A Diabetic Eye Examination Explained: Dilation, OCT, Photography, and Angiography

Blog

A Diabetic Eye Examination Explained: Dilation, OCT, Photography, and Angiography

August 9, 2026

Key Takeaways

  • A diabetic eye examination does more than look for retinopathy. It assesses vision, refraction, lens, pressure, optic nerve, macula, peripheral retina and ocular-surface health.
  • Dilation improves the view of peripheral lesions and neovascularization. Retinal photographs document severity, but their field and image quality determine what they can exclude.
  • OCT provides cross-sectional macular anatomy and is central to diagnosing and monitoring diabetic macular edema. A normal macular OCT does not rule out peripheral proliferative disease.
  • Fluorescein angiography shows vascular leakage and nonperfusion; OCT angiography maps blood-flow signals without dye but cannot display leakage directly. The tests are complementary.
  • Ultra-widefield imaging expands peripheral documentation, while fundus autofluorescence and ultrasound answer selected questions rather than routine staging needs.
  • Screening photographs and autonomous AI improve access but are not identical to a comprehensive dilated examination. Positive or ungradable results require follow-through.
  • Test results should change management: stage each eye, determine DME and neovascular status, set the next interval and identify treatment thresholds.
  • Netra Restoration Therapy (NRT) may use the completed ocular diagnosis to support adjunctive care, but it cannot substitute for retinal imaging, interpret emergency symptoms remotely or prove disease stability.

Patients sometimes leave a diabetic eye visit with a stack of images but no clear explanation. One scan shows colored thickness maps, another shows branching vessels and the examination includes bright lights after dilation. Each tool answers a different question.

The essential outcome is not the number of machines used. It is whether the clinician can state the stage in each eye, whether the fovea is swollen, whether abnormal new vessels or traction are present and when the patient must return.

This guide explains the full examination without duplicating Netra’s diabetic retinopathy condition page.

Preparing for the visit

Bring the date and type of diabetes diagnosis, recent A1C and blood-pressure information, kidney status, pregnancy plans, medication and supplement list and prior eye records. Include anti-VEGF injections, laser, cataract surgery, vitrectomy and glaucoma treatment.

Tell the clinic about sudden floaters, distortion, one-eye loss, pain or neurologic symptoms when scheduling; those may require urgent triage rather than a routine appointment.

Dilation can cause light sensitivity and near blur for several hours. Bring sunglasses and arrange transportation if prior dilation impaired driving. Follow normal food and medication instructions unless the clinic or diabetes team gives a specific plan. Long visits should account for insulin, meals and hypoglycemia risk.

History and symptom review

The clinician asks about blur, distortion, contrast, color, floaters, flashes, haze, field loss, double vision, glare and pain. Absence of symptoms is expected in early retinopathy and does not reduce the need for examination.

Glucose trajectory matters. Rapid change can produce lens refraction shifts and, in high-risk eyes, early retinopathy worsening. Pregnancy, kidney disease, blood pressure and follow-up barriers modify risk.

The history also identifies cataract, dry eye, glaucoma and cranial-nerve problems that can cause vision symptoms unrelated to retinal stage.

Visual acuity

Each eye is tested separately with current correction. Best-corrected acuity estimates central visual function but does not measure peripheral proliferation, contrast or macular ischemia fully.

An eye with PDR can read 20/20 if the fovea is unaffected. An eye with mild NPDR can see poorly from cataract or amblyopia. Acuity is an outcome, not a retinopathy stage.

ETDRS charts used in trials provide standardized letter scoring. Routine clinics may use Snellen or electronic charts. Comparing values requires attention to method and refraction.

Refraction

Refraction determines whether lenses improve blur. Large glucose changes can make the endpoint unstable, so a new prescription may be deferred until relative metabolic stability.

Failure to improve with refraction prompts examination for cataract, DME, ischemia, optic-nerve disease and other causes. A dry ocular surface can also create variable measurements.

Pupils and eye movements

Pupil responses can reveal optic-nerve or severe asymmetric retinal dysfunction. An afferent pupillary defect is not typical of equal mild NPDR and may suggest marked asymmetry, retinal detachment, ischemia or optic neuropathy.

Eye movements are checked when double vision is present. Diabetes-associated cranial-nerve palsy is possible, but aneurysm, stroke, thyroid disease and other causes require appropriate evaluation.

Eye pressure and the anterior segment

Tonometry measures intraocular pressure. Diabetes is associated with open-angle glaucoma risk, and steroid implants or drops used for DME can raise pressure.

Slit-lamp examination assesses cornea, anterior chamber, iris and lens. It identifies dry eye, neurotrophic defects, cataract, inflammation and rubeosis—abnormal iris vessels caused by severe retinal ischemia.

Gonioscopy examines the drainage angle when neovascularization or glaucoma is suspected. New angle vessels can precede painful neovascular glaucoma and require retinal and pressure treatment.

Why dilation is used

Dilating drops enlarge the pupil, allowing a wider retinal view. The clinician examines optic nerve, macula, vessels and peripheral retina with lenses and indirect ophthalmoscopy.

Dilation helps detect hemorrhages, venous beading, IRMA, new vessels, laser scars, tears and detachments. It also improves photographic gradability in small pupils.

Near blur and light sensitivity are expected. Acute angle closure after dilation is rare; anatomy and symptoms guide risk. Severe pain, redness, halos, headache or nausea after dilation requires urgent care.

Direct versus indirect examination

Slit-lamp biomicroscopy with a noncontact or contact lens gives stereoscopic detail of the posterior pole. Indirect ophthalmoscopy provides a wide inverted view of the peripheral retina and can use scleral depression in selected circumstances.

No static photograph perfectly replaces dynamic stereoscopic examination in every patient. Conversely, photography provides documentation that memory and narrative drawing cannot match. The methods complement one another.

Color fundus photography

Color photographs record microaneurysms, hemorrhages, exudates, cotton-wool spots, venous changes, IRMA, neovascularization and laser scars.

The ETDRS seven-field protocol covers standardized retinal regions and underpins research grading. Modern clinical cameras use fewer fields or widefield capture. Image quality, pupil size, cataract, focus and pigmentation affect interpretation.

Serial photographs help determine progression or regression. Comparing images requires similar field and quality; a different camera may make lesions appear changed.

Ultra-widefield imaging

Ultra-widefield systems capture up to approximately 200 degrees in a single image, documenting peripheral lesions and nonperfusion beyond standard fields. Peripheral-predominant lesions may carry prognostic information.

Eyelashes, lids, distortion and peripheral color variation can create artifacts. The far periphery may still need indirect examination. A wide image is not automatically a complete clinical interpretation.

Optical coherence tomography

OCT uses reflected light to create micrometer-scale cross-sections of the retina. It is noncontact, rapid and central to DME care.

OCT shows:

  • central and regional retinal thickness;
  • intraretinal cystoid spaces;
  • subretinal fluid;
  • hyperreflective foci and hard-exudate correlates;
  • vitreomacular adhesion or traction;
  • epiretinal membrane;
  • photoreceptor and outer-retinal integrity; and
  • disorganization of inner retinal layers.

Automated segmentation can fail with edema, traction or poor signal. Clinicians inspect the B-scans rather than relying only on a green or red thickness map.

Center-involved versus noncenter DME

The foveal central subfield determines whether edema is center-involved on OCT. Visual acuity, not thickness alone, influences treatment.

Protocol V showed that many eyes with center-involved DME and 20/25 or better acuity can begin with structured observation and receive aflibercept if vision declines. Vision-reducing center-involved DME commonly receives prompt anti-VEGF.

Noncenter edema may be observed or treated with focal/grid laser in selected cases. The plan depends on leakage, progression, exudates and follow-up.

OCT limitations

OCT scans a defined area and cannot rule out peripheral new vessels. A dry macula does not mean PDR is absent. Thickness can normalize while macular ischemia or photoreceptor damage limits vision.

Signal is reduced by cataract, vitreous hemorrhage and poor fixation. Device normative databases and segmentation differ, so numbers from different machines are not directly interchangeable.

Fluorescein angiography

Fluorescein dye is injected into a peripheral vein, and sequential retinal photographs track its circulation. The test shows microaneurysms, leakage, capillary nonperfusion, neovascularization and laser-treatment patterns.

It is useful when the source of leakage is unclear, peripheral ischemia needs mapping, neovascularization is difficult to distinguish from IRMA or treatment planning requires vascular detail.

Yellow skin or urine discoloration is temporary. Nausea is relatively common; hives and serious allergic reactions are uncommon but possible. Pregnancy, prior reaction and medical history are reviewed. Fluorescein is not the same as iodinated CT contrast.

Angiography interpretation

Hyperfluorescence can result from leakage, staining, pooling, window defects or abnormal vessels. Hypofluorescence can reflect blocked fluorescence or absent perfusion.

