Artificial Intelligence-Assisted Prescription Determination for Orthokeratology Lens Fitting: From Algorithm to Clinical Practice.

OBJECTIVES: To explore the potential of artificial intelligence (AI) to assist prescription determination for orthokeratology (OK) lenses. METHODS: Artificial intelligence algorithm development followed by a real-world trial. A total of 11,502 OK lenses fitting records collected from seven clinical environments covering major brands. Records were randomly divided in a three-way data split. Cross-validation was used to identify the most accurate algorithm, followed by an evaluation using an independent test data set. An online AI-assisted system was implemented and assessed in a real-world trial involving four junior and three senior clinicians. RESULTS: The primary outcome measure was the algorithm's accuracy (ACC). The ACC of the best performance of algorithms to predict the targeted reduction amplitude, lens diameter, and alignment curve of the prescription was 0.80, 0.82, and 0.83, respectively. With the assistance of the AI system, the number of trials required to determine the final prescription significantly decreased for six of the seven participating clinicians (all P <0.01). This reduction was more significant among junior clinicians compared with consultants (0.76±0.60 vs. 0.32±0.60, P <0.001). Junior clinicians achieved clinical outcomes comparable to their seniors, as 93.96% (140/149) and 94.44% (119/126), respectively, of the eyes fitted achieved unaided visual acuity no worse than 0.8 ( P =0.864). CONCLUSIONS: AI can improve prescription efficiency and reduce discrepancies in clinical outcomes among clinicians with differing levels of experience. Embedment of AI in practice should ultimately help lessen the medical burden and improve service quality for myopia boom emerging worldwide.

The efficacy of a device-based approach to microorganism disinfection and protein removal for orthokeratology lenses in varied clinical circumstances.

PURPOSE: RigidCare is an electrolysis-based device that recently obtained approval from the US's FDA to sterilise microorganisms and remove proteins for orthokeratology (O-K) lenses. The study was conducted to investigate the device's performance in varied clinical circumstances. METHODS: Trial lenses and private lenses were employed by O-K lens wearers from five hospitals for an evaluation of disinfection and sterilisation and an assessment of protein removal, respectively. Menicon multipurpose solution and protein remover were selected for use with the control group. Following the instructions, pre-cleaning lens samples, post-cleaning lens samples and residual solution samples of trial lenses of the experimental and control groups were collected for microorganism examinations by an experienced third-party testing organisation. The levels of protein deposition for these two approaches were rated by senior O-K experts. Categorical variables were analysed using statistical tests, such as the chi-squared test and Fisher's exact test. RESULTS: The microbial positive rate detected from the pre-cleaning and post-cleaning lens samples and the residual solution of the trial lenses for the experimental and control group was 4/76 vs 1/74 (P = 0.37), 1/76 vs 0/74 (P = 1.00) and 0/76 vs 8/74 (P = 0.006), respectively. Following protein removal, the experimental group exhibited a significantly higher overall proportion of lenses rated as 'clean' or with a 'mild deposit' (96.4 %, 79/82) compared to the control group (85.7 %, 66/77), with a significant difference (P < 0.05). CONCLUSION: This multi-center study demonstrated that RigidCare exhibited superior efficacy in disinfection, sterilisation and protein removal as compared to Menicon multipurpose solution and protein remover.

[The strategy of determining the customized near addition lens for juvenile esophoric myopes].

OBJECTIVE: To discuss accommodative lag and near phoria varying with the near addition lenses and establish the mathematic equations of near adds and accommodative lag, near phoria respectively in order to getting the optimal adds, which would create the least error in accommodative and convergence responses. METHODS: To measure accommodative lag by Shin-Nippon and near phoria by Cover Test in 33 cm after providing 0D, +0.75D, +1.50D, +2.00D, +2.50D five near adds orderly to myopes after 110 simple myopes with esophoria have been corrected far vision wholly basing on MPMVA norm; to analyze the relation of near adds and accommodative lag, near phoria by linearity regression. RESULTS: While increasing dioptric value of the near adds, accommodative lag and esophoria decrease gradually. Both accommodative lag [F(262.76, 4), P < 0.01]and near phoria [F(223.45, 4), P < 0.01] show statistically significant difference among the five near adds. There is linearity correlation between the optimal adds and initial accommodative lag (r = 0.526, P < 0.01) and between the optimal adds and initial near phoria (r = 0.523, P < 0.01). CONCLUSION: The range of optimal near adds can be gotten basing on initial accommodative lag and near phoria of myopic individual, so that it can be provided referenced strategy of determining the customized near addition lens for juvenile esophoric myopes.

[New progresses of glaucoma research in China and abroad].

Great progresses have recently been made in glaucoma fields, including basic research and clinical management. This review covers major areas in the study of glaucoma, including epidemiology, pathogenesis, integrated imaging technology, medicine, laser and surgical therapy of glaucoma, as well as the new trends of development. This paper provides the future direction in the research of glaucoma.

Effects of the long wavelength-filtered continuous spectrum on natural refractive development in juvenile guinea pigs.

AIM: To investigate the effects of spectral composition and light intensity on natural refractive development in guinea pigs. METHODS: A total of 124 pigmented guinea pigs (2-week-old) were randomly assigned to three groups at high (Hi; 4000 lx), medium (Me; 400 lx) and low (Lo; 50 lx) light intensities under a 12:12 light/dark cycle for 6wk. Each group was subdivided into subgroups with the following spectra: broad spectrum Solux halogen light (BS), 600 nm above-filtered continuous spectrum (600F), 530 nm above-filtered continuous spectrum (530F), and 480 nm above-filtered continuous spectrum (480F; HiBS: n=10, Hi600F: n=10, Hi530F: n=10, Hi480F: n=10, MeBS: n=10, Me600F: n=10, Me530F: n=10, Me480F: n=10, LoBS: n=11, Lo600F: n=12, Lo530F: n=10, Lo480F: n=11). Refractive error, corneal curvature radius, and axial dimensions were determined by cycloplegic retinoscopy, photokeratometry, and A-scan ultrasonography before and after 2, 4, and 6wk of treatment. Average changes from both eyes in the ocular parameters and refractive error were compared among different subgroups. RESULTS: After 6wk of exposure, high-intensity lighting enhanced hyperopic shift; medium- and low-intensity lighting enhanced myopic shift (P<0.05). Under the same spectrum, axial increase was larger in the low light intensity group than in the medium and high light intensity groups (HiBS: 0.65±0.02 mm, MeBS: 0.67±0.01 mm, LoBS: 0.82±0.02 mm; Hi600F: 0.64±0.02 mm, Me600F: 0.67±0.01 mm, Lo600F: 0.81±0.01 mm; Hi530F: 0.64±0.02 mm, Me530F: 0.67±0.01 mm, Lo530F: 0.73±0.02 mm; Hi480F: 0.64±0.01 mm, Me480F: 0.66±0.01 mm, Lo480F: 0.72±0.02 mm; P<0.05). Under 400 lx, there was a faster axial increase in the MeBS group than in the Me480F group (P<0.05). Under 50 lx, axial length changes were significantly larger in LoBS and Lo600F than in Lo530F and Lo480F (P<0.01). CONCLUSION: Under high-intensity lighting, high light intensity rather than spectrum distributions that inhibits axial increase. Under medium- and low-intensity lighting, filtering out the long wavelength inhibits axial growth in juvenile guinea pigs.