CRB1-associated retinal degeneration is dependent on bacterial translocation from the gut.

The Crumbs homolog 1 (CRB1) gene is associated with retinal degeneration, most commonly Leber congenital amaurosis (LCA) and retinitis pigmentosa (RP). Here, we demonstrate that murine retinas bearing the Rd8 mutation of Crb1 are characterized by the presence of intralesional bacteria. While normal CRB1 expression was enriched in the apical junctional complexes of retinal pigment epithelium and colonic enterocytes, Crb1 mutations dampened its expression at both sites. Consequent impairment of the outer blood retinal barrier and colonic intestinal epithelial barrier in Rd8 mice led to the translocation of intestinal bacteria from the lower gastrointestinal (GI) tract to the retina, resulting in secondary retinal degeneration. Either the depletion of bacteria systemically or the reintroduction of normal Crb1 expression colonically rescued Rd8-mutation-associated retinal degeneration without reversing the retinal barrier breach. Our data elucidate the pathogenesis of Crb1-mutation-associated retinal degenerations and suggest that antimicrobial agents have the potential to treat this devastating blinding disease.

Metabolomic profiling of the aqueous humor in patients with pediatric cataract.

Pediatric cataract, including congenital and developmental cataract, is a kind of pediatric vision-threatening disease with extensive phenotypic heterogeneity and multiple mechanisms. We aimed to investigate the metabolite profile of aqueous humor (AH) in patients with pediatric cataracts, and identify underlying mutual correlations between differential metabolites. Metabolomic profiles of AH were analyzed and compared between pediatric cataract patients (n = 33) and age-related cataract patients without metabolic diseases (n = 29), using global untargeted metabolomics with ultra-high-performance liquid chromatography tandem mass spectrometry. Principal component analysis, partial least squares discriminant analysis and heat map were applied. Enriched pathway analysis was conducted using Kyoto Encyclopedia of Genes and Genomes. Receiver-operating characteristic (ROC) analyses were employed to select potential biomarkers. A total of 318 metabolites were identified, of which 54 differential metabolites (25 upregulated and 29 downregulated) were detected in pediatric cataract group compared with controls (variable importance of projection >1.0, fold change ≥1.5 or ≤ 0.667 and P < 0.05). A significant accumulation of N-Acetyl-Dl-glutamic acid was observed in pediatric cataract group. The differential metabolites were mainly enriched in histidine metabolism (increased L-Histidine and decreased 1-Methylhistamine) and the tryptophan metabolism (increased N-Formylkynurenine and L-Kynurenine). 5-Aminosalicylic acid showed strong positive mutual inter-correlation with L-Tyrosinemethylester and N,N-Diethylethanolamine, both of which were down-regulated in pediatric cataract group. The ROC analysis implied 11 metabolites served as potential biomarkers for pediatric cataract patients (all area under the ROC curve ≥0.900). These results illustrated novel potential metabolites and metabolic pathways in pediatric cataract, which provides new insights into the pathophysiology of pediatric cataract.

Privacy preserving technology in ophthalmology.

PURPOSE OF REVIEW: Patient privacy protection is a critical focus in medical practice. Advances over the past decade in big data have led to the digitization of medical records, making medical data increasingly accessible through frequent data sharing and online communication. Periocular features, iris, and fundus images all contain biometric characteristics of patients, making privacy protection in ophthalmology particularly important. Consequently, privacy-preserving technologies have emerged, and are reviewed in this study. RECENT FINDINGS: Recent findings indicate that general medical privacy-preserving technologies, such as federated learning and blockchain, have been gradually applied in ophthalmology. However, the exploration of privacy protection techniques of specific ophthalmic examinations, like digital mask, is still limited. Moreover, we have observed advancements in addressing ophthalmic ethical issues related to privacy protection in the era of big data, such as algorithm fairness and explainability. SUMMARY: Future privacy protection for ophthalmic patients still faces challenges and requires improved strategies. Progress in privacy protection technology for ophthalmology will continue to promote a better healthcare environment and patient experience, as well as more effective data sharing and scientific research.

Nanoporous Gold Ring-Integrated Photothermal Intraocular Lens for Active Prevention of Posterior Capsular Opacification.

Posterior capsular opacification (PCO) is the leading complication after cataract surgery, and is mainly induced by the proliferation and migration of residual lens epithelial cells (LECs). Although numerous attempts have been made to reduce the incidence of PCO, this complication remains a critical challenge in postoperative visual recovery. This study aims to report a functionalized intraocular lens (R-IOL) with a region-confined photothermal effect for the active prevention of PCO after implantation. The outer rim of R-IOL (non-optical area) is decorated with a nanoporous gold (NPG) ring, which can effectively eliminate the LECs around R-IOL, ultimately inhibiting the migration of LECs from the periphery to the visual axis center in the initial stage, and preventing the subsequent PCO. Furthermore, the mechanism of LECs elimination can be attributed to apoptosis induced by mild photothermal therapy. After in vivo implantation for 30 days, PCO is rarely observed in the R-IOL group, whereas the considerably higher incidence of PCO (75%) is found in the pristine IOL (P-IOL) group. The region-confined photothermal effect based on NPG not only provides an active strategy to effectively prevent PCO, but also introduces new opportunities for the treatment of undesirable hyperplasia.

Broadening the genotypic and phenotypic spectrum of MAF in three Chinese Han congenital cataracts families.

Pathogenic variants in the v-maf avian musculoaponeurotic fibrosarcoma oncogene homologue (MAF) encoding a transcription factor (from a unique subclass of basic leucine zipper transcription factors) are associated with isolated congenital cataracts (CCs) and Aymé-Gripp syndrome (AYGRPS). We collected detailed disease histories from, and performed comprehensive ophthalmic and systemic examinations in 269 patients with CCs; we then performed whole-exome sequencing. Pathogenicity assessments were evaluated using multiple predictive tools. The clinical validities of the reported gene-disease relationships for MAF genes (MAF-CCs and MAF-AYGRPS) were assessed using the ClinGen gene curation framework. We identified two novel (c.173C>A, p.Thr58Asn and c.947T>C, p. Leu316Pro) variants and one known (c.173C>T, p.Thr58Ile) MAF missense variant in three patients. We described novel phenotypes including cleft palate, macular hypoplasia, and retinal neovascularization in the peripheral avascular area and analyzed the genotype-phenotype correlations. We demonstrated associations of variants in the MAF C-terminal DNA-binding domain with CCs and associations of variants in the N-terminal transactivation domain of MAF with AYGRPS. We thus expand the genotypic and phenotypic spectrum of the MAF gene. The ClinGen gene curation framework results suggested that variants in different domains of MAF are associated with different diseases.

Extracellular vesicles promote epithelial-to-mesenchymal transition of lens epithelial cells under oxidative stress.

Posterior capsule opacification (PCO), resulting from residual lens epithelial cell (LEC) epithelial-mesenchymal transition (EMT), abnormal proliferation, and migration, is the most common complication of cataract surgery. A recent study determined that extracellular vesicles (EVs) and reactive oxygen species (ROS) regulate the EMT process during cutaneous wound healing and tumour metastasis. However, their underlying mechanism in PCO is unclear. In this study, we examined the secreted EVs from a scratch model in vitro. We found that the production of ROS was increased after mechanical injury, especially at the wound edge, and there was an increased viability of LECs, which can be blocked by diphenyleneiodonium, an NADPH oxidase inhibitor. Cell viability and migration were increased upon treatment with 1 µM H2O2, but significantly reduced when the concentration of H2O2 increased to 100 µM. Transwell assay showed that both post-surgery LECs and LECs treated with 1 µM H2O2 significantly induced the migration of normal LECs by EV secretion. Extraction and quantification of EVs derived from injured and H2O2-treated LECs showed a similar increase in production. Co-incubation of EVs from both injured and H2O2-treated LECs with normal LECs and organ-cultured mouse lenses activated EMT, which was attenuated by a ROS inhibitor. These results suggest that EVs participate in ROS-induced lens EMT, making EVs a potential target for treating PCO.

Lens regeneration using endogenous stem cells with gain of visual function.

The repair and regeneration of tissues using endogenous stem cells represents an ultimate goal in regenerative medicine. To our knowledge, human lens regeneration has not yet been demonstrated. Currently, the only treatment for cataracts, the leading cause of blindness worldwide, is to extract the cataractous lens and implant an artificial intraocular lens. However, this procedure poses notable risks of complications. Here we isolate lens epithelial stem/progenitor cells (LECs) in mammals and show that Pax6 and Bmi1 are required for LEC renewal. We design a surgical method of cataract removal that preserves endogenous LECs and achieves functional lens regeneration in rabbits and macaques, as well as in human infants with cataracts. Our method differs conceptually from current practice, as it preserves endogenous LECs and their natural environment maximally, and regenerates lenses with visual function. Our approach demonstrates a novel treatment strategy for cataracts and provides a new paradigm for tissue regeneration using endogenous stem cells.

Evaluation of integrated modular teaching in Chinese ophthalmology trainee courses.

BACKGROUND: Before attending ophthalmology trainee courses in Zhongshan Ophthalmic Centre, the medical students from Sun Yat-sen University had finished two years of premedical education after the six-year medical courses including basic medical courses, clinical medical courses, clerkship, and research training in medical college. Integrated modular teaching using different problem-based teaching methods in ophthalmology was designed by the teaching steering committee of Zhongshan Ophthalmic Centre. This study aimed to evaluate the effectiveness and satisfaction scales of the integrated modular teaching among the trainee students. METHODS: A total of 100 medical students attending ophthalmology trainee courses in Zhongshan Ophthalmic Centre were enrolled and randomly allocated into 4 groups according to the teaching arrangement. The trainee courses consisted of several sessions delivered in multiple methods, such as "flipped classroom" session and team-based learning session. The pre- and post-class tests were delivered to evaluate the effectiveness of the integrated modular teaching. The satisfaction survey questionnaire was collected from all participants to investigate the degree of satisfaction. RESULTS: Compared with the first-day-test score, the total last-day-test score was significantly improved by a paired t-test (t = 3.288, P = 0.001). Nineteen students obtained a significant improvement in ranking increased by more than 10 in the last-day-test, whereas they failed to obtain a higher average score for daily performance than other students (t = 0.469, P = 0.654). According to the participant satisfaction questionnaires, these innovative teaching methods were considered as effective and satisfactory. CONCLUSIONS: Integrated modular teaching in ophthalmology trainee courses is effective and appreciated by the medical college students.

Loss-of-function mutations in FREM2 disrupt eye morphogenesis.

Cryptophthalmos is a rare congenital disorder characterized by ocular dysplasia with eyelid malformation. Complete cryptophthalmos is characterized by the presence of continuous skin from the forehead over the eyes and onto the cheek, along with complete fusion of the eyelids. In the present study, we characterized the clinical manifestations of three patients with isolated bilateral cryptophthalmos. These patients shared the same c.6499C > T missense mutation in the FRAS1-related extracellular matrix protein 2 (FREM2) gene, while each individual presented an additional nonsense mutation in the same gene (Patient #1, c.2206C > T; Patient #2, c.5309G > A; and Patient #3, c.4063C > T). Then, we used CRISPR/Cas9 to generate mice carrying Frem2R725X/R2156W compound heterozygous mutations, and showed that these mice recapitulated the human isolated cryptophthalmos phenotype. We detected FREM2 expression in the outer plexiform layer of the retina for the first time in the cryptophthalmic eyes, and the levels were comparable to the wild-type mice. Moreover, a set of different expressed genes that may contribute secondarily to the phenotypes were identified by performing RNA sequencing (RNA-seq) of the fetal Frem2 mutant mice. Our findings extend the spectrum of FREM2 mutations, and provide insights into opportunities for the prenatal diagnosis of isolated cryptophthalmos. Furthermore, our work highlights the importance of the FREM2 protein during the development of eyelids and the anterior segment of the eyeballs, establishes a suitable animal model for studying epithelial reopening during eyelid development and serves as a valuable reference for further mechanistic studies of the pathogenesis of isolated cryptophthalmos.

