Smartphone-Based 3D Ocular Reconstruction for Automated Quantitative Blepharoptosis Evaluation

Importance External ocular examination is central to ptosis evaluation, yet it often relies on manual, experience-dependent measurements with variable accuracy and reproducibility. Objective To develop and prospectively evaluate the Digital Rulers of Ptosis (DRP), a smartphone-based 3-dimensional (3D) reconstruction system for automated quantitative external ocular assessment in blepharoptosis, including marginal reflex distance 1 (MRD1) and levator muscle function (LF), and clinical classification. Design, Setting, and Participants This diagnostic/prognostic study was conducted prospectively across 5 hospitals in 2025. Participants 3 years and older underwent smartphone video capture and same-day oculoplastic specialist assessment as the reference standard. Participants undergoing surgery received repeat specialist measurements of MRD1 and LF at 3 months postoperatively to assess recurrence. Exposure DRP uses a 10-second smartphone video for 3D reconstruction with metric unit conversion and real-time quality control. Main Outcomes and Measures Agreement between DRP and specialist measurements was assessed by mean absolute error (MAE) and intraclass correlation coefficient. Diagnostic performance was assessed by area under the receiver operating characteristic curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value. For patients who underwent surgery, 3-month postoperative recurrence was assessed. Results Among 952 participants (1904 eyes), the mean age was 27.9 years (range, 3-86 years); 493 participants were female (51.8%) and 459 were male (48.2%). A total of 1862 eyes were included in MRD1 analysis and 1830 eyes in LF analysis after quality control. DRP demonstrated high agreement with specialist measurements for MRD1 (MAE, 0.52 mm; 95% CI, 0.50-0.54 mm) and LF (MAE, 0.95 mm; 95% CI, 0.91-0.98 mm). For diagnosing ptosis, DRP achieved an AUC of 0.97 (95% CI, 0.96-0.97). For identifying surgical approach, DRP achieved an AUC of 0.98 (95% CI, 0.97-0.98). Performance remained consistent across sex, age, and geographic region. Among 206 surgical participants (248 eyes), the 3-month postoperative recurrence rate was 7.3% (95% CI, 4.4%-11.2%; 18/248 eyes), within the lower range of previously reported rates. Conclusions and Relevance This study found that accurate, standardized quantitative blepharoptosis assessment might be supported by commercially available smartphones, demonstrating the potential of smartphone-based 3D reconstruction for external ocular assessment and clinical decision-making. Further studies are needed to evaluate its clinical impact.