Diabetic retinopathy Eye Diseases Diabetic retinopathy
Conditions
Interventions
In this study, individuals with diabetes will be assessed for diabetic retinopathy with the smartphone-based handheld fundus camera EyeFundusScope (Fraunhofer AICOS, Portugal) operated by non-ophthalm
Sponsors
Fraunhofer Portugal Research
Eligibility
Sex/Gender
All
Inclusion criteria
Inclusion criteria: 1. Diabetes mellitus diagnosis 2. 18-90 years old
Exclusion criteria
Exclusion criteria: 1. Blindness 2. Strabismus 3. Retinal detachment 4. Blepharospasm 5. No autonomy to remain seated 6. Sensitivity to light due to medication, recent photodynamic therapy, or other reason 7. Pregnant or breastfeeding women
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Diagnostic accuracy of EyeFundusScope for diabetic retinopathy screening, measured using sensitivity, specificity, predictive values and likelihood ratios at the time of retinal image acquisition | — |
Secondary
| Measure | Time frame |
|---|---|
| 1. Image quality measured in four levels and classified by ophthalmologists at the time of retinal image acquisition 2. Technological functioning measured using the number of eyes in which it was not possible to acquire images, and the number of technical failures of both the EyeFundusScope and the information system, at the time of retinal image acquisition 3. Usability of the EyeFundusScope measured using the proportion of healthcare professionals who report ease of use, the score of the System Usability Scale, qualitative analyses of the observation and the mean time required to perform retinal image acquisition with the EyeFundusScope, at the time of retinal image acquisition 4. Acceptance of the EyeFundusScope measured using questionnaires and qualitative analysis of interviews and observations at the time of retinal image acquisition 5. The association of patients’ clinical characteristics with the accuracy and the quality of the images of the EyeFundusScope, measured using odds ratios at the time of retinal image acquisition 6. Inter and intra-operator agreement and reliability of the EyeFundusScope measured using kappa statistics and overall observed agreement at the time of retinal image acquisition 7. Ground truth optimization for the training of the algorithms before the calculation of measures listed above | — |
Countries
Portugal
Outcome results
None listed