Psychosocial Factors Affecting Artificial Intelligence Adoption in Health Care in China: Cross-Sectional Study.

BACKGROUND: Poor quality primary health care is a major issue in China, particularly in blindness prevention. Artificial intelligence (AI) could provide early screening and accurate auxiliary diagnosis to improve primary care services and reduce unnecessary referrals, but the application of AI in medical settings is still an emerging field. OBJECTIVE: This study aimed to investigate the general public's acceptance of ophthalmic AI devices, with reference to those already used in China, and the interrelated influencing factors that shape people's intention to use these devices. METHODS: We proposed a model of ophthalmic AI acceptance based on technology acceptance theories and variables from other health care-related studies. The model was verified via a 32-item questionnaire with 7-point Likert scales completed by 474 respondents (nationally random sampled). Structural equation modeling was used to evaluate item and construct reliability and validity via a confirmatory factor analysis, and the model's path effects, significance, goodness of fit, and mediation and moderation effects were analyzed. RESULTS: Standardized factor loadings of items were between 0.583 and 0.876. Composite reliability of 9 constructs ranged from 0.673 to 0.841. The discriminant validity of all constructs met the Fornell and Larcker criteria. Model fit indicators such as standardized root mean square residual (0.057), comparative fit index (0.915), and root mean squared error of approximation (0.049) demonstrated good fit. Intention to use (R2=0.515) is significantly affected by subjective norms (beta=.408; P<.001), perceived usefulness (beta=.336; P=.03), and resistance bias (beta=-.237; P=.02). Subjective norms and perceived behavior control had an indirect impact on intention to use through perceived usefulness and perceived ease of use. Eye health consciousness had an indirect positive effect on intention to use through perceived usefulness. Trust had a significant moderation effect (beta=-.095; P=.049) on the effect path of perceived usefulness to intention to use. CONCLUSIONS: The item, construct, and model indicators indicate reliable interpretation power and help explain the levels of public acceptance of ophthalmic AI devices in China. The influence of subjective norms can be linked to Confucian culture, collectivism, authoritarianism, and conformity mentality in China. Overall, the use of AI in diagnostics and clinical laboratory analysis is underdeveloped, and the Chinese public are generally mistrustful of medical staff and the Chinese medical system. Stakeholders such as doctors and AI suppliers should therefore avoid making misleading or over-exaggerated claims in the promotion of AI health care products.

Prediction of myopia development among Chinese school-aged children using refraction data from electronic medical records: A retrospective, multicentre machine learning study.

BACKGROUND: Electronic medical records provide large-scale real-world clinical data for use in developing clinical decision systems. However, sophisticated methodology and analytical skills are required to handle the large-scale datasets necessary for the optimisation of prediction accuracy. Myopia is a common cause of vision loss. Current approaches to control myopia progression are effective but have significant side effects. Therefore, identifying those at greatest risk who should undergo targeted therapy is of great clinical importance. The objective of this study was to apply big data and machine learning technology to develop an algorithm that can predict the onset of high myopia, at specific future time points, among Chinese school-aged children. METHODS AND FINDINGS: Real-world clinical refraction data were derived from electronic medical record systems in 8 ophthalmic centres from January 1, 2005, to December 30, 2015. The variables of age, spherical equivalent (SE), and annual progression rate were used to develop an algorithm to predict SE and onset of high myopia (SE ≤ -6.0 dioptres) up to 10 years in the future. Random forest machine learning was used for algorithm training and validation. Electronic medical records from the Zhongshan Ophthalmic Centre (a major tertiary ophthalmic centre in China) were used as the training set. Ten-fold cross-validation and out-of-bag (OOB) methods were applied for internal validation. The remaining 7 independent datasets were used for external validation. Two population-based datasets, which had no participant overlap with the ophthalmic-centre-based datasets, were used for multi-resource validation testing. The main outcomes and measures were the area under the curve (AUC) values for predicting the onset of high myopia over 10 years and the presence of high myopia at 18 years of age. In total, 687,063 multiple visit records (≥3 records) of 129,242 individuals in the ophthalmic-centre-based electronic medical record databases and 17,113 follow-up records of 3,215 participants in population-based cohorts were included in the analysis. Our algorithm accurately predicted the presence of high myopia in internal validation (the AUC ranged from 0.903 to 0.986 for 3 years, 0.875 to 0.901 for 5 years, and 0.852 to 0.888 for 8 years), external validation (the AUC ranged from 0.874 to 0.976 for 3 years, 0.847 to 0.921 for 5 years, and 0.802 to 0.886 for 8 years), and multi-resource testing (the AUC ranged from 0.752 to 0.869 for 4 years). With respect to the prediction of high myopia development by 18 years of age, as a surrogate of high myopia in adulthood, the algorithm provided clinically acceptable accuracy over 3 years (the AUC ranged from 0.940 to 0.985), 5 years (the AUC ranged from 0.856 to 0.901), and even 8 years (the AUC ranged from 0.801 to 0.837). Meanwhile, our algorithm achieved clinically acceptable prediction of the actual refraction values at future time points, which is supported by the regressive performance and calibration curves. Although the algorithm achieved balanced and robust performance, concerns about the compromised quality of real-world clinical data and over-fitting issues should be cautiously considered. CONCLUSIONS: To our knowledge, this study, for the first time, used large-scale data collected from electronic health records to demonstrate the contribution of big data and machine learning approaches to improved prediction of myopia prognosis in Chinese school-aged children. This work provides evidence for transforming clinical practice, health policy-making, and precise individualised interventions regarding the practical control of school-aged myopia.

Preoperative profile of inflammatory factors in aqueous humor correlates with postoperative inflammatory response in patients with congenital cataract.

Purpose: To measure the aqueous humor concentrations of inflammatory factors in patients with congenital cataract and to investigate the relationship between the levels and postoperative inflammatory responses. Methods: Aqueous humor samples were prospectively collected from 65 eyes of children with congenital cataracts from January to June 2015. The levels of 41 inflammation-related cytokines, chemokines, and growth factors in aqueous humor were measured using multiplex bead immunoassay. Data on patient demographics and postoperative inflammatory response evaluation of posterior capsule opacification (EPCO) scores were collected for correlation analysis of short- and long-term postoperative inflammatory responses, respectively. Results: Fifteen inflammatory factors were differentially expressed between congenital cataract and age-related cataract. EGF and IL-3 were positively correlated, whereas IL-8 and MCP-1 were negatively correlated with age. TNFα, IL-17A, IL-3, and sCD40L were preferably expressed in specific morphological types of congenital cataract. One month and 3 months postoperatively, PDGF-AA exhibited a positive correlation with the EPCO scores, whereas IL-1RA exhibited a negative correlation. Macrophage-derived chemokine (MDC) showed a positive correlation with the EPCO scores 1 year postoperatively. Conclusions: This study provided a comprehensive preoperative profile of inflammatory factors and their correlations with postoperative inflammatory responses in patients with congenital cataract. These factors may serve as potential biomarkers to predict the postoperative inflammatory response. These findings will also facilitate the development of anti-inflammatory medications in the perioperative period.

Accuracy of intraocular lens power calculation formulas in long eyes: a systematic review and meta-analysis.

IMPORTANCE: Visual outcome after intraocular lens (IOL) implantation in long eyes is considerably affected by IOL power calculation. Various formulas have been designed to achieve an accurate IOL power prediction. However, controversy about the accuracy remains. BACKGROUND: To evaluate the accuracy of IOL power calculation formulas in long eyes. DESIGN: Meta-analysis. PARTICIPANTS: Patients with ocular axial length (AL) over 24.5 mm. METHODS: A comprehensive search in PubMed, EMBASE, Cochrane Data Base of Systematic Reviews and the Cochrane Central Register of Controlled Trials were conducted by September, 2017. The weighted mean differences of mean absolute errors (MAE) and the odds ratio of percentage of eyes within ±0.50D of prediction error among formulas were analysed. MAIN OUTCOMES MEASURES: Between-group differences of MAE among formulas. RESULTS: Eleven observational studies, involving 4047 eyes, were enrolled. Six formulas for IOL power calculation were compared: Barrett Universal II, Haigis, Holladay 2, SRK/T, Hoffer Q and Holladay 1. The MAE of Barrett Universal II was statistically lower than that of Holladay 2 (mean difference, MD = -0.04D, P = 0.0002), SRK/T (MD = -0.05D, P < 0.00001), Hoffer Q (MD = -0.07D, P < 0.00001) and Holladay 1 (MD = -0.07D, P < 0.00001). Barrett Universal II yielded significantly higher percentage of eyes within ±0.50D of the prediction error than the other formulas. The heterogeneity was minimized through dividing eyes into two groups by the AL of 26 mm. CONCLUSIONS AND RELEVANCE: This study demonstrates the superiority of Barrett Universal II over Holladay 2, SRK/T, Hoffer Q and Holladay 1 in predicting IOL power in long eyes.

Meta-analysis of accuracy of intraocular lens power calculation formulas in short eyes.

IMPORTANCE: Intraocular lens (IOL) power selection is a critical factor affecting visual outcome after IOL implantation in short eyes. Many formulas have been developed to achieve a precise prediction of the IOL power. However, controversy regarding the accuracy remains. BACKGROUND: To investigate the accuracy of different IOL power calculation formulas in short eyes. DESIGN: Meta-analysis. PARTICIPANTS: Patients with the axial length of eyes less than 22 mm from previously reported studies. METHODS: A comprehensive search in Pubmed, EMBASE, Cochrane Data Base of Systematic Reviews and the Cochrane Central Register of Controlled Trials was conducted by October 2016. We assessed the methodological quality using a modified QUADAS-2 tool and performed analysis on weighted mean differences of mean absolute errors (MAE) among different formulas. MAIN OUTCOMES MEASURES: The between-group difference of MAE was evaluated with weighted mean difference and 95% confidence intervals. RESULTS: Ten observational studies, involving 1161 eyes, were enrolled to compare six formulas: Haigis, Holladay 2, Hoffer Q, Holladay 1, SRK/T and SRK II. Among them, the Holladay 2 introduced the smallest overall MAE (0.496D) without statistical significance. The difference of MAE is statistically significant between Haigis and Hoffer Q (mean difference = -0.07D, P = 0.003), Haigis and SRK/T (mean difference = -0.07D, P = 0.009), Haigis and SRK II (mean difference = -0.41D, P = 0.01). For publication bias and small-study effect, neither funnel plot nor egger's test detected statistical finding. CONCLUSION AND RELEVANCE: The overall evidence from the studies confirmed the superiority of Haigis over Hoffer Q, SRK/T and SRK II in prediction IOL power in short eyes.

An Interpretable and Expandable Deep Learning Diagnostic System for Multiple Ocular Diseases: Qualitative Study.

