading.
• Therefore, the precision and agreement studies essentially represent the standalone performance of the CASIA2's measurement capabilities compared to the LS900.

7. Type of Ground Truth Used

• The "ground truth" (or clinical reference standard) for comparison was the measurements obtained from the legally marketed predicate/reference device, LENSTAR LS900.
• This is a device-based comparative ground truth, not expert consensus, pathology, or outcomes data. The study aimed to show that the CASIA2's measurements were interchangeable or highly agreeable with those from an established, cleared device.

8. Sample Size for the Training Set

• The document describes a clinical study for validation/testing of the updated software. It does not provide any information about a separate training set size for the development of the algorithms generating these quantitative measurements.
• It only mentions: "The device is a software upgraded version of the predicate K213265 that provides quantitative measurements. All quantitative measurements are derived from OCT images acquired with optical coherence tomography." This implies the software update incorporated algorithms to derive these measurements, but details on their development (including training data) are not provided in this 510(k) summary.

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

• As the document does not describe the training set or its development, there is no information provided on how the ground truth for any training set was established.

Ask a specific question about this device

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 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 super-luminescent diode and a spectral interferometer are used to create the cross-sectional images.
The following modifications have been applied to the device subject of this 510(k):

• Addition of scan acquisitions for the SPECTRALIS OCT Angiography Module (OCTA) at 250 kHz
• Update of the default contrast display setting from 1:4 to 1:2 for the Superficial Vascular Complex (SVC) and the Deep Vascular Complex (DVC) for the acquisition speeds of 125 kHz and 250 kHz

Acceptance Criteria and Study for SPECTRALIS HRA+OCT

The provided FDA clearance letter for the SPECTRALIS HRA+OCT and variants (K250868) describes a retrospective image grading case study (S-2023-1) performed to demonstrate substantial equivalence for modifications to the device. The modifications include the addition of scan acquisitions for the SPECTRALIS OCT Angiography Module (OCTA) at 250 kHz and an update of the default contrast display setting from 1:4 to 1:2 for the Superficial Vascular Complex (SVC) and the Deep Vascular Complex (DVC) for 125 kHz and 250 kHz acquisition speeds.

The study aimed to show that the investigational SPECTRALIS scan types (with 250 kHz acquisition and updated contrast settings) performed similarly to the predicate SPECTRALIS HRA+OCT with OCTA Angiography Module scan types (HR10 @ 85 kHz, HS20 @ 85 kHz) in terms of image quality, visualization of key anatomical vascular structures, and identification of pathologies.

1. Table of Acceptance Criteria and Reported Device Performance

The FDA clearance letter does not explicitly define specific numerical acceptance criteria in the format of a table with pass/fail thresholds. Instead, it reports performance metrics and concludes on "similarity" and "sufficiency" relative to clinical needs and the predicate device. Based on the provided text, the implied acceptance criteria were that the investigational device's performance should be similar to or sufficient for clinical assessment compared to the predicate device.

Here's a summary of the reported device performance, interpreted as meeting these implied criteria:

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

• Sample Size (Test Set): The effectiveness analysis population from the S-2020-5 study consisted of 79 subjects. All 25 Normal subjects and 54 Pathology subjects were included in this retrospective study. However, the exact count for direct comparison between the predicate and investigational device ranged from 74 to 78 subjects depending on the scan type.
• Data Provenance:
  • Country of Origin: United States
  • Retrospective/Prospective: The S-2023-1 image grading case study was retrospective, using clinical data that was collected prospectively in a previous study (S-2020-5). Data was collected at a single clinical site.

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

• Number of Experts: Three independent reviewers from a Reading Center.
• Qualifications of Experts: The document does not explicitly state the specific qualifications (e.g., "radiologist with 10 years of experience") of these reviewers. It only identifies them as "independent reviewers from the Reading Center."

4. Adjudication Method for the Test Set

• The document states that the performance metrics for image quality were "summarized based on the percentage of images graded better than Poor (i.e., Good or Average) on consensus." This indicates that some form of consensus method was used for image quality grading. However, the specific adjudication method (e.g., 2+1, 3+1, majority rule, etc.) for achieving this consensus is not detailed. For visualization of key anatomical structures and identification of pathologies, it indicates agreement analysis between predicate and investigational scan types but does not explicitly describe an adjudication method to establish a single "ground truth" before comparison; rather, it assesses agreement between the two device types.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• The study described is an image grading study involving multiple readers (three independent reviewers) evaluating multiple cases, comparing an investigational device's performance to a predicate device. While it aligns with elements of a comparative effectiveness study, it's not explicitly labeled as a "Multi Reader Multi Case (MRMC) comparative effectiveness study" in the statistical sense (e.g., for ROC analysis). Instead, it focuses on agreement rates, PPA, and NPA.
• Effect Size of Human Readers' Improvement with AI vs. without AI: This study does not involve AI assistance for human readers. The device in question is a medical imaging device (OCTA), not an AI-powered diagnostic tool providing automated interpretations or assisting human readers. Therefore, there is no reported effect size for human readers improving with AI vs. without AI assistance. The study compares two versions of the imaging device.

6. Standalone Performance Study

• Yes, a standalone performance assessment was conducted for the investigational scan types. The reported metrics for "Overall Image Quality" and "Visualization of Key Anatomic Structures" are measures of the algorithm's output (images from the investigational scan types) as graded by experts, independent of a human-in-the-loop scenario for diagnostic decision-making. The agreement analysis is essentially comparing the standalone performance of the investigational scans against the standalone performance of the predicate scans.

7. Type of Ground Truth Used

• The "ground truth" for the test set was primarily established through expert consensus/grading by three independent reviewers.
• The "Normal population" was defined by clinical examination showing no retinal conditions or abnormalities.
• The "Pathology population" had specific retinal conditions (e.g., wet age-related macular degeneration, diabetic retinopathy) and abnormalities (e.g., microaneurysm, choroidal neovascularization, retinal neovascularization) that were identified. This implies medical record review and possibly other diagnostic findings contributed to classifying these subjects, but the direct "truth" for the study's performance metrics (image quality, structure visualization, abnormality identification) came from the expert grading of the OCTA scans.

8. Sample Size for the Training Set

• The document does not specify a sample size for a training set. The study primarily focuses on the validation of modifications to an existing device, and the data listed is related to its verification and clinical evaluation (test set). It is possible that the underlying algorithms within the SPECTRALIS were trained on a separate, unmentioned dataset prior to this 510(k) submission, but this information is not provided in the clearance letter.

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

• As the document does not provide information about a training set, it does not describe how the ground truth for any training set was established.

Ask a specific question about this device

OcuMet Beacon is a confocal scanning ophthalmoscope indicated for infrared (IR) and autofluorescence (AF) imaging of a human retina with or without the use of a mydriatic agent.

Ocumet Beacon is a scanning LED based ophthalmoscope which uses infrared and blue light to obtain confocal images of the retina. Multiple retinal fields can be captured using a programmable internal fixation target. The device integrates a tablet and is provided with an external power supply. The device works with a dedicated software application and operates as a standalone unit.

