Comparison of 21 artificial intelligence algorithms in automated diabetic retinopathy screening using handheld fundus camera.

BACKGROUND: Diabetic retinopathy (DR) is a common complication of diabetes and may lead to irreversible visual loss. Efficient screening and improved treatment of both diabetes and DR have amended visual prognosis for DR. The number of patients with diabetes is increasing and telemedicine, mobile handheld devices and automated solutions may alleviate the burden for healthcare. We compared the performance of 21 artificial intelligence (AI) algorithms for referable DR screening in datasets taken by handheld Optomed Aurora fundus camera in a real-world setting. PATIENTS AND METHODS: Prospective study of 156 patients (312 eyes) attending DR screening and follow-up. Both papilla- and macula-centred 50° fundus images were taken from each eye. DR was graded by experienced ophthalmologists and 21 AI algorithms. RESULTS: Most eyes, 183 out of 312 (58.7%), had no DR and mild NPDR was noted in 21 (6.7%) of the eyes. Moderate NPDR was detected in 66 (21.2%) of the eyes, severe NPDR in 1 (0.3%), and PDR in 41 (13.1%) composing a group of 34.6% of eyes with referable DR. The AI algorithms achieved a mean agreement of 79.4% for referable DR, but the results varied from 49.4% to 92.3%. The mean sensitivity for referable DR was 77.5% (95% CI 69.1-85.8) and specificity 80.6% (95% CI 72.1-89.2). The rate for images ungradable by AI varied from 0% to 28.2% (mean 1.9%). Nineteen out of 21 (90.5%) AI algorithms resulted in grading for DR at least in 98% of the images. CONCLUSIONS: Fundus images captured with Optomed Aurora were suitable for DR screening. The performance of the AI algorithms varied considerably emphasizing the need for external validation of screening algorithms in real-world settings before their clinical application.

What is already known on this topic? Diabetic retinopathy (DR) is a common complication of diabetes. Efficient screening and timely treatment are important to avoid the development of sight-threatening DR. The increasing number of patients with diabetes and DR poses a challenge for healthcare.What this study adds? Telemedicine, mobile handheld devices and artificial intelligence (AI)-based automated algorithms are likely to alleviate the burden by improving efficacy of DR screening programs. Reliable algorithms of high quality exist despite the variability between the solutions.How this study might affect research, practice or policy? AI algorithms improve the efficacy of screening and might be implemented to clinical use after thorough validation in a real-life setting.

Outcomes of 35-year duration of type 1 diabetes and proliferative diabetic retinopathy on functional vision and quality of life: Benefits of good glycemic control.

AIMS: To evaluate clinical outcomes, functional vision and quality of life (QoL) after 35-year duration of type 1 diabetes (T1D) and proliferative diabetic retinopathy (PDR). METHODS: A population-based cohort study of T1D. Data from laboratory tests, ophthalmic examinations, multifunctional vision-test, and 15D-QoL measurements were analysed. RESULTS: 35 % of the original cohort (n = 216) had PDR, and 48 % of them were re-evaluated. They were 41 ± 3 [34-46] years old and 62 % were males. The duration of T1D was 35 ± 4 [29-41] years. 76 % had transdermal glucose monitoring. HbA1c had decreased from 80.1 mmol/mol to 63.6 mmol/mol (p < 0.001). Visual acuity was 73-77 ETDRS-letters. Two patients had visual impairment. Visual field sensitivities were lower in PDR vs. healthy controls (23.2 ± 3.9 dB vs. 26.9 ± 1.0 dB, and 14.9 ± 5.6 dB vs. 21.0 ± 2.0 dB, respectively, p < 0.001). Contrast sensitivity was similar, but the reaction time was longer in the PDR group (490.5 ms vs. 462.8 ms, p = 0.004). QoL-parameters concerning sleeping, usual activities, discomfort and symptoms, and sexual activity had decreased, but improved for mobility and distress. CONCLUSIONS: Long-term visual prognosis and QoL remained good despite the declined functional vision caused by PDR.

Visual outcomes of observation, macular laser and anti-VEGF in diabetic macular edema in type 1 diabetes: a real-world study.

