Eye and Vision最新文献

Background: Artificial intelligence (AI) that utilizes deep learning (DL) has potential for systemic disease prediction using retinal imaging. The retina's unique features enable non-invasive visualization of the central nervous system and microvascular circulation, aiding early detection and personalized treatment plans for personalized care. This review explores the value of retinal assessment, AI-based retinal biomarkers, and the importance of longitudinal prediction models in personalized care.

Main text: This narrative review extensively surveys the literature for relevant studies in PubMed and Google Scholar, investigating the application of AI-based retina biomarkers in predicting systemic diseases using retinal fundus photography. The study settings, sample sizes, utilized AI models and corresponding results were extracted and analysed. This review highlights the substantial potential of AI-based retinal biomarkers in predicting neurodegenerative, cardiovascular, and chronic kidney diseases. Notably, DL algorithms have demonstrated effectiveness in identifying retinal image features associated with cognitive decline, dementia, Parkinson's disease, and cardiovascular risk factors. Furthermore, longitudinal prediction models leveraging retinal images have shown potential in continuous disease risk assessment and early detection. AI-based retinal biomarkers are non-invasive, accurate, and efficient for disease forecasting and personalized care.

Conclusion: AI-based retinal imaging hold promise in transforming primary care and systemic disease management. Together, the retina's unique features and the power of AI enable early detection, risk stratification, and help revolutionizing disease management plans. However, to fully realize the potential of AI in this domain, further research and validation in real-world settings are essential.

Background: Near work is generally considered as a risk factor for myopia onset and progression. This study aimed to investigate the choroidal responses to a brief-period of near work in children and young adults.

Methods: Thirty myopic medical students (aged 18-28 years) and 30 myopic children (aged 8-12 years) participated in this study. The submacular total choroidal area (TCA), luminal area (LA), stromal area (SA), choroidal vascularity index (CVI) and choriocapillaris flow deficit (CcFD), as well as subfoveal choroidal thickness (SFCT) were measured with swept-source optical coherence tomography/optical coherence tomography angiography (SS-OCT/OCTA) before and immediately after 20 min, 40 min, 60 min of near work at a distance of 33 cm.

Results: In adults, 20 min of near work induced a significant reduction in SFCT (- 5.1 ± 6.5 μm), LA [(- 19.2 ± 18.6) × 10³ μm²], SA [(- 8.2 ± 12.6) × 10³ μm²] and TCA [(- 27.4 ± 24.9) × 10³ μm²] (all P < 0.01). After 40 min of near work, LA was still reduced [(- 9.4 ± 18.3) × 10³ μm²], accompanied with a decreased CVI (- 0.39% ± 0.70%) and an increased CcFD (0.30% ± 0.78%) (all P < 0.05). After 60 min of near work, CVI was still reduced (- 0.28% ± 0.59%), and CcFD was still increased (0.37% ± 0.75%) (all P < 0.05). In children, 20 min of near work induced a significant increase in CcFD (0.55% ± 0.64%), while 60 min of near work induced increases in SA [(7.2 ± 13.0) × 10³ μm²] and TCA [(9.7 ± 25.3) × 10³ μm²] and a reduction in CVI (- 0.28% ± 0.72%) (all P < 0.05). Children exhibited lower near work-induced LA and TCA reduction than adults, with a mean difference of - 0.86% and - 0.82%, respectively (all P < 0.05).

Conclusions: The temporal characteristics and magnitude of changes of choroidal vascularity and choriocapillaris perfusion during near work was not identical between children and adults. The initial response to near work was observed in choriocapillaris in children, whereas it was observed in the medium- and large-sized vessels in adults.

Trial registration: Clinical Trial Registry (ChiCTR), ChiCTR2000040205. Registered on 25 November 2020, https://www.chictr.org.cn/bin/project/edit?pid=64501 .

Background: To assess clinical outcomes after implanting toric, extended-depth-of-focus intraocular lenses (IOLs) to correct low corneal astigmatism in eyes with cataracts.

Methods: 47 eyes were implanted with the AcrySof IQ Vivity Toric DFT215 IOL. Main outcome measures were refractive error, monocular uncorrected and corrected distance (UDVA/CDVA), uncorrected and distance-corrected intermediate (UIVA/DCIVA), and uncorrected near and distance-corrected near (UNVA/DCNVA) visual acuities, monocular defocus curve, rotational stability, and IOLSAT and QUVID questionnaires. Patients were assessed at 3 months postsurgery.

