Applied Conversion Equation for Optical Coherence Tomography

Optical Coherence Tomography (OCT) is a type of scan of the back of the eye. It is non-invasive and provides vital information about the health of the eye. It is commonly available in optometrists and hospital eye units. OCT scans are a mainstay for deciding if treatment is needed for Diabetic Macular Oedema (DMO). However, there are many different companies that make these OCT machines and there is variability between the measurements. The amount of variability is unknown. Clinically, this means that patients may be under or over treated unknowingly. From a research perspective, patients are being excluded from studies because measurements from OCT different machines cannot be compared accurately.

There is a need for a conversion equation so measurements between OCT machines can be compared. This would be of benefit for patients, for clinicians making treatment decisions and for researchers

This project addresses the knowledge gap of not knowing the amount of variability in measurements between OCT machines. By knowing this, diseases such as DMO may possibly be detected and treated earlier. It will also make it easier to track progression of disease if a patient moves to a different treatment centre or OCT machine. Crucially, it will help research as patients will not be excluded if participants were scanned on different OCT machines.

The key objective of this study is to collect OCT images for 338 patients (675 eyes) with DMO at the same time and place on the two different OCT machines (Heidelberg Spectralis and Topcon Triton). This will comprise of 450 eyes (225 patients) for the training of the model and a further 225 eyes to validate the model.

Firstly, this will help the investigator to quantify the variability in the measurements between the machines. Secondly it will help the investigator to build a conversion equation for measurements taken on Heidelberg Spectralis and Topcon Triton OCT machines. The final stage is to validate the conversion equation to ensure that the model is suitable to be used.

There may be patients whose disease is under or overestimated due to variability in measurements on the different OCT machines. This project would mean that patients with DMO are diagnosed more accurately and treated is a timelier way which would positively affect their visual outcome.

This project will help further sight-saving research. It would mean that patients are not excluded from studies purely because their OCT images are not comparable as they were taken on different machines. Having more eligible patients for research studies would improve the quality of research into existing and new treatments and better inform treatment guidelines

Inclusion criteria

• Patients aged 18 years or above
• Patients with a diagnosis of Type 1 or Type 2 Diabetes Mellitus
• Patients with a known diagnosis of DMO in at least one eye
• Patients being seen in DMO clinics, Digital Surveillance Clinic or Virtual Diabetic Retinopathy clinics

Exclusion criteria

• Patients under 18 years old
• People unwilling or unable to give informed consent
• Patients in whom it is not possible to take OCT scans, for instance those with:

• Dense cataracts
• Corneal Scarring
• Severe physical disabilities preventing them from adequate positioning.
• People with other ocular pathologies known to impact retinal thickness, including:

• Age related Macular Degeneration (AMD)
• Branch or central retinal vein occlusion (BRVO/CRVO)
• Macular Hole
• Glaucoma
• High Myopia with a spherical equivalent > -5.00 diopters
• Epiretinal membrane
• Eyes with at least one scan containing segmentation errors
• Eyes with at least one scan that is ungradable