Evaluation of photoreceptor features in retinitis pigmentosa with cystoid macular edema by using an adaptive optics fundus camera.

OBJECTIVE: Cystoid macular edema (CME) in retinitis pigmentosa (RP) is an important complication causing visual dysfunction. We investigated the effect of CME on photoreceptors in RP patients with previous or current CME, using an adaptive optics (AO) fundus camera. METHODS: We retrospectively observed the CME and ellipsoid zone (EZ) length (average of horizontal and vertical sections) by optical coherence tomography. The density and regularity of the arrangement of photoreceptor cells (Voronoi analysis) were examined at four points around 1.5° from superior to inferior and temporal to nasal. We also performed a multivariate analysis using CME duration, central macular thickness and transversal length of CME. RESULTS: We evaluated 18 patients with previous or current CME (18 eyes; age, 48.7 ± 15.6 years) and 24 patients without previous or current CME (24 eyes; age, 46.0 ± 14.5 years). There were no significant differences in age, logMAR visual acuity, or EZ length. In groups with and without CME, cell density was 11967 ± 3148 and 16239 ± 2935 cells/mm2, and sequence regularity was 85.5 ± 3.4% and 88.5 ± 2.8%, respectively; both parameters were significantly different. The correlation between photoreceptor density and age was more negative in group with CME. The CME group tended toward greater reductions in duration of CME. CONCLUSION: Complications of CME in RP patients may lead to a decrease in photoreceptor density and regularity. Additionally, a longer duration of CME may result in a greater reduction in photoreceptor density.

Detailed Evaluation of Chromatic Pupillometry and Full-Field Stimulus Testing to Assess Ultralow Vision in Retinitis Pigmentosa.

Purpose: Novel therapeutic options, such as regenerative medicine and gene therapy, are now emerging as viable treatment options for patients with severe visual impairments, such as retinitis pigmentosa (RP). Gradable assessment of patients' visual function is essential to consider treatment options and to evaluate treatment outcomes; however, evaluation of visual function in patients with advanced low vision is often challenging because of patients' poor and sometimes unpredictable responses. In this study, we attempted to accurately assess visual capabilities and disease stage in patients with RP with a visual acuity (VA) of ≤ 0.01. Design: Retrospective analysis of visual function indicators, including VA, retinal thickness, full-field stimulus testing (FST), and chromatic pupillometry. Subjects: Overall, 43 patients (84 eyes) with advanced RP with a VA of ≤ 0.01 visited Kobe City Eye Hospital from 2019 to 2021. Methods: Hierarchical (multilevel) Bayesian modeling was used to estimate individual eye's pupil response and FST threshold, taking into account the ambiguity and randomness often observed in patients with ultralow vision. Using the estimated ability obtained from each test, the correlation between each test and retinal thickness was further analyzed to make a comprehensive assessment of the data. Main Outcome Measures: Visual acuity, retinal thickness, FST threshold, and pupil diameter change to different light stimuli. Results: Full-field stimulus testing and pupillometry measurements were moderately correlated with VA but exhibited a wide range of values within the same VA groups. Full-field stimulus testing was not correlated with central retinal thickness at counting fingers/hand motion VA range and seemed to reflect overall remaining photoreceptor function, including peripheral retina. Pupillometry may be able to distinguish between different levels of inner retinal function. Conclusions: The combination of pupillometry and FST allowed for graded evaluation of visual function within patients grouped in the same VA groups in patients with advanced RP with ultralow vision. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.

Robotic search for optimal cell culture in regenerative medicine.

Induced differentiation is one of the most experience- and skill-dependent experimental processes in regenerative medicine, and establishing optimal conditions often takes years. We developed a robotic AI system with a batch Bayesian optimization algorithm that autonomously induces the differentiation of induced pluripotent stem cell-derived retinal pigment epithelial (iPSC-RPE) cells. From 200 million possible parameter combinations, the system performed cell culture in 143 different conditions in 111 days, resulting in 88% better iPSC-RPE production than that obtained by the pre-optimized culture in terms of the pigmentation scores. Our work demonstrates that the use of autonomous robotic AI systems drastically accelerates systematic and unbiased exploration of experimental search space, suggesting immense use in medicine and research.

Automated evaluation of retinal pigment epithelium disease area in eyes with age-related macular degeneration.

