Background/AimsTo develop a deep learning system for automated glaucomatous optic neuropathy (GON) detection using ultra-widefield fundus (UWF) images.MethodsWe trained, validated and externally ...evaluated a deep learning system for GON detection based on 22 972 UWF images from 10 590 subjects that were collected at 4 different institutions in China and Japan. The InceptionResNetV2 neural network architecture was used to develop the system. The area under the receiver operating characteristic curve (AUC), sensitivity and specificity were used to assess the performance of detecting GON by the system. The data set from the Zhongshan Ophthalmic Center (ZOC) was selected to compare the performance of the system to that of ophthalmologists who mainly conducted UWF image analysis in clinics.ResultsThe system for GON detection achieved AUCs of 0.983–0.999 with sensitivities of 97.5–98.2% and specificities of 94.3–98.4% in four independent data sets. The most common reasons for false-negative results were confounding optic disc characteristics caused by high myopia or pathological myopia (n=39 (53%)). The leading cause for false-positive results was having other fundus lesions (n=401 (96%)). The performance of the system in the ZOC data set was comparable to that of an experienced ophthalmologist (p>0.05).ConclusionOur deep learning system can accurately detect GON from UWF images in an automated fashion. It may be used as a screening tool to improve the accessibility of screening and promote the early diagnosis and management of glaucoma.
Retinal detachment can lead to severe visual loss if not treated timely. The early diagnosis of retinal detachment can improve the rate of successful reattachment and the visual results, especially ...before macular involvement. Manual retinal detachment screening is time-consuming and labour-intensive, which is difficult for large-scale clinical applications. In this study, we developed a cascaded deep learning system based on the ultra-widefield fundus images for automated retinal detachment detection and macula-on/off retinal detachment discerning. The performance of this system is reliable and comparable to an experienced ophthalmologist. In addition, this system can automatically provide guidance to patients regarding appropriate preoperative posturing to reduce retinal detachment progression and the urgency of retinal detachment repair. The implementation of this system on a global scale may drastically reduce the extent of vision impairment resulting from retinal detachment by providing timely identification and referral.
Abstract
Artificial intelligence (AI) based on deep learning has shown excellent diagnostic performance in detecting various diseases with good-quality clinical images. Recently, AI diagnostic ...systems developed from ultra-widefield fundus (UWF) images have become popular standard-of-care tools in screening for ocular fundus diseases. However, in real-world settings, these systems must base their diagnoses on images with uncontrolled quality (“passive feeding”), leading to uncertainty about their performance. Here, using 40,562 UWF images, we develop a deep learning–based image filtering system (DLIFS) for detecting and filtering out poor-quality images in an automated fashion such that only good-quality images are transferred to the subsequent AI diagnostic system (“selective eating”). In three independent datasets from different clinical institutions, the DLIFS performed well with sensitivities of 96.9%, 95.6% and 96.6%, and specificities of 96.6%, 97.9% and 98.8%, respectively. Furthermore, we show that the application of our DLIFS significantly improves the performance of established AI diagnostic systems in real-world settings. Our work demonstrates that “selective eating” of real-world data is necessary and needs to be considered in the development of image-based AI systems.
Medical artificial intelligence (AI) and big data technology have rapidly advanced in recent years, and they are now routinely used for image-based diagnosis. China has a massive amount of medical ...data. However, a uniform criteria for medical data quality have yet to be established. Therefore, this review aimed to develop a standardized and detailed set of quality criteria for medical data collection, storage, annotation, and management related to medical AI. This would greatly improve the process of medical data resource sharing and the use of AI in clinical medicine.
Purpose. To evaluate the astigmatic outcomes of wavefront-guided sub-Bowman keratomileusis (WFG-SBK) for low to moderate myopic astigmatism. Methods. This study enrolled 100 right eyes from 100 ...patients who underwent WFG-SBK for the correction of myopia and astigmatism. The polar value method was performed with anterior and posterior corneal astigmatism measured with Scheimpflug camera combined with Placido corneal topography (Sirius, CSO) and refractive astigmatism preoperatively and 1 month, 3 months, and 6 months postoperatively. Results. Similar results for surgically induced astigmatism (SIA) and error of the procedure in both anterior corneal astigmatism (ACA) and total ocular astigmatism (TOA). There was a minor undercorrection of the cylinder in both ACA and TOA. Posterior corneal astigmatism (PCA) showed no significant change. Conclusions. Wavefront-guided SBK could provide good astigmatic outcomes for the correction of low to moderate myopic astigmatism. The surgical effects were largely attributed to the astigmatic correction of the anterior corneal surface. Posterior corneal astigmatism remained unchanged even after WFG-SBK for myopic astigmatism. Polar value analysis can be used to guide adjustments to the treatment cylinder alongside a nomogram designed to optimize postoperative astigmatic outcomes in myopic WFG-SBK.
