CER-AI is a clinical decision support platform for preoperative corneal ectasia risk screening and refractive surgery evaluation. It analyzes uploaded Pentacam images to support ophthalmologists assessing candidates for LASIK, PRK, and SMILE.
For corneal refractive surgery, CER-AI supports preoperative screening, corneal ectasia risk assessment and separate tissue-safety analysis before LASIK, PRK or SMILE. For refractive lens surgery and cataract surgery, it supports IOL category suitability assessment for monofocal, EDOF and multifocal pathways, lens selection, spherical IOL calculation and toric planning.
Preoperative risk screening is essential for patient safety before laser vision correction (refractive surgery).
CER-AI brings together four literature-informed assessments: Randleman ERSS, CER-AI-adapted NICE, Pentacam Final BAD-D and PS3. Each is evaluated independently, and its score or result is presented transparently to the surgeon.
When screening and separate tissue-safety checks permit treatment, the application performs procedure-specific planning calculations for LASIK, PRK or SMILE according to the surgeon’s chosen procedure and parameters.
Final clinical decisions remain with the surgeon.
CER-AI has two main modules. The first and primary module is Corneal Ectasia Risk Assessment for refractive-surgery screening, including post-LASIK ectasia risk and keratoconus susceptibility review. The second module is IOL Selection & Calculation for cataract planning.
CER-AI does not blend the major ectasia-risk systems into one proprietary score. Each pathway is evaluated independently, with its own inputs, interpretation, and procedural consequence.
Evaluates age, pachymetry, refractive magnitude, residual stromal bed, and anterior topographic evidence.
Final BAD-D is interpreted as its own Pentacam-derived risk channel.
NICE is evaluated independently using required tomographic inputs and a component audit.
PS3 adds independent review of selected corneal, refractive, tomographic, and inter-eye findings.
After the primary ectasia module, CER-AI’s second module supports the cataract-planning sequence from Pentacam and biometry review to lens-category selection and calculation.
Which IOL category is suitable? Pentacam and biometry sources inform lens-category eligibility before the surgeon chooses a preferred lens family.
Toric planning is a capability within the IOL module, not a separate third module. The embedded toric IOL calculator is a test-only prototype requiring manufacturer verification.
A practical workflow for qualified ophthalmic professionals using the clinical application.
Open CER-AI and sign in once with your authorized username and password.
Select Refractive Surgery or IOL Calculation Surgery.
Enter patient name, age, surgeon-defined parameters, and the measurements required by the selected module.
Upload the Pentacam source images required by the selected module.
Decision-critical unreadable data must be entered manually and are never silently estimated.
Review the independent result, reasons, alternatives, and warnings. Final responsibility remains with the surgeon.
CER-AI can be added to your phone's Home Screen and opened like an app. No App Store or Google Play download is required.
1. Open cer-ai.com in Safari.
2. Tap Share.
3. Choose Add to Home Screen.
4. Tap Add.
1. Open cer-ai.com in Chrome.
2. Tap the browser menu ⋮.
3. Choose Install app or Add to Home screen.
4. Confirm.
The platform separates risk assessment, data integrity, and procedure safety so that the final recommendation remains auditable.
ERSS, Final BAD-D, NICE, and PS3 remain independently interpretable.
Important measurements are linked to defined Pentacam source fields.
Tissue and procedure-specific constraints are evaluated separately.
Missing, conflicting, and surgeon-confirmed information remains visible.
In Turkey, LASIK and PRK laser vision correction are commonly called “göz çizdirme”. CER-AI supports ophthalmologists reviewing corneal ectasia risk, keratoconus findings and tissue safety before these procedures.
Searching for “yapay zekâ destekli göz çizdirme”? CER-AI provides AI-assisted reading of Pentacam images and structured preoperative decision support. The ophthalmic surgeon determines eligibility and performs the surgery.
Explore assessment before LASIK / PRK (göz çizdirme)CER-AI is a standalone web platform for corneal ectasia risk assessment before LASIK and PRK (göz çizdirme), including review of risk factors associated with post-LASIK ectasia. It runs outside the imaging device's software.
