Background: To evaluate automated lens average density from IOL Master 700 as a quantitative supplementary measure for cataract assessment, and whether cataract phenotypes and risk profiles differ by axial length (AL). Methods: This retrospective cross-sectional study enrolled 2000 participants; 1906 eyes were analyzed after excluding poor images and ocular pathologies. A Python-OpenCV pipeline automatically extracted lens luminance parameters. Diagnostic performance, repeatability, and biometric associations were assessed using receiver operating characteristic, intraclass correlation coefficient (ICC), and regression analyses. Results: Lens average density showed excellent diagnostic performance (area under the curve = 0.975, 95% confidence interval: 0.969–0.981; cutoff = 68.66; sensitivity 88.5%, specificity 94.4%), correlating well with Lens Opacities Classification System III nuclear opalescence ( ρ = 0.804). It had excellent repeatability (inter-technician ICC = 0.935, coefficient of variation [CV] = 1.85%; inter-time ICC = 0.928, CV = 1.99%). Among cataract patients, nuclear cataract was significantly more frequent in AL ≥26 mm eyes (83.21% vs. 39.83%, P < 0.001). Multivariable logistic regression identified age (odds ratio [OR] = 1.104/year), AL (OR = 1.360/mm), anterior aqueous depth (AQD) (OR = 0.275/mm), and lens thickness (LT) (OR = 2.013/mm) as independent predictors (all P < 0.001). Critically, risk factor profiles diverged by AL: in the non-high myopia group (AL <26 mm), AQD and LT were core predictors, whereas in the high myopia group (AL ≥26 mm), AL and white-to-white (WTW) distance emerged as principal indicators, with LT conferring seven-fold increased odds (OR = 7.012). Conclusions: Automated lens average density provides a reliable, objective biomarker, which enables AL-stratified cataract risk management—prioritize AQD and LT in the non-high myopia group, and AL and WTW distance in the high myopia group, for earlier nuclear cataract detection before vision loss.
No takes yet. Share an insight, caveat, or question.
Wen et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: