Results from new research published in Ophthalmology Science validated the clinical performance and patient usability of Notal Vision’s SCANLY Home OCT (optical coherence tomography) for wet age-related macular degeneration (AMD).
Hold up now, isn’t SCANLY an FDA-cleared device?
Indeed it is. Here’s a little refresh on the De Novo-cleared (in 2024) technology.
What it is: A self-operated (by the patient) and comprehensive remote patient monitoring service that captures spectral-domain (SD-)OCT images in a 10x10° area centered on the point of fixation.
- Important to note: The scan time for these images is < 1 minute per eye
Following this image capture: All images are then immediately transferred to Notal’s in-house Notal Health Cloud—a secure cloud-based platform that stores and analyzes the data—for analysis via a built-in wireless connection.
Explain this analysis process.
Following image receipt, the company’s artificial intelligence (AI)-based Notal OCT Analyzer (NOA)—a deep-learning algorithm that quantifies retinal fluid—then divides and estimates the volume of hypo-reflective spaces (HRS) on the OCT images.
- Defining HRS: These retinal biomarkers are abnormal (dark) areas of reduced light reflectivity within the retina or choroid observed on OCT scans that are indicative of various conditions like wet AMD—and also key in managing the disease.
Once estimated: Physicians review the imaging data and then establish eye-specific notification criteria (such as volume thresholds) and receive notifications via Notal’s HIPAA-compliant web portal.
So … in plain terms …
Essentially: This device is an at-home, patient-operated monitoring system that enables physicians to remotely monitor, receive daily scanning data, and treat potential disease progression.
And wasn’t there already research published on this?
Good memory! If you recall, we previously reported in February 2025 on a prospective quantitative cross-sectional AI-based study evaluating the at-home use of the SCANLY among adult wet AMD patients.
The crux of those findings: The authors determined the SCANLY system to be a “reliable tool to estimate the volume of retinal HRS from serial spectral-domain (SD)-OCT scans while monitoring at home between routine clinical visits.”
- See here for more details.
And take note: This study actually supported the FDA’s clearance of the device as the first-ever approval of an AI algorithm applied to OCT images.
Got it. Now give me a rundown on the premise of this study.
The basics: This was a prospective, 5-week longitudinal, at-home visualization and multicenter study (NCT04907409) conducted across seven retina clinical practices across the United States.
- The participants: 180 adult patients (aged 55–92 years; 57% female) diagnosed with wet AMD, with a few criteria:
- Currently receiving an anti-vascular endothelial growth (VEGF) therapy in at least one eligible eye
- Presenting with a best-corrected visual acuity (BCVA) of 20/320 or better
The design: Patients set up and self-imaged their affected eye(s) at home on a daily basis using the SCANLY OCT for 5 weeks—during which they also had preplanned in-office visits to assess image agreement for comparative OCT scans:
- At Week 1, Week 5, and on an as-needed, interim basis (dependent on if HRS was detected during a clinic visit)
- Take note: These interim visits involved a comparator in-office OCT, and all images were graded by masked reading center (RC) graders for potential HRS
And what was evaluated?
Investigators based their assessments on:
- Patients’ ability to successfully set up the SCANLY device and self-image from their homes
- Positive percent agreement (PPA) of fluid status (≥80%)
- Negative percent agreement (NPA) of fluid status (≥80%)
- Note: Both PPA and NPA were evaluated 1 month following the RC’s grading completion
Real quick: Explain these agreement types.
PPA and NPA measure how accurately two diagnostic tests agree on positive data (particularly when a reference standard is not available).
In other words: These agreements are two approaches that measure how many positives/negatives a test identifies that are in agreement with another test used on the same samples.
- And in this case: PPA and NPA are used in lieu of sensitivity (PPA) and specificity (NPA) because the comparator—in-office OCT versus SCANLY—was known to contain uncertainty.
Gotcha. Now to the findings.
To start, the device met its prespecified endpoint (≥80% agreement) by achieving a:
- PPA of 86.6%
- 95% confidence interval (CI): 80.4%–92.8%
- NPA of 86.1%
- 95% CI: 80.4%–91.8%
Any discrepancies observed (which investigators anticipated) were largely “attributed to (human) grade errors or trace amounts of HRS,” according to Notal Vision.
- In regards to this: There were just 39 pairs of images with discrepancies between SCANLY and the in-office OCT—and aside from the human grade errors, no systemic differences in grading were noted between the two comparators.
As such: The study authors noted that overall image quality and retinal HRS detection by the human readers are similar.
So how many patients were successful in their use of the SCANLY?
Of the 180, 96.1% (95% CI: 92.2%–98.4%)of eligible participants, with SCANLY averaging 5.9 scans per eye per week.
When comparing the analysis of HRS from the at-home device versus in-office OCT—“high concordance (agreement)” was noted in detecting HRS among patients.
- Further in favor of this device: The company reported that patients relying on phone support to configure SCANLY averaged only 0.20 calls per month.
Nice! How was treatment decision-making determined?
In the case of patients needing wet AMD treatment during the study, the authors emphasized that no treatment decisions were made by the study’s protocol—instead, they were made based on the investigator’s determination.
As such: An estimated 50% of the 51 interim visits that took place during the study led to a decision to treat a patient’s wet AMD.
And for the remaining 50%: The authors wrote that “treatment decision-making was not impacted by (SCANLY) findings as the extent of HRS, identified by the RC” because:
- It was not deemed treatment relevant
- The interval from the last treatment was too short to enable further treatment, or
- A treatment was already scheduled for the upcoming Week 5 in-office visit
Duly noted. And what were the final conclusions?
The study authors stated that the findings support SCANLY as a “reliable tool to visualize retinal HRS from serial (SD)-OCT scans while monitoring at home between routine clinical visits as part of the overall intended use.”