Ocumet Beacon offers two different acquisition modalities:

• black and white reflectance images of the retina obtained using infrared illumination;
• black and white fluorescence images of the retina obtained using blue illumination (peak at 458 nm) and a barrier filter to select only autofluorescent emission between 520 and 540 nm.

Clinician Report Generator Software is required to open, review, analyze, and print images obtained from Ocumet Beacon. The software runs on a standalone laptop.

The total light exposure is qualified as Group 1 under ANSI Z80.36-2016

Clinician Report Generator Software is required to open, review, analyze and print a co-registered version of both the infrared and visible wavelength images obtained from Ocumet Beacon. The software runs on a standalone Windows 10 or 11 based laptop computer.

The clinical interpretation of the images acquired by Ocumet Beacon is restricted to licensed eye care practitioners. A device specific training is required for any operator to become able to use the system and certification training is available for operators who will be capturing images in a clinical trial. The interpretation and use in clinical practice is as an adjunctive modality for the clinician to optionally combine with other structural images, and functional studies.

I am sorry, but the provided text from the FDA 510(k) clearance letter for OcuMet Beacon (K241931) does not contain specific details regarding acceptance criteria, study design, or performance metrics from a clinical study proving the device meets acceptance criteria.

The document primarily focuses on establishing substantial equivalence to predicate devices (EIDON FA/AF) and a reference device (Heidelberg Spectralis HRA+OCT) based on:

• Indications for Use: Stating they are similar.
• Device Type: Describing both as confocal scanning ophthalmoscopes.
• Operational Modalities: Comparing features like IR and AF imaging.
• Technical Specifications: Such as field of view (where OcuMet Beacon has a reduced FoV, noted as "Similar" but explainable by multiple acquisitions), light sources, electrical safety, and physical dimensions.
• Safety Standards: Compliance with IEC 60601-1, IEC 60601-1-2, ANSI Z80.36-2016, and ISO 15004-2:2007 (for light hazard).
• Biocompatibility: Stating patient contact points meet requirements.
• System Verification and Validation: Broadly stating the system "was tested and passes."

Crucially, the document does NOT include:

• A table of acceptance criteria with specific quantitative targets (e.g., sensitivity, specificity, accuracy thresholds).
• Reported device performance data against those criteria.
• Details on the sample size used for any test set or the provenance of data (e.g., retrospective/prospective, country of origin).
• Information on the number or qualifications of experts used for ground truth.
• Adjudication methods.
• Any mention of a Multi-Reader Multi-Case (MRMC) comparative effectiveness study, including human reader improvement with/without AI assistance.
• Data on standalone algorithm performance.
• The type of ground truth used (e.g., pathology, expert consensus).
• The sample size for the training set or how its ground truth was established.

Therefore, I cannot fulfill your request to describe the acceptance criteria and the study that proves the device meets them based solely on the provided text. The 510(k) summary provided here focuses on demonstrating substantial equivalence, not on presenting results from a comprehensive clinical performance study with specific acceptance criteria and performance data. Such information would typically be found in a more detailed clinical study report or a different section of the 510(k) submission, which is not included in this excerpt.

Ask a specific question about this device

MAIA is intended for taking digital images of a human retina without the use of a mydriatic agent.

MAIA is intended for taking digital images of a human retina without the use of a mydriatic agent.

The device works with a dedicated software application, operates as a standalone unit, integrates a multitouch display, a push-button and is provided with an external power supply.

MAIA operates in non-mydriatic conditions, i.e. without the need of pharmacological dilation and is intended for prescription use only

The key functional elements of the device are:

• the device base,
• the device optical head,
• the headrest,
• the chinrest,
• the multi-touch display,
• the embedded software,
• patient push-button,
• the power supply and its power cable.

The device acquires confocal retinal images illuminating the retina of the patient's eye, with visible light for imaging purposes and with infrared light for imaging purposes, focusing and retinal tracking.

A specific feature of MAIA is the pupil tracking. The patient's pupils are illuminated and viewed by a different optical system; two cameras track the pupil's movements and allow for automatic alignment of the optical head with the eye position.

From an imaging point of view, an important characteristic of MAIA is that it is a confocal instrument: this means that the illumination of the retina is focused on the same plane of the acquisition (same focus). Compared to traditional imaging, where the entire specimen is flooded evenly in light, a confocal illumination system is able to focus the illuminating light on the same focus plane (conjugate plane) of the acquisition optics.

Another important characteristic of MAIA is the use of white light to illuminate the retina.

The result of the combination of confocal imaging and white light obtained in MAIA is the acquisition of sharp, detailed, naturally colored retinal images.

This medical device product has functions subject to FDA premarket review as well as functions that are not subject to FDA premarket review.

For this application, the product has also functions that are not subject to FDA premarket review; FDA assesses those functions only to the extent that they either could adversely impact the safety and effectiveness of the functions subject to FDA premarket review or they are included as a labeled positive impact that was considered in the assessment of the functions subject to FDA premarket review.

The "other function" of the device is the automatic perimetry (Product Code HPT, 510(K) Exempt), that allows the measurement of retinal threshold sensitivity and the analysis of fixation.

This function consists in projecting light stimuli at different intensities together with a uniform light background; the device records the pressures of the pushbutton by the tested patient when he detects such stimuli, as in Standard Automated Perimetry test.

With its retinal tracking, for all the duration of the exam, MAIA acquires infrared images of the retina and detects its movement, to correct the position for the stimuli, reducing the positioning error that might occur if the patient has poor fixation stability.

The provided text describes the Centervue Spa MAIA (AHMACME001) device, which is an ophthalmoscope intended for taking digital images of a human retina. The 510(k) submission (K243504) seeks to demonstrate substantial equivalence to the predicate device COMPASS (K150320).

However, the provided document explicitly states: "No clinical tests were needed." and "Based on the non-clinical tests (i.e. bench tests), MAIA is safe and performs as intended when used according to its indications for use and in accordance with its labeling. It performs as well as the legally marketed predicate device COMPASS (K150320)." This means that the submission does not contain a study proving the device meets acceptance criteria related to a clinical study involving human performance (e.g., MRMC studies, standalone algorithm performance, expert ground truth).

The acceptance criteria described pertain to design and manufacturing equivalence, and bench testing, rather than clinical performance for a diagnostic device. Since no clinical study was conducted or provided in this submission, it is impossible to populate most of the requested fields regarding performance, sample sizes, expert ground truth, adjudication, or MRMC studies.

Here's how to address the request based only on the provided information, acknowledging the absence of a clinical study:

Acceptance Criteria and Device Performance (Based on Non-Clinical Tests)

Since no clinical studies were performed to establish performance metrics against human-defined acceptance criteria (e.g., sensitivity, specificity, accuracy), the "acceptance criteria" here refer to the successful completion of various engineering and safety tests demonstrating equivalence to the predicate device.