The retinal pigment epithelium (RPE) is essential for the survival and function of retinal photoreceptor cells. RPE dysfunction causes various retinal diseases including age-related macular degeneration (AMD). Clinical studies on ES/iPS cell-derived RPE transplantation for RPE dysfunction-triggered diseases are currently underway. Quantification of the diseased RPE area is important to evaluate disease progression or the therapeutic effect of RPE transplantation. However, there are no standard protocols. To address this issue, we developed a 2-step software that enables objective and efficient quantification of RPE-disease area changes by analyzing the early-phase hyperfluorescent area in fluorescein angiography (FA) images. We extracted the Abnormal region. This extraction was based on deep learning-based discrimination. We scored the binarized extracted area using an automated program. Our program's performance for the same eye from the serial image captures was within 3.1 ± 7.8% error. In progressive AMD, the trend was consistent with human assessment, even when FA images from two different visits were compared. This method was applicable to quantifying RPE-disease area changes over time, evaluating iPSC-RPE transplantation images, and a disease other than AMD. Our program may contribute to the assessment of the clinical course of RPE-disease areas in routine clinics and reduce the workload of researchers.

A Variable Scheduling Maintenance Culture Platform for Mammalian Cells.

Cell culturing is a basic experimental technique in cell biology and medical science. However, culturing high-quality cells with a high degree of reproducibility relies heavily on expert skills and tacit knowledge, and it is not straightforward to scale the production process due to the education bottleneck. Although many automated culture systems have been developed and a few have succeeded in mass production environments, very few robots are permissive of frequent protocol changes, which are often required in basic research environments. LabDroid is a general-purpose humanoid robot with two arms that performs experiments using the same tools as humans. Combining our newly developed AI software with LabDroid, we developed a variable scheduling system that continuously produces subcultures of cell lines without human intervention. The system periodically observes the cells on plates with a microscope, predicts the cell growth curve by processing cell images, and decides the best times for passage. We have succeeded in developing a system that maintains the cultures of two HEK293A cell plates with no human intervention for 192 h.

Artifactual reduction in oxygen saturation readings affected by indocyanine green in ophthalmic angiography.

PURPOSE: To investigate time-dependent changes in oxygen saturation (SpO2) readings during fluorescein angiography (FA) or simultaneous fluorescein and indocyanine green angiography (F/I). STUDY DESIGN: A prospective observational study. METHODS: One hundred twenty-five consecutive patients underwent FA or F/I. After application of the exclusion criteria, 67 patients (28 underwent FA, 39 underwent F/I) were included in the study. During angiography, SpO2 was measured by using a pulse oximeter and recording each second. The arm-to-retina circulation time (ARCT) was measured on the angiography video images. Latency time was defined as the time from intravenous injection of the dye to observation of a 3% decrease in the SpO2 reading, and variation SpO2, as the difference between baseline SpO2 and the lowest SpO2 value. RESULTS: The mean variation SpO2 was - 1.0 ± 0.6 during FA, and - 6.1 ± 1.5 during F/I. Variation SpO2 was within 2% during FA and > 3% during F/I. No corresponding changes were observed in the other vital signs or subjective symptoms. The mean latency time was 33.3 ± 6.4 s. Latency time was positively correlated with the ARCT in the F/I group (rs = 0.766, P < .001). CONCLUSION: Reductions in SpO2 readings were identified during F/I, but not during FA. Our results suggest that clinicians should be mindful that reductions in SpO2 readings during F/I may be artifactually caused by indocyanine green.

Optical Coherence Tomography-Based Deep-Learning Models for Classifying Normal and Age-Related Macular Degeneration and Exudative and Non-Exudative Age-Related Macular Degeneration Changes.

INTRODUCTION: The use of optical coherence tomography (OCT) images is increasing in the medical treatment of age-related macular degeneration (AMD), and thus, the amount of data requiring analysis is increasing. Advances in machine-learning techniques may facilitate processing of large amounts of medical image data. Among deep-learning methods, convolution neural networks (CNNs) show superior image recognition ability. This study aimed to build deep-learning models that could distinguish AMD from healthy OCT scans and to distinguish AMD with and without exudative changes without using a segmentation algorithm. METHODS: This was a cross-sectional observational clinical study. A total of 1621 spectral domain (SD)-OCT images of patients with AMD and a healthy control group were studied. The first CNN model was trained and validated using 1382 AMD images and 239 normal images. The second transfer-learning model was trained and validated with 721 AMD images with exudative changes and 661 AMD images without any exudate. The attention area of the CNN was described as a heat map by class activation mapping (CAM). In the second model, which classified images into AMD with or without exudative changes, we compared the learning stabilization of models using or not using transfer learning. RESULTS: Using the first CNN model, we could classify AMD and normal OCT images with 100% sensitivity, 91.8% specificity, and 99.0% accuracy. In the second, transfer-learning model, we could classify AMD as having or not having exudative changes, with 98.4% sensitivity, 88.3% specificity, and 93.9% accuracy. CAM successfully described the heat-map area on the OCT images. Including the transfer-learning model in the second model resulted in faster stabilization than when the transfer-learning model was not included. CONCLUSION: Two computational deep-learning models were developed and evaluated here; both models showed good performance. Automation of the interpretation process by using deep-learning models can save time and improve efficiency. TRIAL REGISTRATION: No15073.

