What PS3 contributes
PS3 organizes selected corneal, elevation, pachymetric, SRAX, and inter-eye findings as Normal, Moderate, or High factors. CER-AI keeps this pathway independent from ERSS, Final BAD-D, and NICE so the origin and consequence of every finding remain visible.
Automated factor set
The current CER-AI implementation evaluates anterior Km, thinnest pachymetry, front and back elevation at the thinnest point, PPI Average, bilateral asymmetry, and SRAX. Corneal Thickness Map, Relative Thickness Map, and PTI/CTSP morphology remain explicit surgeon-review items and are not silently counted by automation.
Procedure-specific consequence
One Moderate factor ordinarily defers LASIK while allowing PRK and SMILE when the automated set is otherwise complete. CER-AI exception: isolated thickness Moderate at 490-499 µm permits LASIK PASS WITH CAUTION when ERSS, NICE, and Final BAD-D are each PASS or CAUTION. At exactly 500 µm there is no PS3 thickness escalation. Two or more Moderate factors, or any High factor, defer all three procedures. Missing decision-critical factors make the pathway incomplete; an already established defer finding is retained.
Do not add unlike quantities
Moderate and High are categorical findings, not interchangeable numeric points. CER-AI counts factors only within PS3, converts the selected procedure's PS3 result to an independent pathway disposition, and does not add PS3 counts to ERSS or NICE totals.
Evidence and implementation boundary
The public module documents the current CER-AI operational implementation and its source literature. It does not claim that every operational threshold has been prospectively validated for every device, population, or procedure.
Automated PS3 factors in CER-AI
| Factor | Moderate | High |
|---|---|---|
| Anterior Km | 48–50 D | >50 D |
| Thinnest pachymetry | 470–499 µm | <470 µm |
| Elevation at thinnest | — | F.Ele.Th >12 µm or B.Ele.Th >15 µm |
| PPI Average | >1.20; may be ignored only with corneal astigmatism >2 D and surgeon-confirmed otherwise-normal tomography | — |
| Inter-eye asymmetry | 4 of 5 differences reach threshold | 5 of 5 differences reach threshold |
| SRAX | — | >20° with nonnegative signed I-S and required surgeon confirmation; may be ignored only with tomographic astigmatism <1 D and surgeon-confirmed enantiomorphism |
Inter-eye thresholds are anterior Km 0.3 D, posterior Km 0.1 D, thinnest pachymetry 12 µm, front elevation 2 µm, and back elevation 5 µm; equality counts as exceeded. Source footnotes are applied only inside PS3. The population-modification note does not alter CER-AI thickness thresholds. The focal-epithelial-thinning note is documented but creates no duplicate automated I-S factor because CER-AI does not count a separate PS3 I-S/S-I factor.
PS3 procedure disposition
| Factor pattern | LASIK | PRK / SMILE |
|---|---|---|
| No Moderate or High; complete | Allowed | Allowed |
| One Moderate; complete | Defer | Allowed |
| Only Moderate is thickness 490–499 µm; ERSS/NICE/Final BAD-D each PASS or CAUTION | PASS WITH CAUTION | Allowed |
| >=2 Moderate or >=1 High | Defer | Defer |
| Required factor missing | Incomplete unless already deferred | Incomplete unless already deferred |
Manual map-morphology review items remain visible but are not counted as automated factors.
Selected sources
- Elhusseiny et al., Benha Medical Journal 2021
- Sinjab, Step by Step: Reading Pentacam Topography, 3rd ed.
- CER-AI Clinical Evidence
See the complete CER-AI medical reference registry for broader context.