Search

PURPOSE To establish and validate an artificial intelligence (AI)-assisted automatic cataract grading program based on the Lens Opacities Classification System III (LOCSIII). SETTING Eye and Ear, Nose, and Throat (EENT) Hospital, Fudan University, Shanghai, China. DESIGN AI training. METHODS Advanced deep-learning algorithms, including Faster R-CNN and ResNet, were applied to the localization and analysis of the region of interest. An internal dataset from the EENT Hospital of Fudan University and an external dataset from the Pujiang Eye Study were used for AI training, validation, and testing. The datasets were automatically labeled on the AI platform in terms of the capture mode and cataract grading based on the LOCSIII system. RESULTS The AI program showed reliable capture mode recognition, grading, and referral capability for nuclear and cortical cataract grading. In the internal and external datasets, 99.4% and 100% of automatic nuclear grading, respectively, had an absolute prediction error of ≤ 1.0, with a satisfactory referral capability (area under the curve [AUC]: 0.983 for the internal dataset; 0.977 for the external dataset). 75.0% (internal dataset) and 93.5% (external dataset) of the automatic cortical grades had an absolute prediction error of ≤ 1.0, with AUCs of 0.855 and 0.795 for referral, respectively. Good consistency was observed between automatic and manual grading when both nuclear and cortical cataracts were evaluated. However, automatic grading of posterior subcapsular cataracts was impractical. CONCLUSIONS The AI program proposed in this study shows robust grading and diagnostic performance for both nuclear and cortical cataracts, based on LOCSIII.

PURPOSE To compare the rotational stability of a plate-haptic toric intraocular lens (IOL) vs a C-loop haptic toric IOL in myopic cataract eyes. SETTING Eye and Ear, Nose, and Throat Hospital of Fudan University, China. DESIGN Prospective, randomized, controlled study. METHODS Cataract eyes with axial length (AL) more than 24.5 mm were randomly assigned to receive implantation of a C-loop haptic toric IOL (AcrySof Toric IOL) (Group A) or a plate-haptic toric IOL (AT TORBI 709M IOL) (Group B). IOL rotation, residual astigmatism, visual acuity, and higher-order aberrations (HOAs) evaluated with OPD-Scan III aberrometer were compared at 3 months postoperatively. RESULTS In total, 62 eyes of 62 patients were eligible for analysis: 31 in Group A and 31 in Group B. The mean rotation of toric IOLs was greater in Group A than that in Group B (8.00 ± 3.60 degrees vs 4.42 ± 3.24 degrees, respectively, P < .001), especially when IOLs were vertically placed. IOL rotation was positively correlated with AL in Group A, whereas no such correlations were found in Group B. Residual astigmatism in Group A was greater than that in Group B (-0.76 ± 0.30 diopter [D] vs -0.51 ± 0.29 D, respectively, P = .001). Fewer eyes achieved residual astigmatism of 0.50 D or less in Group A than in Group B (38.71% vs 64.52%). Group A had worse postoperative uncorrected visual acuity and higher total HOAs and coma for a 6.0 mm pupil than Group B, whereas postoperative corrected visual acuity was not different between the 2 groups. CONCLUSIONS The plate-haptic toric IOL might be a better choice for myopic cataract eyes with corneal astigmatism because of reduced postoperative rotation.

PURPOSE To describe the distribution of angle α and angle κ in a population with cataract in Shanghai. SETTING Eye and Ear, Nose, Throat Hospital, Fudan University, Shanghai, China. DESIGN Hospital-based, cross-sectional study. METHODS Angle α, angle κ, and other ocular biometric parameters were determined by IOLMaster 700. The distributions of angle α and angle κ and their associations with systemic and ocular parameters were assessed. RESULTS This study included 15 127 eyes of 15 127 cataract patients. The mean angle α and angle κ values were 0.45 ± 0.21 mm and 0.30 ± 0.18 mm, respectively. Angle α and angle κ were both predominantly located temporal to the visual axis. A greater angle α or angle κ was associated with older age, lower corneal power, shorter white-to-white distance, and shallower anterior chamber depth (all P < .05). Angle α correlated positively with angle κ. With increasing axial length (AL), angle α gradually decreased in a nonlinear way and shifted to the nasal side of the visual axis, whereas angle κ decreased in eyes with AL less than 27.5 mm but increased again in eyes with longer AL. CONCLUSIONS Angle α and angle κ, both predominantly located temporal to the visual axis, were influenced by multiple anterior segment parameters. As AL increased, the changes in angle α and angle κ were nonlinear, and their locations gradually shifted from the temporal to the nasal side of the visual axis.