Device-associated tools include the Belin/Ambrósio Display (BAD) in the Pentacam software ecosystem and the SCORE Analyzer on the ANTERION imaging platform. Access to these tools requires the relevant imaging system and applicable software configuration or licensing; equipment and license costs vary by supplier and configuration.
CER-AI adds a separate workflow using existing supported Pentacam reports. It does not replace tomography hardware or generate BAD-D without the device report. Current compatibility is defined for Pentacam sources; support for every imaging device is not claimed.
Current access is free for approved physicians. New demo accounts have a 10-patient allowance and require approval. A low-cost physician subscription is planned for the future; no final tariff or start date has been announced.
OCULUS Pentacam software · Heidelberg Engineering ANTERION workflows
CER-AI presents independent risk systems, canonical Pentacam data, tissue calculations, surgical eligibility, reporting, and planning support in one workflow without blending the risk-system results.
Artificial intelligence should support clinical judgment, not replace it.
CER-AI organizes and cross-checks clinical information, applies predefined safety rules, and documents the reasoning behind an assessment. The final surgical decision always remains with the surgeon.
Choose a module and open the secure clinical application.
CER-AI helps ophthalmologists review findings that may raise suspicion of subclinical keratoconus before LASIK or PRK. It supports the clinician’s diagnostic assessment by organizing Pentacam topography and tomography findings, including corneal asymmetry, thickness distribution and the device-reported Final BAD-D.
Artificial intelligence assists with reading uploaded Pentacam reports; separate, transparent rule-based algorithms organize the clinical risk assessment. Subclinical keratoconus screening is not a confirmed diagnosis. The ophthalmologist interprets the imaging together with the examination and, when needed, repeat or longitudinal measurements. CER-AI does not independently diagnose or exclude keratoconus, and validated CER-AI-specific diagnostic sensitivity or specificity is not claimed.
Research on Pentacam topographic and tomographic indices in subclinical and clinical keratoconus
Hüseyin Cengiz, M.D.
Ophthalmic Surgeon · Developer of CER-AI
After more than thirty years of surgical and administrative experience, I wanted to find a way to pass some of that accumulated knowledge on to the generations that follow.
As a refractive surgeon, one of the complications we fear most is corneal ectasia. My aim was to develop a system that could help identify ectasia susceptibility more reliably before surgery and, ultimately, help protect patients from an avoidable complication. With the rapid development of artificial intelligence, I used established clinical evidence, modern imaging data, and my own surgical experience to design a structured clinical decision-support platform.
CER-AI — Corneal Ectasia Risk Assessment Intelligence — is the single current product name used throughout the platform.
The software developed progressively through versions 0.1, 0.2, 0.3 and subsequent iterations, reaching version 2.0 at the time of writing. It is under continuous development, so by the time you read this, a newer version may already be in use.
During its development, I conducted an extensive review of the medical literature related to corneal ectasia, keratoconus susceptibility, refractive-surgery screening, corneal tomography, topography, biomechanics, and tissue safety. I used this evidence and practical surgical experience to present several ectasia-risk assessment systems within one auditable workflow.
A central principle of CER-AI is that these systems remain independent from one another. CER-AI does not merge, blend, harmonize, average, or add their results into a single opaque score. Each pathway evaluates the case separately so that the surgeon can see where agreement or disagreement exists between different risk-assessment approaches.
The primary purpose of CER-AI is not to make the surgical decision on behalf of the physician. It is designed to support the physician’s decision-making, to organize complex preoperative information, and most importantly, to help identify a finding that may otherwise have been overlooked.
To younger colleagues in particular, I would like to emphasize a principle that has guided me throughout my surgical career: Whatever the procedure, our first responsibility is to avoid harming the patient.
Technology, imaging, scoring systems, and artificial intelligence can assist us, but they should always serve that fundamental clinical responsibility.
We thank our esteemed colleagues at Vizyon Eye Hospital, Dr. Efsane Cengiz, Dr. Ümit Atik, Dr. Ali Özer, Dr. Fatih Baltu, Dr. İdil Göksel and Dr. Ümit Yapıcı, for their contributions to the development and improvement of CER-AI.