Study Details (None for Clinical Performance)

1. Sample size used for the test set and the data provenance: Not applicable. No clinical test set. The submission relies on "non-clinical tests (i.e. bench tests)" and equivalence to the predicate device.
2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable. No clinical test set requiring expert ground truth for performance evaluation.
3. Adjudication method (e.g. 2+1, 3+1, none) for the test set: Not applicable. No clinical test set.
4. 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: Not applicable. No MRMC study was performed. The device is for "taking digital images," not providing AI assistance for diagnosis.
5. If a standalone (i.e. algorithm only without human-in-the loop performance) was done: Not applicable for performance metrics like sensitivity/specificity for a diagnostic outcome. The device's primary function described is image acquisition, a standalone function that was validated via bench testing and comparison to the predicate's technology.
6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.): Not applicable for clinical performance. The ground truth for the "technical performance" and "safety" of the device (such as image resolution, field of view, and electrical safety) would be established by engineering specifications, calibration standards, and regulatory safety standards.
7. The sample size for the training set: Not applicable. No machine learning training set mentioned or implied for a diagnostic algorithm. The device, an ophthalmoscope, acquires images. While it has components like pupil tracking and automatic alignment, these are presented as inherent functions validated through engineering tests, not trained AI models.
8. How the ground truth for the training set was established: Not applicable. No training set for a diagnostic algorithm.

Ask a specific question about this device

The ANTERION is a non-contact ophthalmic imaging and analysis device for the eye. It is intended for visualization and measurement of the anterior segment and measurement of the axial length.
The analysis covers:

• · Cornea Thickness
• · Anterior Segment
  o Anterior chamber width, depth, volume and angle parameters
  o Lens Thickness
• · Axial Length

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/Base) 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 (V.1.5) (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 ANTERION software provides an interface for a Medical Image Management and Processing System.
The ANTERION hardware is separated in three parts: the Base (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 Base and controlled via a joystick. Within its stand, a stepper motor with additional mechanic parts and a controller board are placed, 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 Base 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 and biometry, while the IR camera allows for en-face imaging of a patient's eve.
The ANTERION device provides four separate software functionalities (Apps) to acquire various imaging and measurements of the anterior segment of the eye: (1) the Imaging App. (2) the Cornea App. (3) the Cataract App and (4) the Metrics App. The Cornea App provides tomographic data and measurements for the patient´s individual corneal geometry and corneal characteristics. The Cornea App provides tomographic data and parameters, such as corneal curvature and thickness. The Cataract App provides key measurements for cataract surgery planning, such as corneal thickness, anterior chamber depth and axial length. The Metrics App generates OCT images and scan parameters for the anterior chamber such as anterior chamber angle and volume. The four ANTERION Apps are locked/unlocked independently by a license mechanism for each App. The software implementation of these Apps is realized within the AQM and VWM.
The following modification has been applied to the device, subject of this 510(k):

• · Addition of the Epithelial Thickness Module (separate License in the Cornea App) with maps and parameters of the corneal epithelial and stromal thickness.
  To function as intended, the ANTERION must be connected to a Medical Image Management and Processing system (MIMPS) with compatible interface. To date, HEYEX 2 / HEYEX PACS is the only available MIMPS with compatible interface.

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

The document describes a comparative study, implying that the acceptance criteria for the "ANTERION" device's new Epithelial Thickness Module were based on its precision (repeatability and reproducibility) and agreement with an existing legally marketed device, the "Cirrus HD-OCT 5000 with Anterior Segment Premier Module."

The acceptance criteria are implicitly defined by demonstrating similar or better precision and general agreement with the predicate device across various corneal epithelial thickness measurements and patient populations.

Table of Acceptance Criteria and Reported Device Performance

Study Details:

1. Sample sizes used for the test set and data provenance:

   • Total Subjects: 115 subjects
     • 32 in Normal Cornea population (Group A)
     • 82 in Abnormal Cornea population (Group B), further divided into:
       • 25 Keratoconus subgroup
       • 20 Contact Lens Wearer subgroup
       • 18 Status Post-Keratorefractive Surgery subgroup
       • 19 Dry Eye Disease subgroup
   • Data from 32 Group A and 81 Group B participants were included in precision analyses.
   • Data from 32 Group A and 80 Group B participants were included in agreement analyses.
   • Provenance: Prospective, randomized precision and agreement clinical study conducted at one diverse clinical site in the United States.
   • Scans per parameter/population: Vary (e.g., 285 scans for most Normal Cornea parameters, 725 for most Abnormal Cornea parameters). Each participant had 3 replicates per acquisition type, per configuration from three device-operator configurations.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • This study does not establish ground truth using experts in the traditional sense (e.g., radiologists interpreting images). Instead, it's a comparative effectiveness study where the ANTERION device's measurements are compared to those of a legally marketed predicate device (CIRRUS HD-OCT 5000). The "ground truth" implicitly refers to the measurements obtained by the predicate device, or rather, the comparison is made between the measurements of the two devices, not against an external expert-derived truth. Expertise was involved in operating the devices and assessing image quality by the operator, but not in establishing a separate "ground truth" for the measurements themselves.

3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

   • Not explicitly stated for the test set measurements. The study design involved "three replicates per acquisition type, per configuration" and "image quality was assessed by the operator after each acquisition." This suggests that the operators evaluated the quality of the individual scans, but there is no mention of an independent adjudication process for the actual thickness measurements or a consensus method for defining "ground truth" between multiple readers.

4. 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:

   • This was a comparative effectiveness study involving multiple operators ("three operators") and multiple cases (115 subjects / ~114 study eyes). However, it was not an MRMC study focused on human reader improvement with/without AI assistance. This study compared measurements from one device (ANTERION) to another (CIRRUS HD-OCT 5000) for precision and agreement. The "ANTERION" device is an ophthalmic imaging and analysis device, and while it has "Apps" that perform analysis, the study focuses on the device's measurement performance rather than an AI component assisting human readers in diagnosis.
   • Therefore, there is no effect size reported regarding human readers improving with/without AI assistance in this context.

5. If a standalone (i.e. algorithm only without human-in-the loop performance) was done:

   • Yes, the study primarily evaluates the standalone performance of the ANTERION device's measurement capabilities (specifically the Epithelial Thickness Module) against a predicate device. While human operators collected the images, the reported precision (repeatability and reproducibility) and agreement data are quantifications of the device's algorithmic measurement outputs. The "Cornea App" and its Epithelial Thickness Module perform the thickness calculations.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc):

   • The study uses the measurements from a legally marketed predicate device (CIRRUS HD-OCT 5000) as the reference for comparison, rather than an independent "ground truth" like pathology or expert consensus. The aim is to show that the new device's measurements are sufficiently precise and agree with those from an established device.

7. The sample size for the training set:

   • The document does not provide information on the training set for the ANTERION device's algorithms. The study described is entirely focused on the clinical performance testing (test set) of the device.

8. How the ground truth for the training set was established:

   • As no information on the training set is provided, how its "ground truth" was established is also not available in this document.

Ask a specific question about this device

The SPECTRALIS with Flex Module is a non-contact ophthalmic diagnostic imaging device intended to aid in the visualization of the posterior segment structures of the eye and vasculature of the retina and choroid. SPECTRALIS with Flex Module is intended for imaging of adults and pediatric patients in supine position.