Results: All eyes had a postoperative spherical equivalent (SE) within ± 0.50 D and 97.87% (n = 46) had a refractive cylinder ≤ 0.50 D. The mean SE and refractive cylinder were - 0.10 ± 0.17 D and - 0.16 ± 0.24 D, respectively. The CDVA was ≥ 20/25 and ≥ 20/32 in 95.74% (n = 45) and 97.87% (n = 46) of eyes, respectively. The DCIVA was ≥ 20/32 in 85.11% (n = 40) of eyes and the DCNVA was ≥ 20/40 in 74.47% (n = 35). The mean values of CDVA, DCIVA, and DCNVA were - 0.02 ± 0.08, 0.14 ± 0.09, and 0.23 ± 0.12 logMAR, respectively. The defocus curve revealed good visual acuity at far and intermediate distances with a depth-of-focus of about 1.75 D. IOL rotation was 0.74 ± 1.13 degrees and all eyes had a rotation of less than 5 degrees. Patients reported either good or very good postoperative vision without eyeglasses under bright-light-conditions at distance (87.80%, 36/41) and intermediate distance (92.68%, 38/41). Between about 63.83%-72.34% (30-34) of patients reported no starburst, halos, or glare, or if experienced, were not bothersome.

Conclusions: The Vivity toric IOL implanted in eyes with low-astigmatism provides accurate refractive outcomes, good visual acuity at different distances and excellent rotational stability. Trial Registration The study was registered with the German Clinical Trials Register (DRKS00030579).

Background: Keratoconus is an ectatic, progressive corneal disorder characterized by alterations in the morphology of the corneal tissue that leads to limitation of visual function of the patient. Intracorneal ring segments (ICRS) are small synthetic devices that are implanted in the corneal stromal in order to regularize the morphology of the tissue therefore improving the visual function and the quality of life of the patients.

Main text: The present narrative review summarizes the main scientific articles developed by the authors in relation to the clinical outcomes and long-term results of ICRS in the treatment of keratoconus. It was found that those patients that benefit the most from this surgical intervention are those that have the most severe form of keratoconus. Additionally, patients with good visual function, those with more than 0.9 in the decimal scale are at risk of losing visual acuity after ICRS implantation. In relation to long-term results, scientific investigations published by the authors demonstrate that ICRS is a stable procedure after long period of time in terms of vision, refraction, and topographic variables in those patients with stable keratoconus. However, in patients with keratoconus and signs of progression, ICRS may not have the capability of halting the progression of the disease. Using artificial intelligence to guide ICRS implantation provide better clinical outcomes and improvement in corneal higher-order aberrations in patients with keratoconus in comparison to those treated using the commercial nomogram of implantation.

Conclusions: ICRS is a safe surgical procedure in the treatment of keratoconus. Patients that benefit most from the surgery are those with a significant visual impairment. ICRS should not be considered in patients with good visual function because of the risk of losing lines of vision. Long-term follow-up demonstrate stability of the clinical outcomes in patients with stable keratoconus although ICRS may not have the ability of halting the progression of the disease. New technologies based artificial intelligence improved the indications and the clinical outcomes of keratoconus patients treated with ICRS.

Background: To describe the diagnostic performance of a deep learning (DL) algorithm in detecting Fuchs endothelial corneal dystrophy (FECD) based on specular microscopy (SM) and to reliably detect widefield peripheral SM images with an endothelial cell density (ECD) > 1000 cells/mm².

Methods: Five hundred and forty-seven subjects had SM imaging performed for the central cornea endothelium. One hundred and seventy-three images had FECD, while 602 images had other diagnoses. Using fivefold cross-validation on the dataset containing 775 central SM images combined with ECD, coefficient of variation (CV) and hexagonal endothelial cell ratio (HEX), the first DL model was trained to discriminate FECD from other images and was further tested on an external set of 180 images. In eyes with FECD, a separate DL model was trained with 753 central/paracentral SM images to detect SM with ECD > 1000 cells/mm² and tested on 557 peripheral SM images. Area under curve (AUC), sensitivity and specificity were evaluated.

Results: The first model achieved an AUC of 0.96 with 0.91 sensitivity and 0.91 specificity in detecting FECD from other images. With an external validation set, the model achieved an AUC of 0.77, with a sensitivity of 0.69 and specificity of 0.68 in differentiating FECD from other diagnoses. The second model achieved an AUC of 0.88 with 0.79 sensitivity and 0.78 specificity in detecting peripheral SM images with ECD > 1000 cells/mm².

Conclusions: Our pilot study developed a DL model that could reliably detect FECD from other SM images and identify widefield SM images with ECD > 1000 cells/mm² in eyes with FECD. This could be the foundation for future DL models to track progression of eyes with FECD and identify candidates suitable for therapies such as Descemet stripping only.