Leakage shows where dye escapes but does not perfectly quantify visual potential. Nonperfusion maps ischemia, yet artifact and image timing matter. The test is interpreted with color photographs and OCT.

OCT angiography

OCTA compares repeated OCT scans to detect motion of blood cells and map superficial, deep and choriocapillaris flow signals. It can visualize foveal avascular-zone change, capillary density and neovascular networks without injection.

OCTA does not show leakage directly. Slow flow may appear absent, while motion, projection and segmentation create artifacts. Small scan areas can miss peripheral disease; widefield montages are developing.

OCTA is valuable for selected monitoring but does not universally replace fluorescein angiography.

Fundus autofluorescence

Autofluorescence images naturally fluorescent molecules, primarily lipofuscin-related signals in the retinal pigment epithelium. It is central in many retinal dystrophies and atrophy disorders, but not a routine primary diabetic-retinopathy staging test.

It may document other macular pathology or selected treatment effects. Ordering it should answer a question beyond using another imaging mode.

B-scan ultrasound

Dense vitreous hemorrhage or cataract can block the retinal view. B-scan ultrasound uses sound to detect retinal detachment, traction, vitreous membranes and masses.

It cannot grade small microaneurysms or DME. A negative ultrasound for detachment does not establish that proliferative disease is controlled; retinal treatment remains necessary when the view clears or surgery proceeds.

Visual fields and optic-nerve OCT

These tests evaluate glaucoma and optic neuropathy rather than retinopathy stage. PRP can affect peripheral fields, and advanced retinal disease can complicate interpretation.

Diabetes does not protect a patient from glaucoma. A comprehensive plan separates field loss from retinal blur and avoids attributing every abnormality to one disease.

Screening photography versus comprehensive examination

Nonmydriatic photography in primary care or pharmacy settings can identify referable disease and improve access. Autonomous AI can provide a point-of-care result.

Screening may not assess pressure, anterior segment, peripheral retina beyond the field, macular thickness without OCT or other eye disease. Positive and ungradable results require referral.

The correct comparison is not “camera or eye doctor forever.” It is a connected pathway: accessible detection followed by appropriate comprehensive care.

Test order and workflow

Clinics often measure acuity and pressure before dilation, then perform photographs and OCT. Angiography follows when indicated. If cataract surgery is planned, corneal measurements and surface optimization are added.

The sequence can be adapted for hypoglycemia risk, pregnancy, mobility and urgent symptoms. A long imaging visit should include breaks and access to food according to the patient’s diabetes plan.

How results become a report

A useful diabetic retinal report states:

  • best-corrected acuity in each eye;
  • retinopathy stage in each eye;
  • DME presence and center involvement;
  • neovascularization, hemorrhage or traction;
  • prior laser or injection status;
  • other limiting disease;
  • treatment recommendation; and
  • exact follow-up interval and urgent symptoms.

“Stable” should identify what was compared and over what period. Stable OCT does not necessarily mean stable peripheral PDR.

Imaging artifacts and false certainty

Blink, motion, defocus, small pupils, cataract, hemorrhage and segmentation errors degrade images. AI and automated maps can be confidently wrong.

Clinicians inspect source images and repeat scans when quality changes the decision. Patients should not be sold treatment from a color-coded map without explanation of signal quality and clinical correlation.

Examination during active anti-VEGF treatment

When DME or PDR is treated with injections, each visit asks whether to treat again, extend the interval or change strategy. Acuity and OCT are common monitoring tools for DME. The clinician compares fluid, thickness and retinal structure with prior scans while considering how the patient sees.

PDR requires more than a dry macula. New vessels are inspected clinically or with photography and angiography. Anti-VEGF can suppress them rapidly, but recurrence between visits is possible. PRP completion and traction are documented.

Injection visits also assess pressure, inflammation and symptoms of endophthalmitis. New pain, redness, discharge, photophobia or reduced vision after injection requires urgent contact. A routine next-day floater can be a harmless air bubble in some circumstances, but patients should follow the injecting clinic’s instructions rather than self-diagnose.

Examination after panretinal photocoagulation

PRP creates controlled burns in ischemic peripheral retina to reduce angiogenic drive. Follow-up checks whether neovascularization regressed and whether additional untreated retina requires laser.

Photography documents scar distribution. OCT checks for DME, which can worsen around treatment. Fields may show peripheral change, but routine field testing is not required solely to prove that laser worked.

Laser scars do not mean the eye is discharged. New vessels can recur, vitreous hemorrhage can develop and the fellow eye can progress.

