Which variants of glaucoma does the algorithm detect? 

All types of glaucoma affect the optic nerve head/optic disc and the defects of the optic nerve head can be detected using the RetinaLyze Glaucoma algorithm. This includes:

  • Signs of Open Angle/Chronic Glaucoma

  • Signs of Angle Closure/Acute Glaucoma

  • Signs of Normal-Tension Glaucoma (NTG)

  • Signs of Pigmentary Glaucoma

  • Signs of Pseudo exfoliative Glaucoma (PEX/PES/PSX/PXF)

There has even been a specific study on signs of rare congenital glaucoma. 

It looks like the algorithm detects a border a bit inside the apparent border of the ONH?

The area to be analysed is slightly adjusted inside the apparent edge, i.e. where the inner limit of the Elschnig Scleral Ring lies.

I get an image saturation warning - What do I do?

This refers to the analyzed image a saturation (of colors), which is too high for analysis. Capture a new fundus photo with a lower flash setting and then analyze that instead of the original image. Read more here.

I get the "Image could not be analyzed" error - What do I do?

This refers to an image quality, which is worse than required. The result of the analysis will be displayed as "Ungradable". Capture a new fundus photo and then analyze that instead of the original image. Read more here.

I get the "No ONH" screening result - What do I do?

If you are using RetinaLyze Glaucoma, the Optic Nerve Head (ONH) needs to be present in the image. Capture a new fundus photo (making sure that the ONH is present in the image) and then analyze that instead of the original image. Read more here.

How does the RetinaLyze Glaucoma algorithm work?

RetinaLyze Glaucoma assesses the level of hemoglobin in the ONH and uses the measured values to evaluate the signs of Glaucoma. Read more here.

How has the performance of the algorithm been validated? 

The RetinaLyze Glaucoma algorithm has been clinically validated through several recent studies for several types of glaucoma. Please find the studies on our webpage.

How does the RetinaLyze Glaucoma algorithm compare with other methods of Glaucoma screening?

The results have not only been compared with perimetry, but in several studies with morphological tests (HRT, GDx and OCT-Cirrus and OCT-Spectralis). Not only have advanced glaucomas been analyzed, but also early glaucomas, patients with suspected glaucoma and simple ocular hypertension. A doctoral thesis has been carried out, especially on early glaucoma and ocular hypertension.
Please find the available public studies here:

Which factors can affect the use of the algorithm and the validity of the result? 

Peripapillary atophy/Myopic rarification

The automatic nerve edge delimitation system attempts to rule out peri-papillary atrophy, but the user must verify that it has been done correctly as well as correcting the edge delimitation if it is included erroneously.

Peripapillary myopic atrophy areas should be outside the papillary limit, as should any other type of atrophy or pigmentation.

Myelinated retinal nerve fiber layer

These are rare cases, but if the fibers are myelinized in front of the cribrosa plate the case cannot be analyzed. Other unusual malformations of the nerve, such as colobomas, should not be analyzed either.

Important guidelines about image quality

It is very important that the image is not very luminous. If there are saturated pixels, the user will be prompted to capture an image with a lower flash setting. The illumination should be uniform and without bright spots/reflexes.

Patients with slightly blurred images from initial cataracts can be analyzed.

If the image is so poor that the vessels cannot be clearly identified, the image cannot be analyzed or the results are unreliable.

Any screening or diagnostic system has a limited degree of accuracy, sensitivity and specificity. RetinaLyze Glaucoma is quite resistant to moderate defects in focus or sharpness, but its results will be optimal by selecting the best possible image, just as with manual grading. Some physiological variants such as macro-papilae can more easily produce false positives and as is the case of micro-papilates and oblique papillae with false negatives.

See the in-depth guidelines for image quality here:

What type of fundus images / fundus cameras can be used with the algorithm?

Read more here.

Who developed the algorithm?

The algorithm was developed by INSOFT SL. These are the main doctors and researchers that have been involved in the development:

  • Prof. Manuel Gonzalez de la Rosa (MD,PhD). Department of Ophthalmology of the Universidad de la Laguna.

  • Dr. Marta Gonzalez-Hernandez (OD, PhD). Department of Ophthalmology of the Hospital Universitario de Canarias.

  • Prof. Jose Sigut Saavedra (PhD). Department of Systems Engineering and Automation of the Universidad de La Laguna. 

  • Prof. Silvia Alayon Miranda (PhD). Department of Systems Engineering and Automation of the Universidad de La Laguna.

  • Dr. Carmen Mendez Hernandez (MD,PhD). Departament de Glaucoma, Hospital Clinico San Carlos, Universidad Complutense de Madrid.

Did this answer your question?