Anterior Segment Optical Coherence Tomography Angiography Imaging of Conjunctiva and Intrasclera in Treated Primary Open-Angle Glaucoma.

PURPOSE: To investigate conjunctival and intrascleral vasculature in glaucoma eyes using anterior segment (AS)-optical coherence tomography angiography (OCTA) and assess the factors contributing to the vessel density in AS-OCTA images. DESIGN: Prospective, cross-sectional study. METHODS: Thirty-four patients with primary open-angle glaucoma and 20 healthy subjects were included. A swept-source OCT system was used to obtain the AS-OCTA images of the corneoscleral limbus at the nasal and temporal quadrants. Vessel densities were measured in the superficial (from the conjunctival epithelium to a depth of 200 µm) and deep (from a depth of 200 to 1000 µm) layers. The vessel density was compared between healthy and glaucoma eyes, and the associations of the vessel density with possible confounding factors were analyzed using univariable and multivariable analyses. RESULTS: The vessel density was not significantly different between healthy eyes and eyes with glaucoma. There was a significant association of superficial vessel density with the use of a prostaglandin analog (P = .007) and with nasal location (P = .016) in eyes with glaucoma. Deep vessel density was significantly smaller with advancing age (P = .029) in healthy eyes and greater with higher intraocular pressure (P = .021) in eyes with treated glaucoma. CONCLUSIONS: AS-OCTA images may be useful for the objective assessment of conjunctival hyperemia and helpful for understanding the pathophysiology of post-trabecular aqueous humor outflow.

Deep Learning Classification Models Built with Two-step Transfer Learning for Age Related Macular Degeneration Diagnosis.

The objective of this study was to build deep learning models with optical coherence tomography (OCT) images to classify normal and age related macular degeneration (AMD), AMD with fluid, and AMD without any fluid. In this study, 185 normal OCT images from 49 normal subjects, 535 OCT images of AMD with fluid, and 514 OCT mages of AMD without fluid from 120 AMD eyes as training data, while 49 normal images from 25 normal eyes, 188 AMD OCT images with fluid and 154 AMD images without any fluid from 77 AMD eyes as test data, were enrolled. Data augmentation was applied to increase the number of images to build deep learning models. Totally, two classification models were built in two steps. In the first step, a VGG16 model pre-trained on ImageNet dataset was transfer learned to classify normal and AMD, including AMD with fluid and/or without any fluid. Then, in the second step, the fine-tuned model in the first step was transfer learned again to distinguish the images of AMD with fluid from the ones without any fluid. With the first model, normal and AMD OCT images were classified with 0.999 area under receiver operating characteristic curve (AUC), and 99.2% accuracy. With the second model, AMD with the presence of any fluid, and AMD without fluid were classified with 0.992 AUC, and 95.1% accuracy. Compared with a transfer learned VGG16 model pre-trained on ImageNet dataset, to classify the three categories directly, higher classification performance was achieved with our notable approach. Conclusively, two classification models for AMD clinical practice were built with high classification performance, and these models should help improve the early diagnosis and treatment for AMD.

Unique circumferential peripheral keratitis in relapsing polychondritis: A case report.

RATIONALE: Relapsing polychondritis (RP) is a rare collagen disease characterized by inflammation and destruction of cartilage throughout the body. The paper details the clinical course of a case of RP with unique circumferential peripheral keratitis. PATIENT CONCERNS: A 54-year-old Japanese woman was referred to the hospital presenting with auricular and ocular pain. DIAGNOSES: Based on the auricle biopsy results and the three presenting symptoms (bilateral auricular chondritis, inflammatory arthritis and ocular inflammation), her condition was diagnosed as RP. INTERVENTIONS: The three presenting symptoms gradually improved with prednisolone (PSL), methylprednisolone and cyclophosphamide combination therapy, followed by PSL, methotrexate and infliximab combination therapy. However, one month after the initial visit, despite ongoing treatment, a unique circumferential peripheral keratitis suddenly occurred, in which the corneal infiltration gradually clumped together and shrank at the peripheral area. The eye and ear pain showed exacerbations and remissions on reducing the dosage of steroid drugs. The general condition was improved on altering systemic therapy to PSL, methotrexate and tocilizumab. OUTCOMES: Keratitis gradually disappeared within 10 months of the initial visit. LESSONS: This is the first report of a case of RP causing unique circumferential peripheral keratitis. This keratitis occurred despite use of focal and systemic steroids and showed improvement with general recovery. This may indicate that stabilization of general condition is important for recovery from keratitis in RP.