Examination before and after vitrectomy

When dense hemorrhage blocks the view, B-scan evaluates retinal attachment and traction. If surgery is planned, the clinician reviews macular potential, cataract, neovascularization, prior laser and systemic fitness.

During vitrectomy, blood and vitreous traction are removed; endolaser and membrane peeling may be performed. Postoperative examination monitors pressure, inflammation, recurrent hemorrhage, retinal attachment and infection.

Gas or silicone oil changes vision and precautions. Air travel and anesthesia restrictions apply with intraocular gas. Patients follow the surgeon’s positioning and medication plan; NRT or supplements cannot override it.

Cataract-surgery planning in diabetes

Cataract can obscure retinal grading and reduce OCT quality. Before surgery, the retina is examined for DME and proliferative activity. OCT is especially valuable because visually significant macular edema may not be obvious through a cloudy lens.

Active DME or PDR may be treated before or around cataract surgery. Timing depends on visual need, retinal view, injection schedule and surgical urgency. Biometry also requires a stable corneal surface.

After surgery, inflammation can increase macular edema risk. OCT distinguishes postoperative cystoid edema, DME or a mixed pattern. Improved acuity does not eliminate the need for continued diabetic screening.

Pregnancy-specific examination

People with preexisting type 1 or type 2 diabetes should have a retinal baseline before conception when possible and early in pregnancy. Dilation with commonly used agents is considered by the responsible clinicians, while imaging that avoids intravenous dye is preferred when it answers the question.

Fluorescein angiography and anti-VEGF require pregnancy-specific risk–benefit discussion. OCT and photography are noninvasive. Active PDR may favor timely PRP because follow-up and systemic exposure considerations differ.

Baseline stage determines frequency through pregnancy and postpartum. Gestational diabetes alone has different screening guidance.

Kidney disease and angiography

Patients often equate fluorescein with iodinated CT contrast. The compounds and risk profiles differ. Fluorescein is primarily cleared through the kidneys, can temporarily discolor urine and may remain longer with impairment.

Kidney disease is also a marker of more severe diabetic microvascular risk. The ordering clinician reviews renal status, allergy history, pregnancy and the diagnostic value of angiography. OCTA may answer some perfusion questions without dye but cannot display leakage and has artifacts.

The safest decision is individualized, not a universal statement that angiography is always forbidden or always harmless.

Interpreting “normal” results

A normal OCT means no significant structural macular abnormality was captured in the scanned area at that time. It does not establish normal peripheral retina, normal pressure or normal lens.

A photograph reported as no retinopathy means no qualifying lesions were seen on gradable fields. It does not guarantee lifelong protection. A normal angiogram can be reassuring for the question asked but is not a substitute for future screening.

“Stable” is even more contextual. It should identify the comparison: stable acuity, stable OCT fluid, stable retinopathy photographs or inactive PDR after treatment. Patients can ask which meaning applies.

Interpreting incidental findings

Diabetic examinations often reveal conditions unrelated to retinopathy. Drusen may indicate AMD; an epiretinal membrane can distort the macula; a vein occlusion causes sectoral hemorrhage; a suspicious optic nerve requires glaucoma evaluation; a choroidal nevus needs documentation.

An AI system intended for diabetic retinopathy may not be cleared to diagnose these findings. The report should state whether incidental abnormalities were reviewed and what referral is needed.

The presence of diabetes should not become an anchoring bias that explains every lesion.

Seven common testing scenarios

Good acuity with severe NPDR

Photography shows the 4-2-1 pattern, while OCT is dry and acuity is 20/20. The stage still requires close follow-up because peripheral ischemic risk is high.

Mild NPDR with reduced acuity

OCT is normal but slit-lamp examination shows cataract. Retinopathy stage alone does not explain the visual loss.

DME with good vision

OCT confirms center involvement, but acuity is 20/25. Structured observation can be reasonable under Protocol V principles if follow-up is reliable.

Sudden floaters with a hazy view

Photography is not useful through dense blood. B-scan rules out detachment while the retina team plans treatment for proliferative hemorrhage.

New vessels versus IRMA

Color images are uncertain. Fluorescein leakage, OCTA architecture and stereoscopic examination help determine whether disease is proliferative.

Ungradable primary-care image

Cataract and small pupils block the camera. The patient is referred for dilated examination rather than labeled negative.

Reduced vision with a dry OCT after DME treatment

Macular ischemia, photoreceptor damage, cataract, glaucoma and surface disease are evaluated. More injection is not automatically the answer when fluid is absent.