Purpose To assess the effect of preoperative biometry fixation stability on postoperative refractive errors in highly myopic cataractous eyes. Setting Eye and ENT Hospital of Fudan University, Shanghai, China. Design Prospective cohort study. Methods Eyes of highly myopic patients and emmetropic controls were included. Routine ophthalmologic examinations and measurement of fixation stability in the 63% and 95% bivariate contour ellipse areas (BCEAs) were conducted preoperatively. The refractive error from prediction was calculated 1 month postoperatively with the SRK/T and Holladay 1 formulas. Univariate and multivariable analyses were performed to identify the factors associated with postoperative refractive errors. Results The refractive errors were more widely distributed in the 45 highly myopic eyes than in the 40 emmetropic control eyes: SRK/T, +0.15 diopter [D] ± 0.80 [SD] and −0.16 ± 0.35 D, respectively; Holladay 1, +0.54 ± 0.79 D and −0.23 ± 0.34 D, respectively. In the highly myopic group, 63% BCEA was correlated with axial length (AL) (P = .021) and posterior subcapsular opacity grade (P = .040). With both formulas, refractive errors and absolute refractive errors were positively correlated with 63% BCEA: SRK/T, P = .010 and P = .001, respectively; Holladay 1, P = .006 and P = .003, respectively. Backward multiple linear regression analysis showed that with both formulas, AL and 63% BCEA were significantly associated with postoperative refractive errors. Conclusion Poor preoperative biometry fixation stability correlated with long AL and severe posterior subcapsular opacity contributed to significant deviation of refractive errors after cataract surgery in highly myopic eyes.

Purpose To quantify corneal central epithelial thickness and cell changes 2 weeks after cataract surgery and correlate it with corrected distance visual acuity (CDVA). Setting Eye and ENT Hospital of Fudan University, Shanghai, China. Design Prospective observational study. Methods Patients having cataract surgery were assessed preoperatively and 1, 2, 3, 5, 7, and 14 days postoperatively. Anterior segment spectral-domain optical coherence tomography and corneal confocal laser scanning microscopy were used to measure central epithelial thickness, central corneal thickness (CCT), corneal basal epithelial cell density, and Langerhans cell density. Results The CCT and central epithelial thickness were significantly increased on 1 day postoperatively. From 1 day to 14 days, the CCT gradually decreased but was greater than the baseline value, whereas the central epithelial thickness declined to near baseline values at 3 days, became thinner at 5 and 7 days, and then returned to baseline at 14 days. The corneal basal epithelial cell density increased significantly postoperatively. Significant Langerhans cell infiltration was seen at 3 days and 5 days. Significant correlations were found between CDVA and the central epithelial thickness and CCT. The maximum increase in central epithelial thickness and CCT correlated with the lowest achieved values of CDVA postoperatively. Patients with significant edema at 1 day had worse CDVA at 14 days. Conclusions Near-term epithelial remodeling occurred, initially appearing as a thickening that might be attributed to edema. This was followed by thinning, possibly caused by inflammation, and finally reaching baseline levels. Edema of the corneal epithelium was related to visual recovery after cataract surgery. Financial Disclosure None of the authors has a financial or proprietary interest in any material or method mentioned.

Purpose To compare the accuracy of the Barrett Universal II, Haigis, and Olsen formulas in calculating intraocular lens (IOL) power in eyes with extreme myopia. Setting Eye and Ear, Nose, and Throat Hospital, Fudan University, Shanghai, China. Design Prospective case series. Methods Eyes were divided into 3 axial length (AL) groups as follows: 26.0 to 28.0 mm (control), 28.0 to 30.0 mm (extreme myopia 1), and 30.0 mm or more (extreme myopia 2). The mean error (ME) 1 month postoperatively was adjusted to zero by optimizing the lens factor; then, the median absolute errors (MedAEs) were compared between formulas. Factors associated with postoperative refractive errors were analyzed. Results After optimization, the MEs of the Barrett Universal II, Haigis, and Olsen formulas were 0.04 diopter (D) ± 0.48 (SD), 0.04 ± 0.66 D, and 0.04 ± 0.52 D, respectively, and the MedAEs were 0.37 D, 0.46 D, and 0.39 D, respectively (P = .044; Haigis versus Barrett: P = .038). In the extreme myopia 1 group, all 3 formulas produced small MedAEs (P = .662). In the extreme myopia 2 group, the Haigis formula produced a significantly greater MedAE than the Barrett Universal II formula (P = .007; Haigis versus Olsen: P = .055). The accuracy of the Haigis formula in myopic eyes was affected by the AL and keratometry value, whereas the accuracy of the Barrett Universal II and Olsen formulas was affected by the AL only. Conclusions In eyes with an AL of 28.0 to 30.0 mm, all 3 formulas were accurate. In eyes with AL of 30.0 mm or more, the Barrett Universal II formula was better than the Haigis formula, possibly because there were fewer influencing factors.