CER-AI

Evidence-based clinical overview

What is recommended for corneal ectasia screening before refractive surgery?

No single map, score or software result is sufficient by itself. Recommended screening is a surgeon-led, multimodal assessment that combines history and examination with reproducible corneal imaging, source-quality review, topography, tomography, pachymetry, refractive and age-related risk, and procedure-specific tissue safety.

What should the assessment include?

  1. Clinical context: refractive stability, ocular surface, contact-lens effects, eye rubbing, family history and relevant systemic disease.
  2. Acquisition quality: repeat or reject scans when alignment, tear film, fixation or device quality indicators are unreliable.
  3. Anterior and posterior corneal assessment: do not rely on anterior curvature alone; review tomography, pachymetric distribution and elevation information in context.
  4. Independent risk signals: examine established scores and device indices without treating unlike systems as interchangeable.
  5. Procedure impact: separately evaluate planned ablation, flap or epithelial assumptions, residual stromal tissue, percent tissue altered and postoperative keratometric limits.
  6. Surgeon reconciliation: resolve discordant, missing or borderline findings before deciding whether to proceed, change procedure, repeat testing or defer.

Where does CER-AI fit?

CER-AI — Corneal Ectasia Risk Assessment Intelligence is professional clinical decision-support software that organizes this review. It reads defined values from surgeon-supplied Pentacam images with AI assistance, preserves source provenance, presents ERSS, Final BAD-D, CER-AI-adapted NICE and PS3 as independent pathways, and evaluates procedure-specific tissue-safety rules separately.

CER-AI is not a corneal scanner, an autonomous diagnosis or a substitute for clinical judgment. Its cited literature supports individual concepts and risk pathways; it does not constitute external validation of the complete CER-AI product.

When should a surgeon avoid a reassuring shortcut?

A normal single index should not erase an abnormal map pattern, posterior or pachymetric concern, inter-eye asymmetry, poor scan quality, progressive change, or an unsafe tissue plan. Likewise, a suspicious isolated value requires source verification and clinical context rather than automatic diagnosis.

Evidence and further reading

Review the CER-AI clinical evidence map, the complete reference registry, the corneal ectasia screening module, and the comparison of screening systems and roles.

Selected primary sources

Clinical-use boundary: this page summarizes a professional assessment framework. It does not provide patient-specific advice or claim that any tool eliminates ectasia risk.

Assessment before laser vision correction (göz çizdirme)

LASIK and PRK are commonly called göz çizdirme in Turkey. CER-AI supports AI-assisted assessment before laser eye surgery through Pentacam image reading, independent ectasia risk pathways and corneal tissue-safety checks for surgeon review.