Artificial intelligence (AI) is an emerging field which could make an intelligent healthcare model a reality and has been garnering traction in the field of medicine, with promising results. There have been recent developments in machine learning and/or deep learning algorithms for applications in ophthalmology-primarily for diabetic retinopathy, and age-related macular degeneration. However, AI research in the field of cornea diseases is relatively new. Algorithms have been described to assist clinicians in diagnosis or detection of cornea conditions such as keratoconus, infectious keratitis and dry eye disease. AI may also be used for segmentation and analysis of cornea imaging or tomography as an adjunctive tool. Despite the potential advantages that these new technologies offer, there are challenges that need to be addressed before they can be integrated into clinical practice. In this review, we aim to summarize current literature and provide an update regarding recent advances in AI technologies pertaining to corneal diseases, and its potential future application, in particular pertaining to image analysis.

Background: To evaluate the intraobserver repeatability and interobserver reproducibility of a newly developed dynamic real-time visualization 25 kHz swept-source optical coherence tomography (SS-OCT) based biometer (ZW-30, TowardPi Medical Technology Ltd, China) and compare its agreement with another SS-OCT based biometer (IOLMaster 700, Carl Zeiss Meditec AG, Jena, Germany).

Methods: Eighty-two healthy right eyes were enrolled in this prospective observational study. Measurements were repeated for three times using the ZW-30 and IOLMaster 700 in a random order. Obtained parameters included axial length (AL), central corneal thickness (CCT), aqueous depth (AQD), anterior chamber depth (ACD), lens thickness (LT), mean keratometry (Km), astigmatism magnitude (AST), vector J₀, vector J₄₅, and corneal diameter (CD). The within-subject standard deviation (Sw), test-retest (TRT) variability, coefficient of variation (CoV), and intraclass correlation coefficient (ICC) were adopted to assess the intraobserver repeatability and interobserver reproducibility. The double-angle plot was also used to display the distribution of AST. To estimate agreement, Bland-Altman plots were used.

Results: For the intraobserver repeatability and interobserver reproducibility, the Sw, TRT and CoV for all parameters were low. Meanwhile, the ICC values were all close to 1.000, except for the J₄₅ (ICC = 0.887 for the intraobserver repeatability). The double-angle plot showed that the distribution of AST measured by these two devices was similar. For agreement, the Bland-Altman plots showed narrow 95% limits of agreements (LoAs) for AL, CCT, AQD, ACD, LT, Km AST, J₀, J₄₅, and CD (- 0.02 mm to 0.02 mm, - 7.49 μm to 8.08 μm, - 0.07 mm to 0.04 mm, - 0.07 mm to 0.04 mm, - 0.07 mm to 0.08 mm, - 0.16 D to 0.30 D, - 0.30 D to 0.29 D, - 0.16 D to 0.16 D, - 0.23 D to 0.13 D, and - 0.39 mm to 0.10 mm, respectively).

Conclusions: The newly dynamic real-time visualization biometer exhibited excellent intraobserver repeatability and interobserver reproducibility. The two devices both based on the SS-OCT principle had similar ocular parameters measurement values and can be interchanged in clinical practice.

Background: Corneal cross-linking (CXL) using riboflavin and ultraviolet-A light (UVA) is a treatment used to prevent progression of keratoconus. This ex vivo study assesses the impact on CXL effectiveness, as measured by tissue enzymatic resistance and confocal microscopy, of including a pre-UVA corneal surface rinse with balanced salt solution (BSS) as part of the epithelium-off treatment protocol.

Methods: Sixty-eight porcine eyes, after epithelial debridement, were assigned to six groups in three experimental runs. Group 1 remained untreated. Groups 2-6 received a 16-min application of 0.1% riboflavin/Hydroxypropyl methylcellulose (HPMC) drops, after which Group 3 was exposed to 9 mW/cm² UVA for 10 min, and Groups 4-6 underwent corneal surface rinsing with 0.25 mL, 1 mL or 10 mL BSS followed by 9 mW/cm² UVA exposure for 10 min. Central corneal thickness (CCT) was recorded at each stage. Central 8.0 mm corneal buttons from all eyes were subjected to 0.3% collagenase digestion at 37 °C and the time required for complete digestion determined. A further 15 eyes underwent fluorescence confocal microscopy to assess the impact of rinsing on stromal riboflavin concentration.

Results: Application of riboflavin/HPMC solution led to an increase in CCT of 73 ± 14 µm (P < 0.01) after 16 min. All CXL-treated corneas displayed a 2-4 fold greater resistance to collagenase digestion than non-irradiated corneas. There was no difference in resistance between corneas that received no BSS rinse and those that received a 0.25 mL or 1 mL pre-UVA rinse, but each showed a greater level of resistance than those that received a 10 mL pre-UVA rinse (P < 0.05). Confocal microscopy demonstrated reduced stromal riboflavin fluorescence after rinsing.

Conclusions: All protocols, with and without rinsing, were effective at enhancing the resistance to collagenase digestion, although resistance was significantly decreased, and stromal riboflavin fluorescence reduced with a 10 mL rinse. This suggests that a 10 mL surface rinse can reduce the efficacy of CXL through the dilution of the stromal riboflavin concentration.