Medical imaging technology plays an important role in the screening, evaluation, diagnosis, and monitoring of fundus diseases. In recent years, the development of multi-modality imaging fusion technology, image databases, standardized protocols and image analysis have extracted large amounts of information from imagebased features. Artificial intelligence applications in fundus imaging will bring new opportunities for increasing the rate of disease diagnosis, accelerating the development of new drugs, promoting research and development, improving diagnosis and treatment, and judging the prognosis of patients. This review covers the research status, difficulties and trends of the combination of artificial intelligence and fundus imaging, as well as taking China's national status into consideration, with the goal of presenting an objective update on relevant topics and new prospects in this field.

Objective: To study the correlation between retinoschisis and different types of fundus disease in high myopia and evaluate the different types of morphologic characteristics in highly myopic eyes. Methods:In this retrospective study, 116 patients (174 eyes) with high myopia from the outpatient department were included from October 2015 to December 2017. All participants underwent standardized medical interviews and ophthalmic examinations. An independent samples t test or one-way ANOVA were used to compare the measurement data. A chi square test (Fisher's exact probability method) was used to compare the numerical data. Results: There were five types of fundus disease shown on the fundus photographs: C0, no macular lesions in 10 eyes; C1, tessellated fundus in 74 eyes; C2, diffuse chorioretinal atrophy in 36 eyes; C3, patchy chorioretinal atrophy in 25 eyes and C4, macular atrophy in 29 eyes. Retinoschisis was detected in 70 eyes (40%) among eyes with C1, 2, 3, and 4, and the incidence of retinoschisis was highest in C2 (58%). Conclusions: The International Classification based on fundus color photography combined with OCT shows the highest prevalence of retinoschisis in highly myopia patients with diffuse chorioretinal atrophy

Objective: To observe the morphological changes in choroidal neovascularization (CNV) after anti-vascular endothelial growth factor (VEGF) therapy by using optical coherence tomography angiography (OCTA), and to discuss the advantages of OCTA in diagnosing CNV and judging prognoses. Methods: Seventeen patients (seventeen eyes) diagnosed with CNV in Qingdao Eye Hospital from July 2017 to April 2018 were included in this retrospective case series study. All patients were tested for best corrected visual acuity (BCVA), and regularly examined with spectral domain optical coherence tomography (SD-OCT) and OCTA. All affected eyes were treated with 0.05 ml of intravitreal anti-VEGF agents (Ranibizumab or Conbercept), and all patients were followed up monthly. The changes in BCVA and central macular thickness (CMT) were noted on OCT images and compared with a paired t test. The changes in CNV were observed before and after therapy by using OCTA. Results: The BCVA of (LogMAR) all patients was 0.69±0.35 before treatment, and this indicator was raised to 0.49±0.38 after treatment; the difference was statistically significant (t=3.541, P=0.003). The CMT of all patients decreased from 363±66 μm before treatment to 302±66 μm after treatment, and thedifference was statistically significant (t=4.227, P=0.001). OCTA showed that the density of CNV decreased by different degrees in all patients after treatment. CNV almost disappeared in some patients, meanwhile, neuroepithelial detachment and edema decreased or completely disappeared. Although OCTA showed the thickened and abnormal vessels in some patients, neuroepithelial detachment and edema disappeared completely on OCTA. While neuroepithelial edema vanished completely, marginal vessel anastomosis could still be seen in one patient, and CNV recurred after 3 months in that patient. Conclusions: OCTA clearly demonstrates the morphological changes in CNV after anti-VEGF therapy, and it can be a reliable tool for the clinical diagnosis and therapeutic evaluation of CNV.

Objective: To investigate the effects of the timing of a preoperative intravitreal injection of Conbercept on the expression of vascular endothelial growth factor (VEGF) and the recovery of visual acuity in patients with diabetic retinopathy (DR). Methods: In this prospective clinical study, 109 patients with DR treated in the People's Hospital of Kaizhou District, Chongqing, from May 2015 to May 2017 were randomly divided into A, B and C groups, 32 patients in group A, 43 patients in group B, 34 patients in group C.The three groups were treated by vitrectomy. Patients in group A were not treated with an intravitreal injection of Conbercept before surgery, patients in group B were treated with an intravitreal injection of Conbercept at 3 d before surgery, and patients in group C were treated with an intravitreal injection of Conbercept at 5 d before surgery. The surgery for the three groups was observed, and the surgical time, intraoperative blood loss and incidence of iatrogenic hiatus were recorded. The subretinal neovascular membrane specimens of all patients were collected during the surgery to determine the expression levels of VEGF. The incidence of complications was observed, and changes in the best corrected visual acuity(BCVA) at different times before and after surgery were compared. Data were analyzed using a one-way ANOVA. Results: The surgical times and intraoperative blood loss of groups B and C were shorter and less than those in group A (P<0.05), and the rates of electric coagulation hemostasis during surgery, the incidence rates of the iatrogenic retinal hole and postoperative vitreous hemorrhage were lower than those in group A (P<0.05). The incidence rates of corneal edema, high intraocular pressure and anterior chamber inflammatory reactions in groups B and C after surgery were lower than those in group A (P<0.05), but there was no statistically significant difference between groups B and C (P>0.05). The positive rates of VEGF in the neovascular membranes of group B and group C were lower than that in group A, but the difference between groups B and C (P>0.05) was not significant. The BCVA of the three groups increased at 3 months and 6 months after surgery (P<0.05), and the BCVA of groups B and C at 3 and 6 months after surgery was higher than that of group A (P<0.05), without a statistically significant difference between groups B and C (P>0.05). Conclusions: Intravitreal injections of Conbercept on the 3 d and 5 d before surgery can shorten the surgical time, reduce the incidence of intraoperative and postoperative complications, reduce the expression level of VEGF in the neovascular membrane, and promote the recovery of visual acuity in patients with DR after surgery, without a significant difference between injections 3 d or 5 d before surgery.

Objective: To compare the efficacy of vision training and pencil push-ups as treatments for patients with small-angle exodeviation after surgery for intermittent exotropia. Methods: A controlled clinical trial was performed. Patients who presented with small-angle exodeviation (15△> exotropia >8△) after surgery for intermittent exotropia were recruited in Children's Hospital of Nanjing Medical Uviverstiy from June 2015 to June 2016. Fifty-four patients were divided into a vision training group (28) and a pencil push-up group (26). Patients in the pencil push-up group were instructed to hold a pencil at arm's length directly between their eyes. They were asked to continue moving the pencil slowly toward their nose until it could no longer be seen as a single image and then try to get the pencil point back into a single image. Patients were instructed to do 3 sets of 20 pencil push-ups per day at home, 5 days per week for 6 months. This treatment required an average of 15 minutes per day. Patients in the vision training group received therapy administered by a trained therapist during a weekly 60-minute office visit, with additional procedures to be performed at home for 15 minutes a day, 5 times per week for 6 months. Tests were performed before training and compared to results at 1, 3 and 6 months after training, including exodeviation at distance and near, and stereoacuity at distance. An independent samples t test, two-way repeated measures ANOVA, a paired t test and chi-square test were used for statistical analysis. Results: At 1 month and 6 months after training, the magnitude of the exodeviation between the pencil push-up group and the vision training group at distance and near was not significantly different. However, there was a significant difference between groups at 3 months after training, demonstrating a higher magnitude of exodeviation in the pencil push-up group than in the vision training group (t=3.44, P<0.001; t=2.55, P=0.01). The difference in the magnitude of exodeviation at distance and near between 1 month and 3 months after training was significant (t=-3.17, P=0.003; t=2.32, P=0.02). At 1 month and 3 months after training, the magnitude of stereoacuity at distance between the pencil push-up group and vision training group was not significantly different, However, there was a significant difference between the two groups at 6 months, demonstrating a better stereoacuity in the vision training group than in the pencil push-up group (t=9.33, P<0.001). At 6 months after training, the normal stereoscopic acquisition rate of the vision training group was higher than that of the pencil push-up group (χ2=5.02, P=0.03). Conclusions: Both vision training and pencil push-up training can effectively decrease the magnitude of small-angle exodeviation after surgery for intermittent exotropia. However, vision training has a better effect on stereoacuity.

Objective: To evaluate the corneal biomechanical properties in high myopia using corneal visualization Scheimpflug technology (Corvis ST) and the correlation between corneal biomechanical parameters and intraocular pressure (IOP), axial length (AL), central corneal thickness (CCT), age, and retinal nerve fiber layer thickness. Methods: This was a cross-sectional study. From June to October 2017, 53 patients with high myopia were recruited from Eye Hospital, Wenzhou Medical University. The right eyes of all patients were tested. There were 17 males and 36 females, with a mean age of 33.0±9.6 years (range 18-53 years). Corvis ST was used to obtain the corneal biomechanical parameters, including velocity at the first applanation (A1V), length at the first applanation (A1L), velocity at the second applanation (A2V), length at the second applanation (A2L), deformation amplitude at the highest concavity (DA), radius at the highest concavity (HCR), and peak distance at the highest concavity (PD). The differences in corneal biomechanical parameters were compared between high myopia (-6 D to -10 D, 23 eyes) and extremely high myopia (≤-10 D, 30 eyes). Data were analyzed using independent t test and Pearson correlation.Results: There was a significant difference in A2L (t=1.95, P=0.043) between the high myopia group and extremely high myopia group. There was a negative correlation between IOP and A1V, PD, DA (r=-0.56, -0.46, -0.63, P<0.001) and a positive correlation between IOP and A2V (r=0.37, P=0.005). There was a positive correlation between PD and AL (r=0.43, P=0.001). A1L, A2L, A2V were positively correlated with CCT (r=0.33, P=0.043; r=0.28, P=0.041; r=0.39, P=0.003). PD was negatively correlated with inferior RNFL thickness (r=-0.45, P=0.001). All parameters had no significant correlation with age. Conclusions:Corneal biomechanical parameters in high myopia are positively correlated with AL and CCT, and negatively correlated with IOP and RNFL thickness, suggesting that cornea of high myopia deformed easily, and the inferior RNFL is easily affected.

Objective: To determine if non-cycloplegic refraction on children with incipient myopia is accurate; to compare the differences in non-cycloplegic subjective refraction, cycloplegic autorefraction and cycloplegic subjective refraction in children with incipient myopia. Related factors that influence these measurements
were analyzed to provide evidence for the preferred refractive method in the clinic. Methods: This was a prospective clinical study. Ninety children with incipient myopia aged from 9 to 13 years were included from Eye Hospital, Wenzhou Medical University from May to September 2017. Subjective refraction, phoria,NRA/PRA, and accommodative lag were measured before cycloplegia. 1% cyclopentolate was used 3 times every five minutes. After 45 minutes, measurements were taken with autorefraction (CAR) and subjective refraction (CSR). Only the values of the right eyes were assessed using ANOVA and multiple linear regression for statistical analysis. Results: The mean value of the spherical equivalent (SE) of non-cycloplegic subjective refraction was -1.09±0.30 D. The mean values of cycloplegic autorefraction and subjective refraction were -1.27±0.40 D and -1.09±0.33 D, respectively. There was no statistically significant difference between non-cycloplegic and cycloplegic subjective refraction (P>0.05). The cylinder component from cycloplegic autorefraction was significantly higher than that from subjective
refraction (P<0.001), while the sphere components between them had no significant differences (P>0.05). Multiple linear regression analysis showed that the difference in diopter between SR and CAR was not associated with age (β=0.044, P=0.699), refractive error (β=0.091, P=0.430), phoria (β=-0.059, P=0.599),
accommodative lag (β=-0.064, P=0.576), or positive relative accommodation (β=0.043, P=0.709). Conclusions: Excluding accommodation and vergence anomalies, most children with incipient myopia can be measured accurately with non-cycloplegic subjective refraction.

Objective: To test whether crowding is much more severe under simulated monovision conditions.Methods: This was an experimental study. Twenty subjects participated in this experimental study from March to November 2015 in Wenzhou Medical University. Both eyes viewed clear images under normal conditions. While under the simulated monovision condition, one eye viewed a clear image (OD), and the other eye viewed a blurred image (0.75, 1.50, 2.50, 3.50 D). The eccentricities of the target in this study were 2.5°, 5.0°, and 7.5°. The threshold of the crowding effect was measured by a staircase method. Two-way ANOVA with repeated measures and a t-test with Bonferroni correction were used to test whether crowding was much more severe under the simulated monovision condition. Results: Crowding depended significantly on intraocular blur (F=13.37, P<0.001) and eccentricity (Ecc) (F=296.90, P<0.001). Significant interactions between intraocular blur and Ecc were found (F=4.03, P<0.001). When the eccentricity was 7.5° or 5.0°, intraocular blur significantly affected the critical gap size of crowding (F=8.59, P<0.001; F=3.74, P=0.01). However, intraocular blur did not affect the critical gap size of crowding when the eccentricity was 2.5°(F=1.13, P=0.34). Conclusions: Crowding is much more severe under simulated monovision conditions than normal conditions, and the difference is dependent on intraocular blur and eccentricity.

Objective: To evaluate the prediction accuracy of three intraocular lens power calculation formulas (SRK/T, Haigis and Barrett Universal Ⅱ) for postoperative refractive power using a Pentacam AXL panoramic biometer in cataract patients with ultra-high myopia. Methods: This was a prospective clinical study. Thirty-two cataract patients (56 eyes) with ultra-high myopia [axial length (AL) ≥28 mm] who underwent cataract surgery in the Ophthalmology Department of the First Affiliated Hospital of Chongqing Medical University from October 2017 to March 2018 were selected. The Pentacam AXL panoramic biometer was used to measure the anterior and posterior corneal surface curvature, anterior chamber depth and AL. An intraocular lens (IOL) with appropriate residual refractive power was individually selected for each eye, and the postoperative residual refractive powers of the three formulas were obtained. The refractive powers of the patients at three months after surgery were measured. Then the difference between the preoperative residual refractive power and postoperative refractive power, which was called the mean refractive error, the absolute value of the difference, which was called the mean absolute refractive error (MAE), were calculated. The quantitative data of non-normal distributions were expressed by the median and percentile (P25, P75), and the measurement data were tested by a Mann-Whitney U test or chi-square test. Linear regression analysis was used to evaluate the correlation between AL and MAE. Results: The median and percentile (P25, P75) of the mean refractive errors of the SRK/T, Haigis and Barrett Universal Ⅱ formula were -0.18(-0.53, 0.23), -0.18(-0.54, 0.09), -0.11(-0.49, 0.15), respectively. The mean refractive error of the Barrett Universal Ⅱ formula was significantly lower than that of SRK/T formula (Z=-2.517, P=0.012). However, the Barrett Universal Ⅱ formula agreed with the Haigis formula, the Haigis formula agreed with the SRK/T formula, and the differences in mean refractive error were not statistically significance (P>0.05). The median and percentile (P25, P75) of MAE of the SRK/T, Haigis and Barrett Universal Ⅱ formulas were 0.40(0.20, 0.61), 0.32(0.14, 0.64), 0.27(0.12, 0.55), respectively. And there were no significant differences among the Barrett Universal Ⅱ, Haigis and SRK/T formulas (P>0.05). Moreover, the MAE is related to AL. In the SRK/T formula, an increase of 1 mm in axial length led to an increase of 0.23 D in MAE; in the Haigis formula, an axial length increase by 1 mm led to a MAE increase of 0.04 D; but in the Barrett Universal Ⅱ formula, an axial length increase of 1 mm led to an absolute error reduction of 0.01 D. Conclusions: The three IOL power calculation formulas are relatively accurate for postoperative refractive error in cataract patients with ultra-high myopia. The Barrett Universal Ⅱ formula has more advantages than the SRK/T formula and Haigis formula.

Objective: To observe the changes in the central corneal epithelium thickness (CCET) after transepithelial photorefractive keratectomy (TPRK) in low and moderate myopia patients. Methods: This was a retrospective case series study. Sixty-two myopes (114 eyes) were recruited who underwent TPRK at the Eye Hospital of China Academy of Traditional Chinese Medicine from January 2015 to June2016. Patients were divided into two groups based on spherical equivalence (SE) refractive error: a low myopia group (-3.00 D to -1.25 D) and a moderate myopia group (-6.00 D to -3.25 D). Both groups underwent testing for uncorrected visual acuity (UCVA), SE, and CCET and were compared preoperatively and postoperatively at 1 week, 2 weeks, 1 month, 3 months, 6 months and 1 year. Preoperative and postoperative changes were analyzed with two independent samples t tests and repeated measures analysis of variance. Results: The two groups had significant differences in both UCVA and SE preoperatively (t=6.605, P=0.01; t=15.374, P=0.01). There were no significant differences between BCVA and CCET preoperatively (both P>0.05). The CCET values for the low myopia group were relatively higher than that for the moderate myopia group 1 week postoperation (t=3.362, P=0.01), but there were no significant differences at the other time points. Between the two groups and within each group at postoperative observation time points, UCVA and SE were not statistically significant. For CCET, except for between postoperative 1 month and 2 weeks, between postoperative 6 months, 1 year and preoperative without difference. And there was with difference between time points (P<0.05). Conclusions: In the early stage after TPRK, CCET in the low myopia group recovers faster than that in the moderate myopia group. The CCET in both groups is restored to preoperative levels after 6 months, and remains stable. TPRK surgical treatments for visual acuity in low and moderate myopia has a high predictability and a stable SE.

Objective: To evaluate optical quality and correlation factors in thyroid-associated ophthalmopathy (TAO) patients using a double-pass system. Methods: In this case control study, 31 TAO patients (62 eyes), including 28 eyes in an active stage and 34 eyes in an inactive stage, and 31 normal controls (62 eyes) were recruited from Sichuan Provincial People's Hospital from October 2016 to June 2017. Three Optical Quality Analysis System (OQAS) values at 100%, 20%, 9% contrast (OVl00%, OV20%, and OV9%), objective scatter index (OSI), modulation transfer function cutoff (MTFcutoff), Strehl ratio (SR), and mean OSI reflecting the change of tear film were measured with an OQAS visual quality analyzing system. Clinical activity score (CAS) and NO SPECS grading were performed for TAO patients. Visual acuity, intraocular pressure (IOP), height of the palpebral fissure, eyelid lag, proptosis, lagophthalmos, Schirmer Ⅰand tear film break-up time (BUT) were evaluated. An independent samples t test or Mann-Whitney U test were used to compare the quantitative data between the two groups. A Chi-square test was used to compare the qualitative data between the two groups. The Spearman correlation test was performed for data analysis. Results: Optical quality parameters including OVl00%, OV20%, OV9%, MTFcutoff, and SR had significantly decreased, and OSI and mean OSI had significantly improved in TAO patients compared with that of normal controls (all P<0.05). No statistically significant differences were found in any optical quality parameters or the Schirmer Ⅰ test in TAO patients between the active stage group and inactive stage group. However, the differences in the height of the palpebral fissure, eyelid lag, proptosis, lagophthalmos,IOP and BUT were statistically significantly between the two groups (all P<0.05). The OSI in patients with TAO was negatively correlated with the height of the palpebral fissure (r=-0.296, P=0.020), and positively correlated with NO SPECS, age and mean OSI (r=0.469, P<0.001; r=0.429, P=0.001; r=0.836, P<0.001). The mean OSI in patients with TAO was negatively correlated with the height of the palpebral fissure(r=-0.252, P=0.048), and positively correlated with NO SPECS, age and IOP (r=0.404, P=0.001; r=0.347,
P=0.006; r=0.311, P=0.014). Both OSI and mean OSI were not significantly correlated with eyelid lag, proptosis, lagophthalmos, CAS, BUT, or SchirmerⅠ. Conclusions: TAO patients have a lower optical quality and more serious dry eye compared with that of normal people. Age, height of palpebral fissure, NO SPECS and mean OSI are related to optical quality in TAO patients.

Objective: To evaluate the clinical characteristics of focal choroidal excavation (FCE). Methods: This was a retrospective observational case series study. The medical records of 12 patients (22 eyes) with FCE were reviewed from October 2016 to November 2017 in Jinhua Eye Hospital. All patients underwent the following examinations: Slit lamp biomicroscopy, indirect ophthalmoscopy, best corrected visual acuity (BCVA), optic coherence tomography (OCT) and ultra-wide angle laser fundus scans (Daytona). Fundus fluorescein angiography (FFA) was also performed on some patients. Results: The 12 patients included 3 females (3/12) and 9 males (9/12). The average age was 45.3±14.2 years (range from 25 to 59 years). Ten patients had bilateral involvement. Ten eyes (10/22) were emmetropic, and the others were myopic. The mean BCVA (LogMAR) was 0.18±0.11 (range from 0 to 0.5). Clinical examination and color photography showed that 17 eyes (17/22) were normal. Two eyes (2/22) had varying degrees of foveal pigment alterations or a yellowish spot at the edge of the fovea. Two eyes (2/22) showed macular edema seen as a disc and 1 eye had both macular hemorrhage and exudation. OCT scans showed that 22 eyes (22/22) had a single FCE. It was located in the subfovea in 6 eyes (6/22) and was eccentric in the other 16 eyes (16/22). In 2 eyes (2/22) with central serous choroidal retinopathy, there was no obvious change in the local choroidal depression after treatment of the hydroepithelia. Choroidal neovascularization was also present in 1 eye (1/22), after repeated anti-vascular endothelial growth factor treatment, the disease reoccured. Conclusions: Most patients have bilateral involvement and secondly, unilateral involvement. Visual acuity is normal or only moderately impaired to different degrees. FCE is related to refraction. Clinical examination shows that the fundus of most patients showed normal. When complications occur, corresponding symptoms appear.

Age-related macular degeneration (AMD) is one of the major causes of blindness around the world.Anti-vascular endothelial growth factor (anti-VEGF) therapy has become the first-line therapy for treating neovascular AMD (nAMD). Considering the burden of the high cost of anti-VEGF medicine, long-term and often lifelong repeated injections required of patients, it is important to understand the patterns of anti-VEGF therapy for nAMD patients in real-world settings, which will optimize the delivery of healthcare for nAMD patients, improve clinical pathways and may even help with decision-making in healthcare policy. Recently, data analysis studies on nAMD, based on electronic information platforms, have unveiled the epidemiology of nAMD in different countries and regions, the anti-VEGF therapy patterns in the real
world and the medical expenditure related to nAMD. This article reviews research progress in real-world studies on nAMD based on electronic information platforms.