AOdetect™

Add investigational value to your rtx1™ AO Retinal Camera

AOdetect* is a semi-automated application for the segmentation and analysis of ultrahigh resolution images acquired with the rtx1 AO retinal camera.

It computes local descriptors of the distribution of parafoveal cone cells and the structure of retinal arterioles. These metrics provide candidate AO biomarkers for clinical investigations.

With AOdetect, you are in complete control as all automated segmentation can be manually corrected in just a few clicks.

*AOdetect is for research use only

2-in-1 analysis application

AOdetect* is a semi-automated application for the segmentation and analysis of ultrahigh resolution images acquired with the rtx1 AO retinal camera.

It computes local descriptors of:

  • the structure of retinal arterioles,
  • the distribution of parafoveal cone cells.

These metrics provide candidate AO biomarkers for clinical investigations.

With AOdetect, you are in complete control as all automated segmentation can be manually corrected in just a few clicks.

*AOdetect is for research use only

Quick 2-step analysis

Define a region of interest
Review and save the results

Simple workflow

Capture images with the rtx1 software
Open AOdetect with a single click

Accurate and reliable

Good reproducibility
Proven by peer-reviewed publications

AOdetect Mosaic

For images of parafoveal cone cells

Automated mosaic segmentation and Voronoi analysis

Manual correction : simple clicks to add and remove cells while monitoring the Voronoi segmentation

Mosaic descriptors

  • Voronoi cell density
  • Inter-cell spacing
  • Regularity index
  • Dispersion index

Color-coded Voronoi diagrams

  • Regularity
  • Density
  • Spacing

AOdetect Mosaic

For images of parafoveal cone cells

Automated mosaic segmentation and Voronoi analysis

Manual correction : simple clicks to add and remove cells while monitoring the Voronoi segmentation

Mosaic descriptors

  • Voronoi cell density
  • Inter-cell spacing
  • Regularity index
  • Dispersion index

Color-coded Voronoi diagrams

  • Regularity
  • Density
  • Spacing

AOdetect Walls

For images of retinal arterioles

Automated wall segmentation and thickness computation

Manual correction : click-and-drag to displace the wall borders while monitoring their position in the gradient profile

Walls descriptors

  • Internal and external diameter
  • Wall thickness
  • Wall Cross Section Area (WSCA)
  • Wall to Lumen Ratio (WLR)

rtx1 + AOdetect : cellular follow-up made easy

Thanks to its high-precision alignment algorithm, the rtx1 delivers follow-up images that are perfectly aligned with the baseline image.

AOdetect enables analyzing the exact same region of interest in baseline and follow-up images.

This is how you can easily monitor a specific group of cells or a chosen vascular section over time.

rtx1 + AOdetect : cellular follow-up made easy

Thanks to its high-precision alignment algorithm, the rtx1 delivers follow-up images that are perfectly aligned with the baseline image.

AOdetect enables analyzing the exact same region of interest in baseline and follow-up images.

This is how you can easily monitor a specific group of cells or a chosen vascular section over time.

Enable clinical investigations

AOdetect has empowered clinical researchers to investigate a wide array of candidate biomarkers for assessing the micro-structural integrity of the retina.

A number of peer-reviewed publications describe various new insights into the retina in various pathologies, enabled by the rtx1’s unrivaled resolution and by AOdetect segmentation possibilities.

The quantitative assessment of photoreceptor survival or loss, based on analysis of adaptive optics retinal images, was valuable to monitor disease progression at a cellular level.

Ziccardi et al.American Journal of Ophthalmology 160, 301-312.e6Read more in the article

In arterial hypertension, WLR is a robust, dimensionless parameter that can be measured on large cohorts of  nondilated patients.

Paques et al.Progress in Retinal and Eye Research. DOI: 10.1016/j.preteyeres.2018.07.001Read more in the article

The AO images revealed that the parafoveal cone density was reduced even with good visual acuity at stages 1 and 2 of CACD.

Gocho et al.Ophthalmic Surg Lasers Imaging Retina 47, 1115- 1126

High-resolution retinal imaging analysis, such as that by AO, would be helpful in identifying patients with CEP250- associated disease.

Kubota et al.Ophthalmic Genetics, 2018Read more in the article

It could represent, in the near future, an evaluation to be performed in all hypertensive patients.

Agabiti-Rosei et al.Journal of Hypertension 35, 914–921

The difference of two measurements [] was less than 6% confirming the good repeatability of the rtx1. We also tested the reproducibility of the method [] We observed an averaged difference of less than 4% between [three] investigators.

Legras et al.PLOS ONE 13, e0191141Read more in the article

In early MacTel 2 eyes with near-normal vision, AO is able to show cone density loss in the macula, without loss of the ellipsoid zone on OCT.

Jacob et al.Jacob et al. Retina 36, 545–551Read more in the article

Links & Downloads

Product brochure
AOdetect biomarker software


AOdetect software is not part of the rtx1 product. AOdetect software has not received FDA clearance for clinical use. It is for research use only.

Computer requirements
OSWindows XP-SP3, Vista-SP2, 7-SP1, 8
RAM2 Gbyte or more
CPUIntel i3 or higher
Export formats
Analysis resultsXLS, CSV spreadsheets
SegmentationJPEG
Reproducibility information
AOdetect Mosaic4% for cone density1
AOdetect Walls3.2% for WLR2
1.3% for lumen diameter2

1 Legras et al. Distribution of Cone Density, Spacing and Arrangement in Adult Healthy Retinas with Adaptive Optics Flood Illumination. PLOS ONE 13, e0191141.
2 Rosenbaum et al. Imaging of retinal arteries with adaptative optics, feasibility and reproducibility. Ann Cardiol Angeiol (Paris) 2013; 62:184-188.

Contact us for more information