Coming soon: real clinical case examples

De-identified examples selected from real clinical cases will be added to this section soon. Until then, the worked cases below remain clearly labeled synthetic teaching cases and do not contain real patient data.

Case A: concordant concern

Anterior curvature asymmetry, corresponding posterior-elevation deviation, and abnormal pachymetric progression point toward the same region. The learning task is to verify acquisition quality and source identity, then explain why multiple independent channels are concordant rather than merely counting abnormal labels.

Case B: isolated composite alert

Final BAD-D is outside the device reference band while the anterior curvature map appears regular. The learning task is to inspect the component deviations, raw elevation and pachymetry displays, quality status, and fellow eye before deciding whether the composite is coherent or isolated.

Case C: reassuring shape, unfavorable plan

Topography and tomography appear reassuring, but planned flap and ablation geometry leave limited predicted stromal reserve. The learning task is to keep tissue safety independent from ectasia-pattern screening and reconsider the proposed procedure rather than allowing normal maps to override the plan.

Case D: missing decision-critical source

A required display is absent or unreadable. The learning task is not to infer a favorable result from missingness. Identify the exact field and source needed, obtain a repeat examination or surgeon-confirmed value when appropriate, and preserve the uncertainty in documentation.

Worked synthetic cases

Selected sources

  1. Randleman et al., Ophthalmology 2008
  2. Chan et al., JCRS 2018
  3. Moshirfar et al., Ophthalmology and Therapy 2021

See the complete CER-AI medical reference registry for broader context.