The Heidelberg Engineering SPECTRALIS with Flex Module is based on the predicate SPECTRALIS HRA+OCT consisting of an accessory device mount allowing imaging of patients in supine position. The SPECTRALIS with Flex Module is intended for visualization of the posterior segments of the human eye. The SPECTRALIS with Flex Module is using the identical technologies as the predicate SPECTRALIS tabletop configuration (K223557), i.e. it is a combination of a confocal laser-scanning ophthalmoscope (cSLO, the HRA portion) and a spectral-domain optical coherence tomographer (SD-OCT).

Here's the breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

Device Name: SPECTRALIS with Flex Module
Predicate Device: SPECTRALIS HRA+OCT (K223557)

• 100% of OCT images were assessed as having sufficient overall image quality.
• 100% of IR cSLO images were assessed as having sufficient overall image quality.
• 97.2% of OCTA images were assessed as having sufficient overall image quality. (Page 14) |
  | Visibility of Key Anatomic Structures | Key anatomic structures should be clinically acceptably visualized. | In the effectiveness population (All Eyes):
• Clinically acceptable visualization of key anatomic structures was achieved on OCT in 100% of assessments.
• On OCTA, clinically acceptable visualization was achieved in 94.4% to 97.2% of assessments. (Page 14)
  The results generally demonstrate that investigational SPECTRALIS with Flex images provide similar visualization as compared to the predicate SPECTRALIS. (Page 14-15) |
  | Identification of Pre-specified Abnormalities | The device should have comparable ability to identify pre-specified abnormalities (structural via OCT, vascular via OCTA) compared to the predicate device. | - Agreement rates (between investigational device and predicate for same result):
  • Structural abnormalities on OCT: ≥ 84.3%.
  • Vascular abnormalities on OCTA: ≥ 85.1%.
• Negative Percent Agreement (NPA) in All Eyes for all pre-specified abnormalities:
  • On OCT: at least 87.8%.
  • On OCTA: at least 92.6%.
• Positive Percent Agreement (PPA) for All Eyes for all pre-specified abnormalities (with more than 2 cases identified on the predicate):
  • On OCT: at least 75%.
  • On OCTA: at least 75%.
    These results indicate the ability to identify each pre-specified abnormality is similar between devices. (Page 15) |
    | Safety | The device should not introduce new safety concerns. | No adverse events occurred during the course of the study. (Page 14) |
    | Substantial Equivalence (General) | The device should be as safe and effective as the predicate devices. | The study concludes that the SPECTRALIS with Flex Module is substantially equivalent to the predicate SPECTRALIS with regards to image quality, visibility of key anatomic structures, and identification of structural and vascular abnormalities, and supports its safety. (Page 15) |
    | Supine & Pediatric Patients | Demonstrate effective and safe imaging for pediatric and adult patients in the supine position. | The clinical study included adult patients (22 years or older) and the literature review provides evidence for safe and effective use in pediatric conscious or unconscious patients in the supine position. The device shares the patient profile (conscious/unconscious pediatrics and adults in supine position) with the secondary predicate device (Bioptigen ENVISU). (Page 15-16) |

Study Details

1. Sample size used for the test set and the data provenance:

   • Subjects Enrolled: 88
   • Subjects Completed Study: 84
   • Effectiveness Analysis Population: 76 participants (25 Normal subjects, 51 Posterior Segment Abnormality subjects). The exact count varied slightly for each image type based on abnormalities of interest and acceptable acquisitions.
   • Data Provenance: Single clinical site, located in the United States. The study was prospective and observational.
   • Patient Demographics:
     • Mean age: 57.3 ± 19.2 years (overall)
     • Gender: 59.2% female, 40.8% male
     • Ethnicity: 96.1% did not identify as Hispanic or Latino
     • Race: 80% White, 18.4% Black/African American, 2.6% Asian
   • Study Populations:
     • Adult Normal Eyes (no posterior segment abnormalities)
     • Adult Posterior Segment Abnormality Eyes (structural and/or vascular posterior segment abnormalities)

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Number of Experts: Three independent readers.
   • Qualifications: Referred to as "independent readers from a reading center." No specific qualifications (e.g., years of experience, subspecialty) are provided in this extract.

3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

   • The text states that the "proportion of images graded better than Poor on consensus" was used. For abnormality identification, "agreement rates (based on the abnormalities identified by the reading center on the predicate, percentage of eyes with the same result on the investigational device)" were utilized. This implies that the three readers likely had a consensus process for the overall image quality and visibility assessments, and for abnormality identification, agreement with the predicate's findings by the reading center was important. The specific 2+1 or 3+1 method is not explicitly mentioned, but the term "consensus" suggests an agreed-upon finding among the readers.

4. 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:

   • This was not an MRMC comparative effectiveness study involving AI assistance. The study was designed to show substantial equivalence between the investigational device (SPECTRALIS with Flex Module) and its predicate (SPECTRALIS HRA+OCT) regarding image quality and ability to identify abnormalities. There is no mention of AI or its impact on human reader performance.

5. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:

   • No, this was not a standalone (algorithm-only) study. The study involved human readers (three independent readers) to grade images obtained from the device.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • The ground truth for image quality, visibility of structures, and identification of abnormalities was established through expert grading by three independent readers from a reading center, likely using expert consensus to interpret findings. The predicate device's findings also served as a reference for agreement in abnormality detection.

7. The sample size for the training set:

   • The provided text does not contain information about a training set size. The clinical study described is for validation and comparison to a predicate, not for training a new algorithm.

8. How the ground truth for the training set was established:

   • As no training set is mentioned, this information is not applicable based on the provided document.

Ask a specific question about this device

· 3D OPTICAL COHERENCE TOMOGRAPHY 3D OCT-1(type: Maestro2)

The Topcon 3D Optical Coherence Tomography 3D OCT-1 (Type:Maestro2) is a non-contact, high resolution tomographic and biomicroscopic imaging device that incorporates a digital camera for photographing, displaying and storing the data of the retina and surrounding parts of the eye to be examined under mydriatic conditions.

It is indicated for in vivo viewing, axial cross sectional, and three-dimensional imaging and measurement of posterior ocular structures, including retinal nerve fiber layer, macula and optic disc as well as imaging of anterior ocular structures.

It also includes a reference database for posterior ocular measurements which provide for the quantitative comparison of retinal nerve fiber layer, optic nerve head, and the human retina to a database of known normal subjects.

It is indicated for use as a diagnostic device to aid in the diagnosis, documentation and management of ocular health and diseases in the adult population.

All the above functionalities and indications are available in combination with IMAGEnet 6.

· Indications for Use of the combination of the Maestro2 in conjunction with IMAGEnet6

Maestro2 in combination with IMAGEnet 6 is indicated as an aid in the visualization of vascular structures of the posterior segment of the eye including the retina, optic disc and choroid.

