Ultra-widefield diabetic retinopathy screening

  • Research type

    Research Study

  • Full title

    Ultra-widefield retinal imaging for diabetic retinopathy screening

  • IRAS ID

    156927

  • Contact name

    Nigel Davies

  • Contact email

    nigel.davies@chelwest.nhs.uk

  • Duration of Study in the UK

    0 years, 11 months, 28 days

  • Research summary

    Across the UK, patients known to have diabetes are invited at least annually to attend diabetic retinopathy screening. A fundus camera is used to produce two overlapping images of the posterior of the retina. Following a nationally agreed protocol, patients in whom referable diabetic retinopathy is detected in their images are referred to the hospital eye service.

    An alternative imaging device for diabetic retinopathy screening is potentially a "Daytona" (Optos plc, Dunfermline) ultra-widefield scanning laser ophthalmoscope (UWF-SLO). This would have the advantage of obtaining a larger retinal field being imaged in a single photograph for each eye. Optos plc received CE mark clearance for the Daytona UWF-SLO in November 2011.

    The aim of this study will be to evaluate the effectiveness of a computer program designed to assess diabetic retinopathy in images captured by the Daytona UWF-SLO.

    Patients attending screening for diabetic retinopathy and who consent to participate will be imaged by the Daytona device at the start of their appointment. They will then continue to their routine screening photography.

    Approximately 600 patients (including 100 with referable diabetic retinopathy) attending Kensington Chelsea & Westminster Diabetic Eye Screening Service, a service contracted to 1st Retinal Screen Ltd, will be imaged in this way. 1st Retinal Screen staff will operate the Daytona device, undertake the routine appointment and grade the images.

    Anonymised images will be processed by Optos automated diabetic retinopathy grading software. The result will be compared to the manual grading allowing the sensitivity (true positive rate) and specificity (true negative rate) of the software for the detection for diabetic retinopathy to be estimated. The automated grading result will not be linked to the patient identifiable data and will not influence patient care.

    1st Retinal Screen Ltd will receive funding from Optos plc for this study.

  • REC name

    East Midlands - Nottingham 1 Research Ethics Committee

  • REC reference

    14/EM/1196

  • Date of REC Opinion

    29 Oct 2014

  • REC opinion

    Further Information Favourable Opinion