For educational and research purposes only. Not a medical device.
External validation

APEX formula accuracy
in real clinical practice

The APEX formula was built on a sample drawn from a clinical base of 42,000 surgeries. On held-out eyes that were not used to develop the formula, APEX matches the Kane formula in accuracy and outperforms the classic formulas.

42 000
surgeries
in the APEX clinical base
76%
of eyes within ±0.5 D
for the table leader
№1
rank by the FPI index
among the compared formulas

Formula comparison

Accuracy of hitting the target refraction, measured on 960 held-out eyes that were not used to develop the formula. Each formula was computed with its own optimized constant under equal conditions. The Kane formula was included as an external benchmark. Results are sorted by the FPI index.

Single-center retrospective data. The full methodology and statistical analysis will be published separately.

Formula MAE (D) MedAE ±0.5 D FPI vs APEX
1APEX 0.348 0.278 76% 0.617 leader
2PEARL-DGS 0.370 0.296 75% 0.571 comparable
3Kaneexternal 0.352 0.288 75% 0.567 comparable
4Castrop 0.408 0.332 69% 0.454 lower
5Hoffer Q 0.423 0.345 68% 0.415 lower
6T2 0.415 0.335 68% 0.374 lower
7SRK/T 0.427 0.347 67% 0.357 lower
8MIKOF 0.427 0.348 68% 0.356 lower
9Haigis 0.399 0.327 69% 0.347 lower
10Holladay 1 0.402 0.322 70% 0.339 lower
Optimization The formula constants were estimated on a sample drawn from a clinical base of 42,000 surgeries.
Validation 960 held-out eyes not used to develop the formula. All formulas were computed on the same sample, each with its own optimized constant (Hoffer–Savini 2021 protocol).
FPI A composite accuracy index for the formula. Higher values reflect better overall accuracy.
“Comparable” The difference from APEX is not statistically significant.