• · IMAGEnet6 Ophthalmic Data System
  The IMAGEnetto Ophthalmic Data System is a software program that is intended for use in the collection, storage and management of digital images, patient data, diagnostic data and clinical information from Topcon devices.

It is intended for processing and displaying ophthalmic images and optical coherence tomography data.

The IMAGEnet6 Ophthalmic Data System uses the same algorithms and reference databases from the original data capture device as a quantitative tool for the comparison of posterior ocular measurements to a database of known normal subjects.

· Indications for Use of the combination of the Maestro2 in conjunction with IMAGEnet6

Maestro2 in combination with IMAGEnet 6 is indicated as an aid in the visualization of vascular structures of the posterior segment of the eye including the retina, optic disc and choroid.

3D OPTICAL COHERENCE TOMOGRAPHY 3D OCT-1 (type: Maestro2) with system linkage software (herein referred to as "Maestro2") is a non-contact ophthalmic device combining spectral-domain optical coherence tomography (SD-OCT) with digital color fundus photography. Maestro2 includes an optical system of OCT, fundus camera (color, IR and Red-free image), and anterior observation camera. Maestro2 is used together with IMAGEnet6 by connecting via System linkage software which is a PC software installed to off the shelf PC connected to Maestro2 is capable of OCT imaging and color fundus photography. For this 510(k) notification, Maestro2 has been modified to allow for OCT "angiographic" imaging (only in conjunction with IMAGEnet6).

IMAGEnet6 is a software program installed on a server computer and operated via web browser on a client computer. It is used in acquiring, storing, managing, processing, measuring, displaying of patient information, examination information and image information delivered from TOPCON devices.

When combined with Maestro2, IMAGEnet6 plays an essential role as the user interface of the external PC by working together with the linkage software of Maestro2. In this configuration, IMAGEnet6 performs general GUI functions such as providing of the log-in screen, display of the menu icons, display function, measurement, analysis function, image editing functions, storing and management of data of the captured OCT scans and provides the reference database for quantitative comparison. For this 510(k) notification, IMAGEnet6 has been modified to allow for OCT "angiographic" imaging on the Maestro2.

Here's a summary of the acceptance criteria and study details for the Topcon Corporation's 3D OCT-1 (Maestro2) and IMAGEnet6 Ophthalmic Data System, primarily focusing on the new OCT Angiography functionality:

Acceptance Criteria and Device Performance

The study primarily focused on comparing the performance of the Maestro2 (with IMAGEnet6) against the predicate CIRRUS HD-OCT, particularly for its new OCT Angiography (OCTA) imaging capabilities. The acceptance criteria were implicitly defined through the evaluation of "response rates" and agreement metrics.

Table of Acceptance Criteria and Reported Device Performance

• 3x3-mm macular scan: 75.0%
• 6x6-mm macular scan: 71.0%
• 4.5x4.5-mm disc scan: 71.0%
  Pathology Group:
• 3x3-mm macular scan: 75.6%
• 6x6-mm macular scan: 77.9%
• 4.5x4.5-mm disc scan: 74.4% |
  | Visibility of Key Anatomical Vascular Features Response Rate (FAZ, large, medium, small vessels/capillaries - Maestro2 scans same or better grade than CIRRUS HD-OCT) | High percentage indicating comparable or superior visibility of features. | Entire Cohort:
• 3x3-mm Macular Scan: FAZ visibility 87.1%, medium vessels 87.9%, small vessels/capillaries 82.3%
• 6x6-mm Macular Scan: FAZ visibility 81.5%, large vessels 87.1%, medium vessels 77.4%, small vessels/capillaries 79.8%
• 4.5x4.5-mm Disc Scan: large vessels 83.9%, medium vessels 80.6%, small vessels/capillaries 80.6% |
  | Positive Percent Agreement (PPA) for Pathological Vascular Features (Maestro2 vs. FA/ICGA) | High PPA indicating good sensitivity in identifying pathologies. | Microaneurysms (MAs):
• 3x3-mm macular scans: 96.6%
• 6x6-mm macular scans: 96.6%
• 4.5x4.5-mm disc scans: 73.9%
  Retinal Ischemia/Capillary Dropout (RI/CD):
• 3x3-mm scans: 93.1%
• 6x6-mm scans: 100%
• 4.5x4.5-mm disc scans: 75.0%
  Choroidal Neovascularization (CNV):
• 3x3-mm macular scans: 88.9%
• 6x6-mm macular scans: 84.2%
• 4.5x4.5-mm disc scans: 66.7% |
  | Negative Percent Agreement (NPA) for Pathological Vascular Features (Maestro2 vs. FA/ICGA) | High NPA indicating good specificity in ruling out pathologies. | Microaneurysms (MAs):
• 3x3-mm macular scans: 92.7%
• 6x6-mm macular scans: 92.7%
• 4.5x4.5-mm disc scans: 100%
  Retinal Ischemia/Capillary Dropout (RI/CD):
• 3x3-mm scans: 85.4%
• 6x6-mm scans: 87.8%
• 4.5x4.5-mm disc scans: 85.7%
  Choroidal Neovascularization (CNV):
• 3x3-mm macular scans: 82.7%
• 6x6-mm macular scans: 84.3%
• 4.5x4.5-mm disc scans: 98.4% |
  | Response Rate for Identification of Pathologies (Maestro2 scans same or better outcome than CIRRUS HD-OCT) | High percentage indicating comparable or superior ability to identify specific pathologies. | Microaneurysms (MAs) response rates:
• 3x3-mm macular scans: 81.0%
• 6x6-mm macular scans: 82.1%
• 4.5x4.5-mm disc scan: 75.6%
  RI/CD response rates:
• 3x3-mm macular scans: 76.2%
• 6x6-mm macular scans: 79.8%
• 4.5x4.5-mm disc scans: 71.8%
  CNV response rates:
• 3x3-mm macular scans: 79.8%
• 6x6-mm macular scans: 81.0%
• 4.5x4.5-mm disc scan: 75.3% |

Study Details

1. Sample size used for the test set and the data provenance:

   • Sample Size: 124 eligible eyes from 122 subjects. This included 38 "normal" eyes and 86 "pathology" eyes.
   • Data Provenance: Prospective, multi-center, observational study. The country of origin of the data is not explicitly stated in the provided text, but "multi-center" suggests data from various clinical sites.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Number of Experts: Not explicitly stated. The text mentions that images were sent to an "independent reading center (RC) for image grading," implying multiple experts, but the exact number is not provided.
   • Qualifications of Experts: Not explicitly stated. The nature of the study (comparing OCTA images and identifying vascular pathologies) suggests that the graders at the independent reading center would be ophthalmologists or trained image graders with expertise in retinal imaging and pathology.

3. Adjudication method for the test set:

   • The document implies that an "independent reading center (RC)" performed "image grading." The specific adjudication method (e.g., 2+1, 3+1, none) is not explicitly stated.