How imaging guides anti-VEGF choice and interval

Anti-VEGF drugs differ in approved dosing, molecular targets, duration and cost. Protocol T showed that baseline acuity influenced relative average outcomes among aflibercept, bevacizumab and ranibizumab during early DME treatment, with differences narrowing over time.

Faricimab targets VEGF-A and angiopoietin-2, while higher-dose aflibercept aims for longer intervals in selected eyes. Imaging supports treat-and-extend or protocol-based decisions, but an OCT thickness target is not the only outcome.

Clinicians also consider prior response, systemic history, insurance, injection burden and fellow-eye schedule. Switching drugs is a clinical trial in the individual, documented with comparable scans and acuity.

Why more imaging is not always better

Repeated angiography adds time, cost and dye exposure without value if the result will not change management. Monthly widefield photographs may be unnecessary when OCT and examination answer the active DME question.

Conversely, omitting imaging to reduce cost can miss treatable edema or progression. The appropriate set is the smallest that safely answers the current questions.

Patients can ask whether a scan is screening, baseline documentation, treatment monitoring or investigation of unexplained loss. That purpose determines frequency.

Data ownership and portability

Retinal images are part of the medical record. Patients can request reports and, where available, image files. Portability helps when transferring retina care or obtaining a second opinion.

Raw OCT formats may not open across manufacturers, but PDFs and key B-scans still provide context. A report should include device, date, eye, signal quality and interpretation.

AI systems require privacy, cybersecurity and clarity about whether images are retained or used for model improvement. Consent and data governance belong in implementation, particularly in retail and mobile settings.

Quality assurance in screening programs

A program should monitor ungradable rates, referral completion, false negatives, time to treatment and performance across demographic and clinical groups. Camera maintenance and operator training matter as much as algorithm accuracy.

Thresholds balance sensitivity and referral burden. A highly sensitive system may refer many false positives; a highly specific system may miss disease. Local capacity and patient risk influence design.

No algorithm should quietly convert an ungradable image into a negative result. Human oversight and a clear fallback pathway protect patients.

Examination burden and accessibility

Low vision, mobility limits, dialysis schedules, cognitive impairment, language and caregiving can make a long visit difficult. Clinics can consolidate tests, provide accessible equipment, schedule around meals and arrange interpretation.

The technically perfect examination is ineffective if the patient cannot return. Treatment selection and interval should reflect transportation and cost. Durable PRP may be favored over injection-only PDR care when repeated attendance is unrealistic.

Respectful care asks about barriers rather than treating missed visits as lack of concern. Imaging can be moved closer to patients, but complex disease still needs specialist access.

What to do after the examination

Before leaving, confirm whether driving is safe after dilation, whether a same-day treatment is planned and which drops or activities apply. Obtain the stage and DME status for each eye in writing.

Share advanced or changing findings with the diabetes, primary-care and kidney teams. The eye report should not simply say “abnormal”; it should state the urgency and return date. If a referral is placed, know which clinic will contact whom.

Store injection and laser dates, drug names and any dye or medication reactions. These details become important during travel, emergency care and insurance changes.

If the explanation and images appear inconsistent, ask the clinician to point to the lesion or scan feature and explain how it changes care. A second opinion is reasonable for major surgery or costly long-term treatment, but active PDR, retinal detachment or infection should not be delayed while records circulate.

Finally, put the next appointment in more than one calendar and arrange transportation early. Diabetic retinal care is time-sensitive but often asymptomatic, so memory cannot depend on vision getting worse. If insurance, work or caregiving makes the interval impossible, tell the clinic before the visit is missed; an alternate site, schedule or treatment strategy may be safer.

Clear communication about those constraints is clinically relevant information, not an inconvenience, because follow-up reliability directly influences which retinal treatment can protect vision safely over time.

That planning deserves explicit documentation.

Where Netra Restoration Therapy may fit

Netra Restoration Therapy (NRT) may use the confirmed ocular findings to frame adjunctive goals involving sleep, stress, activity, nutrition and adherence. It cannot substitute symptom interpretation for imaging.

NRT does not stage retinopathy, measure DME or determine whether new vessels regressed unless qualified retinal examination and appropriate imaging show it. TCM evidence for diabetic retinopathy remains heterogeneous and methodologically limited. Changes in vision after adjunctive care must be verified and never used to cancel injections or laser independently.

Oral herbs can alter glucose, bleeding and drug metabolism. Disclose them before angiography, injections and surgery. Sudden loss or floaters goes directly to urgent retinal care.

