Synthetic teaching case
Case 2: two independent caution pathways
Synthetic LASIK example showing why CER-AI does not hide discordance inside one average score.
Not real patient dataThis worked example was created to teach the current CER-AI v2.0 rules. It is not a clinical decision.
Source inputs
| Field | Value |
|---|---|
| Procedure | LASIK |
| Age / thinnest | 25 years / 505 µm |
| I-S / SRAX | +0.80 D / 15° |
| Manifest MRSE | -5.00 D |
| Flap / ablation / RSB | 110 / 70 / 325 µm |
| Final BAD-D | 2.00 |
| NICE inputs | K2 46 D; central 505 µm; B.Ele.Th 16 µm; I-S 0.80 D |
Pathway evaluation
| System | Calculation / finding | Result |
|---|---|---|
| ERSS | Topography 1 + pachymetry 1 = 2 | PASS |
| BAD-D | 2.00 suspicious | CAUTION |
| NICE | 2 + 2 + 2 + 1 = 7 | CAUTION |
| PS3 | Complete; no Moderate/High | PASS |
| Tissue safety | Independent checks complete | PASS |
Final CER-AI result
CAUTION
Report summary
The report lists BAD-D and NICE as separate caution drivers. Because exactly two of the four completed scoring systems are CAUTION and no stop is present, the combined result is CAUTION.
Learning points
- NICE points are not added to ERSS points.
- BAD-D component context does not create extra caution counts.
- The surgeon sees which pathways disagree and reviews the underlying maps.