4. 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:

   • This was a comparative effectiveness study, but it primarily compared device performance (Maestro2 vs. CIRRUS HD-OCT) and visualization capabilities against a reference standard (dye-based angiography), rather than assessing improvement in human readers with AI assistance versus without. The OCTA function itself is part of the imaging device, providing images for human interpretation, not an AI assisting human reads. Therefore, an effect size of human readers improving with/without AI assistance is not applicable to this study design as described. The study aims to demonstrate that the new device's OCTA images are comparable or superior to the predicate device and aid in visualizing vascular structures.

5. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:

   • The document describes the device as providing visualization capabilities ("indicated as an aid in the visualization"). The PPA/NPA results compare the device's diagnostic capability for certain pathologies against dye-based angiography, suggesting a standalone assessment of the image data's ability to reveal these pathologies. However, the exact methodology for pathology identification (e.g., whether it relied purely on automated detection within the device or expert interpretation of the OCTA images generated by the device) is not fully detailed. Given the context of "image grading" by a reading center, it strongly suggests expert interpretation of the images produced by the Maestro2 (device-only output).

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • The ground truth for identifying key pathological vascular features (Microaneurysms, Retinal Ischemia/Capillary Dropout, and Choroidal Neovascularization) was established by dye-based angiography (e.g., fluorescein angiography [FA] and indocyanine green angiography [ICGA]). This is referred to as the reference standard against which the OCTA images from Maestro2 and CIRRUS HD-OCT were compared. Additionally, "clinically significant pathology" as determined by clinical assessment was used to categorize the "Pathology Population."

7. The sample size for the training set:

   • The document only describes a clinical performance test study. Information regarding a specific training set size is not provided within this document. The device uses algorithms and reference databases, suggesting prior training, but details on that process or data size are absent.

8. How the ground truth for the training set was established:

   • As no information on a specific training set is provided, how its ground truth was established is also not detailed in this document. The document mentions "reference databases for posterior ocular measurements" (e.g., for normal subjects), implying ground truth for these databases would have been established through prior clinical studies or expert consensus on normal ocular anatomy.

Ask a specific question about this device

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 imaqinq
• · 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 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 crosssectional images.

The following modifications have been applied to the device subject of this 510(k):

• . Addition of scan acquisitions for the SPECTRALIS OCT Angiography Module (OCTA) at 125 kH
• Addition of a General-Purpose Graphics Processing Unit (GPGPU)

Here's a breakdown of the acceptance criteria and study details based on the provided FDA 510(k) summary:

Device: SPECTRALIS HRA+OCT and variants

1. Table of Acceptance Criteria and Reported Device Performance

The 510(k) summary doesn't explicitly state "acceptance criteria" with numerical thresholds in the typical sense for a pass/fail. Instead, it demonstrates similarity and non-inferiority to a predicate device. The performance metrics reported serve as evidence that the new modifications do not negatively impact the device's functionality compared to the predicate.

Overall Conclusion from Study: The investigational SPECTRALIS OCTA images provide similar visibility as compared to the predicate (85 kHz) and the ability to identify each pre-specified vascular abnormality is similar between the predicate and investigational scan types.

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

• Sample Size (Effectiveness Analysis Population): 79 subjects. However, the exact count for direct comparison between the predicate and investigational device varied depending on the scan type.
• Data Provenance:
  • Country of Origin: United States
  • Retrospective or Prospective: Prospective
  • Study Design: Observational Case Study (S-2020-5)

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

• Number of Experts: Three independent reviewers.
• Qualifications of Experts: The document states they were "from a reading center" but does not specify their individual qualifications (e.g., "radiologist with 10 years of experience").

4. Adjudication Method for the Test Set

The document explicitly states that the analyses were based on "the grading results from the effectiveness analysis population," and imagery was "graded better than Poor on consensus." This implies a consensus-based adjudication method for image quality and visualization of structures, and agreement analysis for vascular abnormalities. It does not specify a 2+1 or 3+1 rule, but highlights the "consensus" aspect.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No, a typical MRMC comparative effectiveness study, designed to measure how much human readers improve with AI vs. without AI assistance, was not conducted. This study's primary goal was to demonstrate that modifications to an existing device (adding new scan acquisitions, increasing scan speed, and GPU processing) did not negatively impact its performance compared to its predicate. It assesses the similarity of the device's output (images) between the investigational and predicate versions.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance)

Since the device in question is an imaging device (OCT) that captures and processes images for clinical evaluation by a human, and the study involves human graders assessing image quality and identifying pathologies, a "standalone algorithm-only" performance evaluation (without human-in-the-loop) in terms of clinical interpretation was not the primary focus or design of this particular study. The assessment revolves around the quality of the device's output for human interpretation.

7. Type of Ground Truth Used

The ground truth was established by expert consensus from the three independent reviewers from a reading center, who graded the OCTA scans on image quality, visibility of key anatomical vascular structures, and identification of pathologies.

8. Sample Size for the Training Set

The document does not provide information about a training set. This study is a clinical performance evaluation of an updated imaging device, not typically a machine learning model that requires a discrete training set for its core function. The modifications involve hardware (scan speed, GPU) and an investigational scan type, and the performance assessment is against a predicate device.

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

As no training set is mentioned or implied for this type of device modification study, this information is not applicable.

Ask a specific question about this device

The CIRRUS™ HD-OCT is a non-contact, high resolution tomographic and biomicroscopic imaging device. It is indicated for in-vivo viewing, axial cross-sectional, and three-dimensional imaging and measurement of anterior and posterior ocular structures, including cornea, retinal nerve fiber layer, ganglion cell plus inner plexiform layer, macula, and optic nerve head.

The CIRRUS™ HD-OCT Reference Database is a quantitative tool used for the comparison of retinal nerve fiber layer thickness, macular thickness, ganglion cell plus inner plexiform layer thickness, and optic nerve head measurements to a database of healthy subjects.

CIRRUS™ HD-OCT AngioPlex angiography is indicated as an aid in the visualization of vascular structures of the retina and choroid.

The CIRRUS™ HD-OCT is indicated for use as a diagnostic device to aid in the detection and management of ocular diseases including, but not limited to, macular holes, cystoid macular edema, diabetic retinopathy, age-related macular degeneration, and glaucoma.

The subject device is a computerized instrument that acquires and analyses cross-sectional tomograms of anterior ocular structures (including comea, retinal nerve fiber layer, macula, and optic disc). It employs non-invasive, non-contact, low-coherence interferometry to obtain these high-resolution images. CIRRUS 6000 has a 100kHz scan rate for all structural and angiography scans.

The subject device uses the same optical system, and principle of operation as the previously cleared CIRRUS 6000 (K222200) except for the reference database functionality.

The subject device contains a newly acquired reference database which was collected on K222200. This study data compares macular thickness, ganglion cell thickness, optic disc and RNFL measurements to a reference range of healthy eyes as guided by the age of the patient and /or optic disc size. Reference database outputs are available on Macular Cube 200x200, and Optic Disc Cube 200x20 scan patterns. All other technical specifications have remained the same as the predicate K222200.

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided text:

Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly met by the successful development of the CIRRUS™ HD-OCT Reference Database (RDB) and its ability to provide normative data for comparison. The study aims to establish these reference limits.