To evaluate anterior segment variations after posterior chamber phakic intraocular lens (pIOL) implantation in myopic eyes.
Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China.
...Cohort study.
Patients with high myopia were scheduled for nontoric Implantable Collamer Lens pIOL or toric Implantable Collamer Lens pIOL implantation. Anterior segment optical coherence tomography was performed to evaluate the anterior segment variations over time and the impact of physiologic accommodation and change in brightness after pIOL implantation. Slitlamp photography of the anterior segment was taken after pupil dilation to calculate pIOL rotation.
There was significantly improved visual acuity and refractive status after implantation of both pIOLs. Anterior segment axial variations had good stability 6 months postoperatively. However, nontoric pIOL rotation occurred in 1 eye 7 months postoperatively and rotation of the toric pIOL occurred in 2 eyes at 3 months and 6 months. Stimulations of -4.0 diopters (D) and -8.0 D led to significant changes in anterior chamber depth (ACD) based on the lens (ACD-lens), ACD after pIOL implantation, distance between the pIOL posterior surface and the lens anterior surface (pIOL-lens), and pupil diameter. Increasing ambient light brightness could reduce the ACD-lens and pIOL-lens.
Both pIOLs had good axial stability in myopic eyes 6 months postoperatively; however, rotational stability over time could not be determined. The physiologic adjustment and change in brightness could influence the anterior segment significantly in eyes with a pIOL.
Purpose. To evaluate the safety of high-intensity focused ultrasound keratoplasty as a treatment for presbyopia by examining its effect on the rabbit anterior segment. Methods. The right corneas of ...36 New Zealand rabbits were treated with HIFU keratoplasty. The animals were sacrificed at 1, 7, 15, 30, 60, and 90 days after operation. Collagen type I, MMP-2, and MMP-9 were evaluated using immunohistochemistry. For the detection of apoptosis, the TUNEL method was applied. The SOD and MDA levels were analyzed with assay kits. Results. Collagen type I, MMP-2, and MMP-9 levels were altered after the operation but returned to normal within 90 days. The apoptotic index (AI) of the corneal cells decreased from 1 to 30 days gradually. No apoptosis was observed in the epithelial cells of the lens, and the SOD and MDA levels were normal at any time point. Conclusion. After HIFU keratoplasty, the histomorphology of the cornea changed, the corneal collagen type I levels decreased, the corneal MMP-2 and MMP-9 levels increased, and the corneal cells underwent apoptosis for a period of time. Ninety days after the operation, the levels returned to normal, and the lenses were not affected. Thus, HIFU presents good biological safety for eyes.
Purpose. To evaluate high intensity focused ultrasound (HIFU) as an innovation and noninvasive technique to correct presbyopia by altering corneal curvature in the rabbit eye. Methods. Eighteen ...enucleated rabbit eyes were treated with a prototype HIFU keratoplasty. According to the therapy power, these eyes were divided three groups: group 1 (1 W), group 2 (2 W), and group 3 (3 W). The change in corneal power was quantified by a Sirius Scheimpflug camera. Light microscopy (LM) and transmission electron microscopy (TEM) were performed to determine the effect on the corneal stroma. Results. In the treated eyes, the corneal curvature increases from 49.42 ± 0.30 diopters (D) and 48.00 ± 1.95 D before procedure to 51.37 ± 1.11 D and 57.00 ± 1.84 D after HIFU keratoplasty application in groups 1 and 3, respectively. The major axis and minor axis of the focal region got longer when the powers of the HIFU got increased; the difference was statistically significant ( p < 0.05 ). LM and TEM showed HIFU-induced shrinkage of corneal stromal collagen with little disturbance to the underlying epithelium. Conclusions. We have preliminarily exploited HIFU to establish a new technique for correcting presbyopia. HIFU keratoplasty will be a good application prospect for treating presbyopia.
In recent years, the incidence of myopia has increased at an alarming rate among children and adolescents in China. The exploration of an effective prevention and control method for myopia is in ...urgent need. With the development of information technology in the past decade, artificial intelligence with the Internet of Things technology (AIoT) is characterized by strong computing power, advanced algorithm, continuous monitoring, and accurate prediction of long-term progression. Therefore, big data and artificial intelligence technology have the potential to be applied to data mining of myopia etiology and prediction of myopia occurrence and development. More recently, there has been a growing recognition that myopia study involving AIoT needs to undergo a rigorous evaluation to demonstrate robust results.