BACKGROUND: The treatment for diabetic macular edema (DME) has revolutionized during the last 15 years after the introduction of intravitreal anti-VEGF agents. The aim of the current study is to evaluate the real-world visual outcomes of diabetic macular edema (DME) treatment in patients with type 1 diabetes (T1D) in long-term follow-up. METHODS: A real-world, descriptive, population-based cohort and follow-up of all patients with T1D and DME in 2006-2020 in 34 communities of the Northern Ostrobothnia Hospital District. The main outcome measures included age, gender, duration of T1D at the onset of DME, stage of retinopathy, treatment of DME (observation, laser, intravitreal treatments, combination), and visual outcomes. RESULTS: A total of 304 eyes of 206 T1D patients with DME were included. 75% (n=155) had non-proliferative diabetic retinopathy during the onset of DME. 15% of the cases were observed, 33% had macular laser, 41% intravitreal anti-VEGF and 12% combination of laser and intravitreal injections. Patients in anti-VEGF and in combination groups gained 4.9 and 5.5 ETDRS letters after the initial DME episode (p<0.001 and p<0.001), and the long-term visual improvements were 4.1 and 5.1 ETDRS letters (p<0.001 and p<0.001), respectively. In observation and laser groups the initial gain of 0.1 (p>0.90) and loss of 0.4 ETDRS letter (p=0.61), respectively, was noted. After the follow-up, a 3.7 ETDRS letter decrease was documented in the observation group (p>0.90) and a 1.1 (p=0.14) ETDRS letter decline in the laser group of patients. At the beginning of treatment, eyes subjected to anti-VEGF alone or in combination with laser had lower visual acuity compared to eyes subjected to observation or macular laser. The average of a 6.1±4.8 anti-VEGF injections were needed to dry DME. Visual impairment due to DME decreased from 2.4% to 1.0% during the 15-year period. CONCLUSIONS: Anti-VEGF alone or in combination with macular laser seems to be beneficial in terms of visual outcomes and treatment stability in T1D patients with central DME. Moreover, satisfying long-term visual outcomes were achieved with anti-VEGF treatment in a real-world setting.

Handheld fundus camera performance, image quality and outcomes of diabetic retinopathy grading in a pilot screening study.

PURPOSE: To compare the performance and image quality of the handheld fundus camera to standard table-top fundus cameras in diabetic retinopathy (DR) screening. The reliability and diagnostic accuracy of DR grading performed by an ophthalmologist and a photographer reader were evaluated. MATERIALS AND METHODS: 157 patients with diabetes, attending screening or follow-up of DR, were evaluated by fundus photographs taken in mydriasis by Optomed Aurora and Canon or Zeiss Visucam fundus cameras. The image quality and the severity of DR were evaluated independently by an ophthalmologist and experienced photographer. The sensitivity, specificity and reliability of the assessments were determined. RESULTS: 1884 fundus images from 314 eyes were analysed. In 53% of all eyes, DR was not present. 10% had mild non-proliferative diabetic retinopathy (NPDR), 16% moderate NPDR, 6% severe NPDR and 16% proliferative diabetic retinopathy (PDR). The DR grading outcomes by Aurora highly equalled to those of Canon or Zeiss (κ = 0.93, 95% CI 0.91 to 0.94), and there was almost perfect agreement in grading between the ophthalmologist and photographer (κ = 0.96, 95% CI 0.95 to 0.97). The image quality of Aurora was sufficient for reliable assessment according to both graders in 84-88% of the cases. CONCLUSION: The Optomed Aurora fundus camera seems appropriate for DR screening. The sufficient image quality and high diagnostic accuracy for DR grading are supportive for a less expensive and easily transportable screening system for DR. Immediate image grading carried out by a photographer would further improve and speed up the screening process in all settings.

Intravitreal bevacizumab improves the clearance of vitreous haemorrhage and visual outcomes in patients with proliferative diabetic retinopathy.

OBJECTIVE: To evaluate the occurrence of vitreous haemorrhage (VH) secondary to proliferative diabetic retinopathy (PDR) and the efficacy of intravitreal bevacizumab (IVB) for VH in 5-year real-life data. METHODS AND ANALYSIS: 850 adult patients with type 1 (T1D) or type 2 diabetes (T2D) with PDR were screened for VH. The effect of IVB was evaluated by the clearage of VH and the change in best corrected visual acuity (BCVA). The rates of VHs, reinjections, macular oedema, complications, additional treatments and outcomes of spontaneous resorption, panretinal photocoagulation or pars plana vitrectomy (PPV) for VH were also investigated. RESULTS: VH occurred in 16% of patients with T1D and 9% of patients with T2D with PDR. 336 VHs in 140 eyes of 103 patients were documented. VH was cleared in 92% of cases in less than 3 months by the initial IVB. IVB was superior to other treatment methods in shortening the time for clearance of VH (Kaplan-Meier, p<0.0001). The average rate of IVB reinjections was 1.7±1.1 and the reinjection interval was 7.2±3.9 weeks. BCVA increased 0.73±0.04 logarithm of the minimum angle of resolution units after IVB (generalised estimating equations, p=0.0004). In 5 years, the patients had 2.2±2.7 recurrence of VHs. A simultaneous 72% decrease in the rate of PPVs was documented (p<0.0001). CONCLUSION: VH occurs mostly in patients with T1D. The therapeutic effect of IVB for VH was significant and led to improved clearance of VH and visual outcome. Moreover, IVB prevented persistent and recurrent VHs and decreased the need for costly PPV.