Study Details

1. Sample Size and Data Provenance:

   • Test Set (for RDB establishment): 870 subjects had one eye included in the analysis from an initial enrollment of 1000 subjects.
     • Data Provenance: Prospective, multi-site study conducted at eight (8) clinical sites across the USA.

2. Number of Experts and Qualifications for Ground Truth:

   • The document does not specify the number or qualifications of experts used to establish the ground truth for the test set regarding the "healthiness" of the subjects. The eligibility and exclusion criteria (e.g., "presence of any clinicant vitreal, retinal optic nerve, or choroidal disease in the study eye, including glaucoma or suspected glaucoma. This was assessed based on clinical examination and fundus photography.") imply that ophthalmologists or optometrists would have made these clinical judgments, but the specific number or their experience level is not detailed.

3. Adjudication Method for the Test Set:

   • The document does not explicitly describe an adjudication method for determining the "healthy" status of the subjects. It states that inclusion/exclusion was "assessed based on clinical examination and fundus photography" by unnamed personnel at the clinical sites. There is no mention of a consensus process, independent review, or other adjudication for the ground truth.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

   • No MRMC comparative effectiveness study was done to assess how human readers improve with AI vs. without AI assistance. The study focuses solely on establishing the normative reference database for the device's measurements. The RDB itself is a tool to be used by clinicians, but its impact on clinical decision-making or reader performance was not evaluated in this submission.

5. Standalone Performance:

   • This is a standalone performance study in the sense that the device, equipped with the new reference database, generates the normative values and compares patient data to them. It's the performance of the device's RDB calculation and display, not an AI algorithm performing diagnostic tasks without human input.

6. Type of Ground Truth Used:

   • Clinical Ground Truth: The ground truth for defining "healthy subjects" was based on extensive clinical examination, fundus photography, and adherence to strict inclusion/exclusion criteria (e.g., no known ocular disease, specific visual acuity, IOP, refraction limits). This represents a clinically defined healthy population.

7. Sample Size for the Training Set:

   • The term "training set" is not explicitly used in the context of a machine learning model, as the primary objective was to establish a statistical reference database. The entire dataset of 870 subjects (with qualified scans) was used to develop the reference database. So, the sample size for developing the reference database was 870 subjects.

8. How Ground Truth for the Training Set Was Established:

   • The "ground truth" for the subjects included in the reference database was established by defining them as "healthy subjects" through rigorous inclusion and exclusion criteria applied at 8 clinical sites across the USA. These criteria included:
     • Age 18 years and older
     • Best corrected visual acuity (BCVA) of 20/40 or better in either eye
     • IOP

Ask a specific question about this device

The P200TE is a non-contact scanning laser ophthalmoscope and optical coherence tomographer. It is intended for in-vivo viewing, digital imaging, and measurement of posterior ocular structures, including the retinal nerve fiber layer, ganglion cell complex (GCC) and optic disc under mydriatic and nonmydriatic conditions.

P200TE is indicated for producing high resolution, ultra-widefield, en face reflectance images, autofluorescence images, axial cross-sectional images, three-dimensional images, retinal layer boundary analysis, optic nerve head analysis and thickness maps.

The P200TE includes a Reference Database that enables the results of OCT segmentation analysis to be compared to reference data, including Full retinal thickness, Ganglion Cell Complex thickness, Retina Nerve Fiber Layer thickness and Optic Nerve Head metrics.

The P200TE is indicated for use as a device to aid in the detection, diagnosis, documentation and management of retinal health and diseases that manifest in the retina.

The P200TE is a desktop retinal imaging device that can perform ultra-widefield scanning laser ophthalmoscopy and optical coherence tomography. Ultra-widefield images can be captured in less than half a second. The device is intended to be used by ophthalmic and optometry health care professionals.

The P200TE delivers images in the following image modes:

• . Scanning Laser Ophthalmoscopy
• . Reflectance imaging
• . Autofluorescence imaging
• . Optical Coherence Tomography

The P200TE instrument uses red and green laser illumination for reflectance imaging, enabling it to image pathology throughout the layers of the retina, from the sensory retina and nerve fiber layer, through the retinal pigment epithelium (RPE) and down to the choroid. The image can be separated to present the distinct retinal sub-structures associated with the individual imaging wavelengths.

The P200TE instrument uses green laser illumination to excite autofluorescence (AF) emission from the naturally occurring lipofuscin in the fundus.

The P200TE instrument uses a broadband near-infrared (N-IR) super-luminescent diode (SLD) light source for optical coherence tomography allowing a depth profile of the reflectance of the fundus to be recorded. The P200TE instrument uses N-IR laser illumination for reflectance imaging simultaneously with OCT imaging. Reflectance images are used to track eye position during OCT imaging and are not available to the user.

The P200TE images the eye via two ellipsoidal mirrors arranged so that a focal point of one of the mirrors coincides with a focal point of the other mirror; a mirrored scanner is also located at this common focal point. The pupil of the subject's eye is placed at one of the other focal points. A second mirrored scanner is located at the remaining focal point; a laser or SLD reflected off this scanner is relayed onto the second scanner by the first ellipsoidal mirror and from there is reflected through the pupil and into the eve by the second ellipsoidal mirror. The second scanning element is different for OCT and SLO imaging. The energy reflected back from the retina, or emitted by fluorophores, returns through the same path to the detectors; the images are generated from the captured detector data.

P200TE OCT images are automatically segmented to identify and annotate retinal layers and structures, enabling practitioners to efficiently assess retinal structures in support of detecting, monitoring and documentation. A Reference Database enables the automatic annotation of OCT segmentation results to provide comparison to a known healthy population. Segmentation outcomes are recorded as annotations and support adjustment as deemed necessary by the clinician.

P200TE automatic seqmentation provides comprehensive retinal and optic nerve head information, including:

• . Full Retinal Thickness (FRT)
• . Ganglion Cell Complex Thickness (GCC)
• ONH Analysis
• ONH Nerve Fiber Layer Thickness

The P200TE refers to the scan head component of the system. together with touchscreen and hand controller. The device is supported by an image server which delivers patient management and image storage, as well as interfacing with the business systems and Electronic Medical Record systems.

The images are captured by the scan head under operator control and then automatically saved to the image server that uses a database structure to hold the images and patient information. For subsequent image review, a number of viewing PCs are connected remotely or via a local area network to the image server. The patient records and images are then accessible in a distributed format suited to the physical layout of the eye-care practice.

Images can be reviewed through OptosAdvance review software (K162039) either on the image server, or on individual review stations, or other compatible PACS viewers.

The provided text describes the P200TE device, a non-contact scanning laser ophthalmoscope and optical coherence tomographer. The primary purpose of the submission K233602 is to introduce a Reference Database (RDB) feature to the previously cleared P200TE, enabling the comparison of OCT segmentation analysis results to a known healthy population.

