What can an eye measurement tell you about health?

“Eye scan” can mean several different things. Measuring the clear tissue at the front of the eye, identifying damage in a photograph of the back of the eye and estimating a substance through light signals are different tasks. Before interpreting a result, identify the device, the measurement and the decision it is meant to support. [1] [3] [8] [9]

Longevitytech.fund — supporting longevity research

Longevitytech.fund is proud of its backing for Occuity’s research into noncontact optical measurements of the eye. Occuity announced the fund’s investment on September 27, 2021, with an update on August 24, 2022. We support the ambition to make useful health measurements simpler and less invasive. [11] [1] [8] [9]

What is established, and what is still being tested?

• Established examples: a published PM1 study assessed corneal thickness measurements, and FDA records support a defined US use for IDx-DR, now called LumineticsCore, in diabetic eye screening. [1] [3] [4] [5]

• Development programs: Occuity's Indigo, SD1 and Alzheimer's-oriented reader address different proposed measurements. Company reports through September 22, 2026 do not establish clinical performance of the final devices for these uses. [8] [9] [10] [12]

• What would strengthen the evidence: independent studies of the exact device and intended users, with individual errors, missed cases, unsuccessful readings and consequences for care clearly reported.

Different devices do different jobs

Corneal thickness: a useful measurement within eye care

The cornea is the clear tissue at the front of the eye. A pachymeter measures its thickness. That measurement can inform an eye-care assessment, but it cannot diagnose glaucoma alone. Glaucoma involves damage to the optic nerve and visual function. The National Eye Institute describes a comprehensive examination with dilated pupils and visual-field testing to check the area you can see. [1] [2]

A 2023 study assessed the handheld, noncontact PM1 in 105 adults with normal corneas. Examiners measured each person's right eye three consecutive times with each instrument, comparing PM1 with ultrasound and two established desktop optical devices. PM1 repeatability was similar to ultrasound; the desktop optical devices were more repeatable. [1]

The comparison method changed the answer about agreement. PM1 and ultrasound were not directly interchangeable despite a small average difference. Agreement with Lenstar was closer, and the authors considered those measurements potentially interchangeable in clinical use. The practical lesson is to check which two instruments are being compared. [1]

Occuity funded the study; the authors reported no proprietary interest. It was a single-visit measurement study and excluded people with prior eye surgery or corneal surface disease. It supplies useful evidence for that measurement task, but not performance in every eye condition, accuracy over years or improved patient outcomes. [1]

For follow-up, ask whether the instrument and measurement conditions were comparable. A changed number after switching devices may reflect the method rather than a change in your cornea.

Similar repeated readings are only part of accuracy

Repeatability asks whether readings stay close under similar conditions. Agreement asks whether they are sufficiently close to a suitable comparison method. Clinical validity asks whether the result identifies the condition or risk claimed. Clinical usefulness asks whether acting on it improves decisions and outcomes.

Imagine a hypothetical instrument that always reads too high. It can be very repeatable while still being wrong. Likewise, a small average difference in a study can conceal errors large enough to matter for particular people. The spread of differences matters as well as the average. [1]

When evaluating a performance claim, look for who was tested, the comparison method, failed measurements and the size of individual errors. Measurements made by experienced examiners in normal eyes do not automatically establish reliable unsupervised use or performance in eye disease. [1]

Retinal screening: a defined use with a follow-up pathway

The retina is the light-sensitive tissue at the back of the eye. Diabetes can damage its blood vessels. IDx-DR, now marketed as LumineticsCore, analyzes retinal photographs to identify more-than-mild diabetic retinopathy, a specified threshold of diabetic eye disease. [3] [4] [5]

The FDA granted the original De Novo authorization on April 11, 2018. The main study enrolled 900 people at ten US primary-care sites; 819 had fully analyzable data. The comparison used an independent reading center and established retinal imaging methods. The review considered both diagnostic performance and whether an interpretable result could be obtained. Not all incomplete analyses reflected a failed device reading; some lacked the comparison data or other information needed. [3]

A June 10, 2021 FDA clearance covered an updated IDx-DR system. The US manufacturer labeling available on September 22, 2026 specified adults aged 22 or older with diabetes, no previous diagnosis of diabetic retinopathy and use of the Topcon NW400 camera. Eligibility and image quality are part of using the test correctly. [4] [5]

The labeling excludes people with visual symptoms, those who are pregnant and people with certain retinal conditions or previous retinal treatments. New blurred vision or floaters need an eye-care assessment; this screening pathway is not intended to explain those symptoms. [5]

A positive result needs referral for evaluation and possible treatment. An unreadable result is not a negative result: the labeling calls for immediate retesting or referral. False positives and false negatives are possible, and the system neither diagnoses diabetes nor screens comprehensively for other eye diseases. [5]

This is evidence for a particular task, population and equipment. It cannot validate another camera, algorithm or optical device. Diagnostic performance also does not by itself tell us how much a whole screening program reduces later vision loss. [3] [4] [5]

Glucose through the eye: separate stages of development

Occuity's Indigo program aims to estimate glucose without a blood sample. As of September 22, 2026, its product page described it as being in research and development. A December 11, 2025 company update described early investigations in people using a prototype and problems including alignment and movement. It then described a revised optical approach studied in laboratory solutions, with further human work planned. These are distinct stages; the early prototype does not establish the revised device's clinical performance. [8]

The company reported encouraging correspondence with blood glucose in early work. The update, however, did not provide a complete, independently assessed comparison of final-device readings with a clinical reference method. Correlation means that two readings tend to rise and fall together. They can still be too far apart for a safe treatment decision. [8]

A useful glucose validation study would need to assess errors across the intended range of values and users, especially when an error could prompt a wrong action. It would also need to include unsuccessful readings and ordinary use conditions. These are questions for testing, not findings supplied by the announcement.

