Symmetry of the Upper Eyelid Following Bilateral Blepharoptosis Repair Using a Minimally Invasive Conjoint Fascial...

Aim to use an artificial intelligence (AI) approach to automatically measure the marginal reflex distance 1 (MRD1), corneal exposure rate (CER) and percentage of overlapping curvatures (POC) to assess symmetry in patients with mild to moderate bilateral asymmetric ptosis, and to identify the optimal parameters for evaluating symmetry.

A paired t test was employed to compare alterations in MRD1, CER, and POC prior to and following surgery. The AI measurement outcomes were then compared with manual measurements utilizing ImageJ to ascertain reliability and accuracy. Finally, the area under the receiver operating characteristic curve (AUC) reflecting diverse eyelid symmetry parameters was analyzed.

The MRD1 symmetry difference prior to surgery was measured at 0.89 ± 0.73mm, and postoperatively, this value decreased significantly to 0.53 ± 0.49mm (P < 0.0001). The preoperative CER demonstrated a statistically significant increase, from 75.01% ± 11.71% to 82.03% ± 9.20% (P < 0.0001). The POC value increased from an initial 90.62% ± 6.28% to a final 94.33% ± 3.61% (P = 0.004). Furthermore, the MRD1 symmetrical value was highly sensitive (98.48%), but less specific (37.88%). In addition, the study showed that the sensitivity of CER was 76.79%, while its specificity was 58.93%.

Minimally invasive CFS suspension effectively corrects mild-to-moderate bilateral blepharoptosis, with CER emerging as the most reliable metric for postoperative symmetry. Integration of AI tools enhance the objectivity of outcome assessments. A multi-parameter approach combining CER and POC is recommended for comprehensive clinical evaluation of eyelid symmetry.

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .