The adaptive optics (AO) can be used to compensate for ocular aberrations to achieve near diffraction limited high-resolution retinal images. However, many factors such as the limited aberration measurement and correction accuracy with AO, intraocular scatter, imaging noise and so on will degrade the quality of retinal images. Image post processing is an indispensable and economical method to make up for the limitation of AO retinal imaging procedure. In this paper, we proposed a deep learning method to restore the degraded retinal images for the first time. The method directly learned an end-to-end mapping between the blurred and restored retinal images. The mapping was represented as a deep convolutional neural network that was trained to output high-quality images directly from blurry inputs without any preprocessing. This network was validated on synthetically generated retinal images as well as real AO retinal images. The assessment of the restored retinal images demonstrated that the image quality had been significantly improved.

Higher-order aberrations (HOAs) and intraocular scatter lead to the degradation of image quality on the retina, and consequently deteriorate subjective visual performance. In this article, we modified an adaptive optics double-pass system to combine objective and subjective visual testing capabilities. Employing the modified DP system, we investigated the effects of HOAs and intraocular scatter on contrast sensitivity. Contrast sensitivity measurements were performed with HOAs either retained or corrected by adaptive optics, and with scatter either remaining at the natural eye-induced level or further enhanced by a set of three different scatter filters. Contrast sensitivity was found to be worse when HOAs were uncorrected or scatter increased. Quantitative analysis indicated that the joint effect of HOAs and scatter on contrast sensitivity was not a simple summation of each contributing factor, suggesting a potential compensatory mechanism between HOAs and intraocular scatter on contrast sensitivity.

Measurement of the double-pass (DP) point-spread function (PSF) can provide an objective and non-invasive method for estimating intraocular scatter in the human eye. The objective scatter index (OSI), which is calculated from the DP PSF images, is commonly used to quantify intraocular scatter. In this article, we simulated the effect of higher-order ocular aberrations on OSI, and the results showed that higher-order ocular aberrations had a significant influence on OSI. Then we developed an adaptive optics DP PSF measurement system (AO-DPPMS) which was capable of correcting ocular aberrations up to eighth-order radial Zernike modes over a 6.0-mm pupil. Employing this system, we obtained DP PSF images of four subjects at the fovea. OSI values with aberrations corrected up to 2nd, 5th and 8th Zernike order were calculated respectively, from the DP PSF images of the four subjects. The experimental results were consistent with the simulation, suggesting that it is necessary to compensate for the higher-order ocular aberrations for accurate intraocular scatter estimation.

To analyze the incidence and progression rates of myopia among children aged 6-8 years in Chengdu, and to evaluate the influence of baseline parameters and dynamic changes in ocular biometric parameters on new-onset myopia risk. A total of 26,169 students aged 6-8 years from eight administrative districts in Chengdu were followed annually for 4 years. All students underwent uncorrected distance visual acuity testing, non-cycloplegic autorefraction, and ocular biometric parameters. Potential risk factors for myopia onset were screened through multivariate logistic regression analysis. The influence of each factor was evaluated using the area under the curve (AUC). Results showed that the baseline prevalence of myopia was 17.8%. The cumulative incidence rates of myopia were 8.1-13.4% (2nd year), 8.6-12.2% (3rd year), and 11.5-16.2% (4th year) across different age groups, with higher rates in older children and females. The mean myopia progression rate was - 0.45 ± 0.32 D/year in the persistent myopia (PM) group. The annual growth rates of axial length(AL) and axial length/corneal radius (AL/CR) ratio in both the new-onset myopia (NOM) group and PM group were higher than those in the persistent non-myopia (PNM) group. The NOM showed significantly higher annual reduction rates in lens thickness and lens power(LP), and higher annual increase rate in anterior chamber depth compared to both PNM and PM. Both baseline parameters (e.g., AL, LP) and their changes 1-3 years before myopia onset significantly influenced the risk of myopia onset (all P < 0.001). The model combining baseline values and annual changes demonstrated good performance with cross-validated AUC values of 0.853. The study demonstrates that baseline values and changes 1-3 years before myopia onset in ocular biometric parameters have important influence on myopia onset, suggesting their incorporation into comprehensive myopia prevention and control systems for children.

The non-invasive measurement of hemoglobin oxygen saturation (SO₂) in retinal vessels is based on spectrophotometry and the absorption spectral characteristics of the tissue. The dual-wavelength retinal images are simultaneously captured via retinal oximetry. SO₂ is calculated by processing a series of images and by calculating the optic density ratio of two images. However, existing SO₂ research is focused on the thick vessels in the high-clarity region of retinal images. However, the thin vessels in the low-clarity region could provide significant information for the detection and diagnosis of neovascular diseases. To this end, we proposed a novel hybrid vessel segmentation algorithm. Firstly, a median filter was employed for image denoising. Secondly, high- and low-clarity region segmentation was carried out based on a clarity histogram. The vessels in the high-clarity areas were segmented after implementing a Gaussian filter, a matched filter, and morphological segmentation. Additionally, the vessels in the low-clarity areas were segmented using a guided filter, matched filtering, and dynamic threshold segmentation. Finally, the results were obtained through image merger and morphological operations. The experimental results and analysis show that the proposed method can effectively segment thick and thin vessels and can extend the measuring range of dual-wavelength retinal oximetry.