A research demonstration is not a basis for replacing prescribed glucose monitoring or changing medication. Those decisions need evidence and authorization for the particular device and use.

Diabetes risk is a different target from glucose measurement

Occuity's SD1 program studies a signal from the eye's lens. The company describes fluorescence—light emitted after illumination—related to advanced glycation end products, compounds formed through chemical reactions involving sugars. Its aim is to help identify people who may need further diabetes-risk assessment. The grant announcement places further development and clinical validation among the work ahead and describes supporting existing diagnostic pathways. [9]

A November 2023 company report described an internal prototype study of 27 people, including two with known diabetes. It compared the optical signal with age and highlighted results from those two participants. The account did not report screening sensitivity or specificity: how often the test would detect previously undiagnosed diabetes, and how often it would correctly give a negative result in people without diabetes. It demonstrates early human feasibility work, not established screening accuracy. [12]

A screening study would need to show who is correctly referred, who is missed, what else causes positive results and whether the added test improves the existing pathway. Shared optical components with a validated corneal instrument do not answer those questions.

An Alzheimer's research aim also needs its own test

Occuity announced a research contract for an Amyloid Beta Plus Reader in March 2024. That announcement describes development, not a clinical accuracy result or regulatory authorization. Evidence that a device measures corneal thickness cannot establish that a lens signal identifies Alzheimer's pathology or predicts dementia. [10]

The same boundary applies to other types of brain-age research. Blood-protein models and MRI brain-volume measurements use different biological information. Neither supplies validation for an optical eye device.

How Healthy Longevity Clinic experts evaluate the evidence

For someone asking whether an eye measurement can reveal diabetes, the first distinction is between detecting diabetic damage in an already diagnosed person and diagnosing diabetes itself. LumineticsCore addresses the former within its intended patient group. A retinal screening result cannot be read as a general diabetes test, and an unreadable image needs action rather than reassurance. [3] [4] [5]

For someone tracking a corneal measurement, the meaningful question is whether the same instrument and conditions were used. The PM1 study's closer agreement with Lenstar and lack of direct interchangeability with ultrasound show why the particular pair of instruments matters. For someone considering a new glucose or lens-risk device, the question changes again: has the exact version been tested for the decision being proposed? [1] [8] [9]

A useful consultation identifies the anatomical measurement or screening target, confirms that the person fits the studied use and explains what each result would lead to. Stronger evidence for an emerging device would show reliable individual performance in its intended users and a beneficial effect on care. A successful reading, a grant or a related product's authorization alone does not establish that benefit.

Read regulatory announcements precisely

Occuity announced CE marking for PM1 on October 1, 2024. This is a manufacturer-reported European conformity milestone for that pachymeter. It does not extend to Indigo, SD1 or the Alzheimer's project. For actual use, the relevant facts are the device's current certificate, intended purpose and local requirements. [6]

The company's September 15, 2023 announcement described an FDA 510(k) submission for PM1. A submission is an application, not a clearance decision; that announcement cannot serve as evidence of US authorization. [7]

Research grants, patents, trial permissions and availability describe different aspects of development. None alone proves that testing improves health. An informative report states its actual clinical role, uncertainty and next step, without turning a single eye signal into a measure of whole-body health.

Three questions for a consultation

1. Is this test measuring corneal thickness, screening for diabetic retinal damage or estimating a different signal—and which exact device and version produced it?

2. Was that use tested in people with my diabetes status, age and eye history, and how are failed or unreadable measurements handled?

3. What would a positive, negative or changed result mean for care, and who would arrange any needed eye assessment or further testing?

Common questions

Can a corneal thickness result tell me whether I have glaucoma?

Not on its own. It can contribute to eye care, while glaucoma assessment considers the optic nerve and visual function as part of a fuller examination. [1] [2]

Does diabetic retinal screening diagnose diabetes?

No. The LumineticsCore use described here is for eligible people who already have diabetes. It looks for a specified level of diabetic eye disease. [5]

Is an unreadable retinal result reassuring?

No. It means a valid result was not obtained. The US labeling calls for immediate retesting or referral, rather than treating it as a negative screen. [5]

Does the PM1 study validate Occuity's glucose or Alzheimer's projects?

No. It studied corneal thickness in adults with normal corneas. The other projects measure different signals and need their own validation for the intended clinical decisions. [1] [8] [9] [10]