CER-AI Clinical Evidence

Clinical Evidence

CER-AI is a clinical decision-support platform for structured preoperative corneal ectasia risk assessment. Its public evidence layer identifies the literature that informs individual risk pathways and related safety concepts, while keeping those pathways independently interpretable.

Evidence boundary: the publications below support specific concepts, risk systems, or clinical variables used by CER-AI. They should not be interpreted as external validation of CER-AI as a complete software product unless explicitly stated. CER-AI does not replace surgeon judgment.
Clinical reviewer: Last clinical review: 12 September 2026Editorial and evidence policy

Medical-quality and evidence standards

CER-AI separates source quality, relevance, implementation and product validation so that one type of evidence is not presented as another.

Source selection

Primary peer-reviewed publications are preferred for clinical concepts. Textbooks and manufacturer documentation are identified separately and used only for their appropriate educational or device-definition roles.

Claim-to-source traceability

Selected publications are mapped to the specific pathway or safety concept they support. The complete citation remains in the single medical reference registry.

Limitations retained

Associations, retrospective studies, case reports and device documentation are not presented as equivalent forms of evidence, and published thresholds are not generalized beyond their stated context.

Independent product validation

No publication listed here is described as external validation of the complete CER-AI software unless a dedicated CER-AI validation study is performed and published.

Core evidence mapped to CER-AI pathways

Randleman / ERSS

Original ectasia risk framework

Randleman JB, Woodward M, Lynn MJ, Stulting RD. Risk assessment for ectasia after corneal refractive surgery. Ophthalmology. 2008;115(1):37-50. DOI: 10.1016/j.ophtha.2007.03.073.

Supports the classical multivariable ectasia-risk framework involving topographic pattern, residual stromal bed, age, corneal thickness and refractive magnitude.

Randleman / ERSS

Independent validation

Randleman JB, Trattler WB, Stulting RD. Validation of the Ectasia Risk Score System for Preoperative Laser In Situ Keratomileusis Screening. Am J Ophthalmol. 2008;145(5):813-818.e2. DOI: 10.1016/j.ajo.2007.12.033.

Supports use of the ERSS as an independently evaluated preoperative LASIK screening system. CER-AI retains ERSS as a distinct pathway rather than mathematically blending it with tomography systems.

PRK context

Ectasia risk after PRK

Sorkin N, Kaiserman I, Domniz Y, Sela T, Munzer G, Varssano D. Risk Assessment for Corneal Ectasia following Photorefractive Keratectomy. J Ophthalmol. 2017;2017:2434830. DOI: 10.1155/2017/2434830.

Provides PRK-specific context for ectasia risk and demonstrates why a LASIK-derived risk score should not be treated as an absolute individual risk calculator for PRK.

NICE implementation source

Post-LASIK ectasia risk factors

Navarro-Naranjo PI, Bareño Silva J, Trujillo-Quijano MA. Assessment of Preoperative Risk Factors for Post-LASIK Ectasia Development [Letter]. Clin Ophthalmol. 2024;18:881-883. DOI: 10.2147/OPTH.S464217.

This is the explicit literature source used in the current CER-AI NICE implementation documentation. CER-AI displays the NICE pathway separately from ERSS and BAD-D.

PS3

Practical subjective scoring system

Elhusseiny FY, Hamed AM, Sinjab MM, Salem TI, El-Shahed AF. Ectasia risk score system and practical subjective scoring system in screening of keratorefractive surgery candidates. Benha Med J. 2021. DOI: 10.21608/bmfj.2021.100688.1503.

Provides the identified publication for the PS3 pathway. CER-AI keeps PS3 findings visible as their own risk channel and documents triggered factors rather than presenting PS3 as a replacement for tomography or tissue-safety review.

Tomographic AI

Enhanced ectasia susceptibility detection

Lopes BT, Ramos IC, Salomão MQ, et al. Enhanced Tomographic Assessment to Detect Corneal Ectasia Based on Artificial Intelligence. Am J Ophthalmol. 2018;195:223-232. DOI: 10.1016/j.ajo.2018.08.005.

Supports the broader role of multivariate tomographic analysis and artificial intelligence in detecting ectasia susceptibility. CER-AI does not claim to reproduce proprietary indices from this study unless their required inputs and implementation are explicitly available.

BAD / Pentacam tomography

Belin/Ambrósio ectasia screening

Belin MW. Ectasia screening with the Pentacam Scheimpflug tomographer: the Belin Ambrosio module. Acta Ophthalmologica. 2025. DOI: 10.1111/aos.16814.

Provides contemporary background on the Belin/Ambrósio Display and the role of tomographic parameters in ectasia screening. In CER-AI, Final BAD-D remains an independent Pentacam-derived risk channel.

Tissue safety

Percent Tissue Altered

Santhiago MR, Smadja D, Gomes BF, et al. Association Between the Percent Tissue Altered and Post-LASIK Ectasia in Eyes With Normal Preoperative Topography. Am J Ophthalmol. 2014;158(1):87-95.e1. DOI: 10.1016/j.ajo.2014.04.002.

Supports the clinical relevance of procedure-related tissue alteration in post-LASIK ectasia risk. CER-AI keeps structural/tissue-safety calculations conceptually separate from intrinsic corneal susceptibility pathways.

Tissue safety

PTA in suspicious topography

Santhiago MR, Smadja D, Wilson SE, Krueger RR, Monteiro MLR, Randleman JB. Role of Percent Tissue Altered on Ectasia After LASIK in Eyes With Suspicious Topography. J Refract Surg. 2015;31(4):258-265. DOI: 10.3928/1081597X-20150319-05.

Extends the tissue-alteration literature to eyes with suspicious preoperative topography and is relevant to CER-AI's separation of structural impact from topographic/tomographic susceptibility.

Pentacam source provenance

Manufacturer field definitions

OCULUS Optikgeräte GmbH. Pentacam, Pentacam AXL and Pentacam AXL Wave instructions and user manuals. Manufacturer document portal; revision varies by device and software version.

Supports device-specific screen and field terminology only. Manufacturer documentation is identified separately from peer-reviewed clinical outcome evidence and does not validate CER-AI's risk classification.

Evidence interpretation principles

  • Independent channels: ERSS, Final BAD-D, NICE and other supported pathways are interpreted separately rather than collapsed into an opaque single score.
  • Source provenance: Pentacam-derived variables should retain their defined screen/source context.
  • Procedure safety is separate: residual stromal bed, pachymetry, ablation planning and postoperative keratometry constraints are not overridden by a reassuring ectasia score.
  • No unsupported extrapolation: CER-AI does not present proprietary indices, unpublished formulas, or unavailable device outputs as though they had been reproduced.
  • Clinical decision support: the software is intended for qualified ophthalmic professionals and is not an autonomous diagnosis or treatment system.

Medical References

The complete CER-AI literature registry is maintained as a single consolidated reference list, including ERSS, NICE, PS3, BAD-D, Pentacam tomography, PRFI, PTA, RTA, SCORE, biomechanics, PRK/LASIK ectasia and related background evidence.

Open References