BACKGROUND: Although artificial intelligence performs promisingly in medicine, few automatic disease diagnosis platforms can clearly explain why a specific medical decision is made. OBJECTIVE: We aimed to devise and develop an interpretable and expandable diagnosis framework for automatically diagnosing multiple ocular diseases and providing treatment recommendations for the particular illness of a specific patient. METHODS: As the diagnosis of ocular diseases highly depends on observing medical images, we chose ophthalmic images as research material. All medical images were labeled to 4 types of diseases or normal (total 5 classes); each image was decomposed into different parts according to the anatomical knowledge and then annotated. This process yields the positions and primary information on different anatomical parts and foci observed in medical images, thereby bridging the gap between medical image and diagnostic process. Next, we applied images and the information produced during the annotation process to implement an interpretable and expandable automatic diagnostic framework with deep learning. RESULTS: This diagnosis framework comprises 4 stages. The first stage identifies the type of disease (identification accuracy, 93%). The second stage localizes the anatomical parts and foci of the eye (localization accuracy: images under natural light without fluorescein sodium eye drops, 82%; images under cobalt blue light or natural light with fluorescein sodium eye drops, 90%). The third stage carefully classifies the specific condition of each anatomical part or focus with the result from the second stage (average accuracy for multiple classification problems, 79%-98%). The last stage provides treatment advice according to medical experience and artificial intelligence, which is merely involved with pterygium (accuracy, >95%). Based on this, we developed a telemedical system that can show detailed reasons for a particular diagnosis to doctors and patients to help doctors with medical decision making. This system can carefully analyze medical images and provide treatment advices according to the analysis results and consultation between a doctor and a patient. CONCLUSIONS: The interpretable and expandable medical artificial intelligence platform was successfully built; this system can identify the disease, distinguish different anatomical parts and foci, discern the diagnostic information relevant to the diagnosis of diseases, and provide treatment suggestions. During this process, the whole diagnostic flow becomes clear and understandable to both doctors and their patients. Moreover, other diseases can be seamlessly integrated into this system without any influence on existing modules or diseases. Furthermore, this framework can assist in the clinical training of junior doctors. Owing to the rare high-grade medical resource, it is impossible that everyone receives high-quality professional diagnosis and treatment service. This framework can not only be applied in hospitals with insufficient medical resources to decrease the pressure on experienced doctors but also deployed in remote areas to help doctors diagnose common ocular diseases.

Dynamic response to initial stage blindness in visual system development.

Sensitive periods and experience-dependent plasticity have become core issues in visual system development. Converging evidence indicates that visual experience is an indispensable factor in establishing mature visual system circuitry during sensitive periods and the visual system exhibits substantial plasticity while facing deprivation. The mechanisms that underlie the environmental regulation of visual system development and plasticity are of great interest but need further exploration. Here, we investigated a unique sample of human infants who experienced initial stage blindness (beginning at birth and lasting for 2-8 months) before the removal of bilateral cataracts. Retinal thickness (RT), axial length (AL), refractive status, visual grating acuity and genetic integrity were recorded during the preoperative period or at surgery and then during follow-up. The results showed that the development of the retina is malleable and associated with external environmental influences. Our work supported that the retina might play critical roles in the development of the experience-dependent visual system and its malleability might partly contribute to the sensitive period plasticity.

Automatic diagnosis of imbalanced ophthalmic images using a cost-sensitive deep convolutional neural network.

BACKGROUND: Ocular images play an essential role in ophthalmological diagnoses. Having an imbalanced dataset is an inevitable issue in automated ocular diseases diagnosis; the scarcity of positive samples always tends to result in the misdiagnosis of severe patients during the classification task. Exploring an effective computer-aided diagnostic method to deal with imbalanced ophthalmological dataset is crucial. METHODS: In this paper, we develop an effective cost-sensitive deep residual convolutional neural network (CS-ResCNN) classifier to diagnose ophthalmic diseases using retro-illumination images. First, the regions of interest (crystalline lens) are automatically identified via twice-applied Canny detection and Hough transformation. Then, the localized zones are fed into the CS-ResCNN to extract high-level features for subsequent use in automatic diagnosis. Second, the impacts of cost factors on the CS-ResCNN are further analyzed using a grid-search procedure to verify that our proposed system is robust and efficient. RESULTS: Qualitative analyses and quantitative experimental results demonstrate that our proposed method outperforms other conventional approaches and offers exceptional mean accuracy (92.24%), specificity (93.19%), sensitivity (89.66%) and AUC (97.11%) results. Moreover, the sensitivity of the CS-ResCNN is enhanced by over 13.6% compared to the native CNN method. CONCLUSION: Our study provides a practical strategy for addressing imbalanced ophthalmological datasets and has the potential to be applied to other medical images. The developed and deployed CS-ResCNN could serve as computer-aided diagnosis software for ophthalmologists in clinical application.