The acceptance criteria for the P200TE with the Reference Database are not explicitly stated in a quantitative table format with pass/fail metrics. Instead, the document focuses on demonstrating substantial equivalence to predicate devices (Optovue iVue and previous P200TE) by showing that the updated P200TE has the same intended use, technological characteristics, principles of operation, and similar indications. The "acceptance criteria" are implied to be centered around the clinical performance testing of the Reference Database, specifically its ability to establish reliable cut-off values for various retinal and optic nerve head measurements based on a healthy cohort, and the effective color-coding of results for interpretation.

Here's an attempt to structure the information based on the provided text, inferring acceptance criteria from the study's design and reported results:

1. Table of Acceptance Criteria (Inferred) and Reported Device Performance

• Central Foveal Thickness: Mean 247.04, 1st %ile: 195.74, 5th %ile: 212.32, 95th %ile: 283.34, 99th %ile: 302.84.
• Superior Parafoveal Thickness: Mean 315.00, 1st %ile: 278.64, 5th %ile: 290.54, 95th %ile: 339.33, 99th %ile: 351.57.
• And similar detailed statistics for Temporal, Inferior, Nasal Parafoveal, and Perifoveal thicknesses (see Table 4).
  Color-coding implemented: Values below 5% light blue, below 1% dark blue; values above 95% amber, above 99% red. |
  | Ganglion Cell Complex (GCC) Thickness Measurements:
  Ability to define statistically sound cut-off values for GCC measurements to differentiate "normal" from "borderline" or "outside normal." | GCC Thickness Measurements:
• Total Average: Mean 108.03, 1st %ile: 88.01, 5th %ile: 95.31, 95th %ile: 120.45, 99th %ile: 124.09.
• Superior Hemiretina: Mean 107.90, 1st %ile: 87.88, 5th %ile: 95.65, 95th %ile: 120.09, 99th %ile: 124.37.
• Inferior Hemiretina: Mean 108.16, 1st %ile: 87.22, 5th %ile: 94.20, 95th %ile: 121.44, 99th %ile: 125.07.
  Color-coding implemented: Values below 5% amber, below 1% red. |
  | Retinal Nerve Fiber Layer (RNFL) Measurements:
  Ability to define statistically sound cut-off values for RNFL measurements, considering optic disc size, to differentiate "normal" from "borderline" or "outside normal." | RNFL Measurements (stratified by optic disc size):
• Average RNFL, Temporal, Superior, Inferior, and Nasal Quadrant RNFL over TSNIT Circle (see Table 6 for detailed percentiles and means for Small, Medium, and Large optic disc groups).
  Color-coding implemented: Values below 5% amber, below 1% red. |
  | Optic Nerve Head (ONH) Analysis:
  Ability to define statistically sound cut-off values for ONH parameters (C/D Vertical, C/D Horizontal, C/D Area, Rim Area), considering optic disc size, to differentiate "normal" from "borderline" or "outside normal." | ONH Measurements (stratified by optic disc size):
• C/D Vertical, C/D Horizontal, C/D Area, Rim Area (see Table 7 for detailed percentiles and means for Small, Medium, and Large optic disc groups).
  Color-coding implemented: Values above 95% amber, above 99% red. |
  | Non-clinical Performance:
  Device meets defined functional and non-functional specifications.
  Compliance with relevant electrical, laser, and EMC standards.
  Software verification and validation. | Non-clinical system testing showed the system met defined specifications, with no concerns.
  Type tested in accordance with IEC 60601-1, IEC 60601-1-2, IEC 60825-1, and ANSI Z80.36-2021.
  Software V&V conducted according to QMS processes.
  All testing passed with no additional safety or performance concerns. |

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

• Test Set Sample Size: 860 eyes for full retina thickness and GCC measurements. For RNFL and ONH measurements, the sample was stratified by optic disc size:
  • Small optic disc ( 2.15 mm2): 254 eyes for RNFL, 255 eyes for ONH.
  • Overall, 879 eyes were enrolled in the study.
• Data Provenance: The document does not explicitly state the country of origin. It describes the age distribution as "skewed toward older eyes...to better match the age distribution typically found in eye clinics," implying a focus on a patient population that would likely use the device. It also mentions "good racial and ethnic diversity" (15% Hispanic, 16% Asian, 14% Black or African American, and 57% White), which suggests a US-based or diverse multinational cohort. The study was prospective in the sense that it established a reference database from healthy eyes for future comparison, effectively a one-time data collection for the database.

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

• The document does not specify the number or qualifications of experts used to establish the ground truth for the test set. The ground truth for the reference database was established by collecting data from "healthy eyes." The criteria for defining an eye as "healthy" for inclusion in the reference database are not detailed (e.g., based on clinical examination by ophthalmologists, absence of specific diseases).

4. Adjudication Method for the Test Set

• The document does not mention an adjudication method for the test set. The study describes the creation of a "reference database of healthy eyes" and the use of "non-parametric analysis to determine the 1%, 5%, 95%, and 99% cut off values." This suggests a statistical approach to defining "normalcy" from the collected data, rather than individual case adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size of Human Improvement with AI vs Without AI Assistance

• No, an MRMC comparative effectiveness study was not done. The document focuses on the technical performance and the establishment of a reference database for interpretation by human users, not on the improvement of human readers with AI assistance beyond providing a data comparison tool. The device provides "automatic segmentation" but the core of the submission (K233602) is the RDB facilitating comparison guidance, not an AI-assisted diagnostic workflow assessment involving human readers.

6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done

• Yes, in essence, the study evaluated the standalone performance of the algorithm's segmentation and RDB comparison feature by establishing the statistical distribution and cut-offs from a healthy population. The "clinical performance testing" section primarily details the generation of these normative values and their statistical validity. The device performs "automatic segmentation," and the RDB then compares these segmented measurements to a normative dataset. The study does not describe a human-in-the-loop performance study for this specific submission's purpose (adding the RDB).

7. The Type of Ground Truth Used

• The ground truth for the Reference Database was derived from clinical data collected from a cohort of "healthy eyes." The specific criteria for "healthy" status and whether this involved expert consensus, clinical outcomes, or other methods are not detailed. It is explicitly stated that the "age distribution was intentionally skewed toward older eyes... Because of this, age-related corrections were not necessary for this database," suggesting the definition of "healthy" for each age group was considered stable across the cohort.

8. The Sample Size for the Training Set

• The document refers to the data collected as the basis for the "reference database." This dataset of 879 eyes (with 860 eyes analyzed for FRT and GCC, and stratified numbers for RNFL and ONH based on disc size) effectively serves as the "training set" or "normative dataset" from which the cut-off percentiles were derived.

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

• The ground truth for the "training set" (the healthy reference database) was established through a clinical study that enrolled "healthy eyes." The data from these eyes were then used in a non-parametric statistical analysis to determine the 1st, 5th, 95th, and 99th percentile cut-off values for various OCT measurements. The document states, "The cut-off values were determined using a non-parametric analysis to reduce potential bias introduced when using standard parametric techniques due to their underlying assumptions." For RNFL and ONH measurements, the database was further partitioned into three strata based on optic disc size to account for its influence. This statistical derivation from a healthy population defines the "ground truth" for the RDB's comparative function.

Ask a specific question about this device