Learn about Netra Restoration Therapy for diabetic retinopathy, Netra Eye Institute’s approach, the silent beginning of diabetic eye disease and how to request an appointment.

Questions to ask about an image

  • What question did this test answer?
  • Was image quality adequate?
  • What is the stage in each eye?
  • Does OCT show center-involved DME?
  • Is there leakage, nonperfusion or neovascularization?
  • How does the finding change treatment?
  • Which test will be repeated and when?
  • What symptom should prompt an earlier visit?

Frequently asked examination questions

Do I need both dilation and photographs?

Often they provide complementary information: examination gives a dynamic stereoscopic view; photographs document lesions. The exact combination depends on findings and resources.

Does OCT require dye?

Standard OCT and OCTA do not require injected dye. Fluorescein angiography does.

Can OCT detect PDR?

It may show preretinal tissue or neovascular complexes, but a standard macular OCT cannot exclude peripheral PDR. Dilation and wide retinal assessment remain necessary.

Is fluorescein safe with kidney disease?

Fluorescein differs from iodinated contrast, but medical history and prior reactions still guide use. The ordering clinician decides whether the benefit justifies the test.

Can an AI camera replace all annual eye visits?

No. It can fulfill retinal screening in appropriate settings but does not evaluate every eye structure or manage positive disease.

Can NRT imaging prove healing?

Only validated retinal examination and comparable imaging can document anatomy. NRT itself cannot redefine a stage or prove capillary regeneration.

The central idea

Each part of a diabetic eye examination has a job. Dilation and photography stage the retina; OCT measures macular structure; angiography maps leakage and perfusion; ultrasound sees through opaque media; pressure and anterior examination find other complications.

The best visit ends with an integrated plan, not merely images. NRT may support whole-person care after that plan is established, but never replaces the diagnostic tools that make silent retinal disease visible.

References

  1. National Eye Institute. Diabetic Retinopathy. Updated 2025.
  2. American Academy of Ophthalmology. Diabetic Retinopathy Preferred Practice Pattern. 2024.
  3. American Diabetes Association. Standards of Care in Diabetes—2025. Diabetes Care. 2025.
  4. Early Treatment Diabetic Retinopathy Study Research Group. Grading diabetic retinopathy from stereoscopic color photographs. Ophthalmology. 1991;98:786-806.
  5. Silva PS, Cavallerano JD, Sun JK, et al. Peripheral lesions identified by ultra-widefield imaging and risk of retinopathy worsening. Ophthalmology. 2015;122:949-956.
  6. Sun JK, Radwan SH, Soliman AZ, et al. Neural retinal disorganization as a robust marker of visual acuity in DME. Diabetes. 2015;64:2560-2570.
  7. Baker CW, Glassman AR, Beaulieu WT, et al. Initial management of CI-DME with good visual acuity. JAMA. 2019;321:1880-1894.
  8. Glassman AR, Baker CW, Beaulieu WT, et al. Protocol V observation approach. JAMA Ophthalmology. 2020;138:341-349.
  9. de Carlo TE, Romano A, Waheed NK, Duker JS. A review of optical coherence tomography angiography. International Journal of Retina and Vitreous. 2015;1:5.
  10. Spaide RF, Fujimoto JG, Waheed NK, Sadda SR, Staurenghi G. OCT angiography. Progress in Retinal and Eye Research. 2018;64:1-55.
  11. Vujosevic S, Aldington SJ, Silva P, et al. Screening for diabetic retinopathy. Lancet Diabetes & Endocrinology. 2020;8:337-347.
  12. Joseph S, et al. Real-world accuracy of AI diabetic retinopathy screening. 2024.
  13. Abràmoff MD, et al. Autonomous AI adoption lessons. 2025.
  14. Ling J, Xie Z, Luo X, et al. Evidence map of TCM for diabetic retinopathy. Systematic Reviews. 2025;14:45.

Medical Disclaimer: This article provides general education and is not medical advice, diagnosis or a personal imaging schedule. Sudden floaters, dark haze, distortion, a curtain, marked vision loss, eye pain, redness or neurologic symptoms requires urgent care. Tell clinicians about pregnancy, prior dye reactions, kidney disease and every medicine or supplement. Netra Restoration Therapy is adjunctive and cannot replace dilation, photography, OCT, angiography, ultrasound, anti-VEGF, laser, surgery or emergency evaluation.

More From the Blog

Request Consultation
Conditions We Treat
\n