Search Results

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