Off the Cuff: The Power and Perils of Leadership

This cross-sectional study looked at patients who received a COVID-19 diagnosis between Dec. 30, 2019, and Feb. 7, 2020, at Tongji Hospital. A total of 102 patients (48 males [47%] and 54 females [53%]) with clinical symptoms, Rt, and chest computed tomography (CT) abnormalities were identified with a clinical diagnosis of COVID-19. Patients had a mean [SD] gestational age of 57.63 [14.90] years.

Of a total of 102 patients identified, 72 patients (36 men [50%] and 36 women [50%]; mean [SD] age, 58.68 [14.81] years) were confirmed to have COVID-19 by laboratory diagnosis with a SARS-CoV-2 RT-PCR assay. Only two patients (2.78%) with conjunctivitis were identified from 72 patients with a laboratory confirmed COVID-19. Of those two patients, SARS-CoV-2 RNA fragments were found in ocular discharges by SARS-CoV-2 RT-PCR in only one patient.

Our findings suspect the incidence of SARS-CoV-2 infection through the ocular surface is extremely low, while the nosocomial infection of SARS-CoV-2 through the eyes after occupational exposure is a potential route. To lower the SARS-CoV-2 nosocomial infection, all health care professionals should wear protective goggles. The inefficient diagnostic method and the sampling time lag may contribute to the lower positive rate of conjunctival swab samples of SARS-CoV-2.

This is a systematic review on the main algorithms using machine learning (ML) in retinal image processing for glaucoma diagnosis and detection. ML has proven to be a significant tool for the development of computer-aided technology. Furthermore, secondary research has been widely conducted over the years for ophthalmologists. Such aspects indicate the importance of ML in the context of retinal image processing. The publications that were chosen to compose this review were gathered from Scopus, PubMed, IEEE Xplore and ScienceDirect databases. Then, the papers published between 2014 and 2019 were selected. Research that used the segmented optic disc method was excluded. Moreover, only the methods that applied the classification process were considered. The systematic analysis was performed in such studies and, thereupon, the results were summarized.

Based on architectures used for ML in retinal image processing, some studies applied feature extraction and dimensionality reduction to detect and isolate important parts of the analyzed image. Differently, other works utilized a deep convolutional network. Based on the evaluated research, the main difference between the architectures was the number of images demanded for processing and the high computational cost required to use deep learning techniques.

All of the analyzed publications indicated it was possible to develop an automated system for glaucoma diagnosis. Researchers wrote that the disease’s severity and its high occurrence rates have justified the research that has been carried out. Recent computational techniques, such as deep learning, have been shown to be promising technologies in fundus imaging. However, such techniques require an extensive database and high computational cost, and studies have shown that the data augmentation and transfer learning techniques have been applied as an alternative way to optimize and reduce network training.

Diabetic retinopathy is a microvascular disease associated with changes in peripapillary retinal nerve fiber layer thickness, with the changes being more pronounced in proliferative diabetic retinopathy (PDR) patients undergoing laser photocoagulation. Investigators aimed to assess changes in peripapillary retinal nerve fiber layer thickness in proliferative diabetic retinopathy patients using optical coherence tomography (OCT). A database search was conducted from June 2018 through October 2018, and included PubMed, Medline, Ovid and Google Scholar. A meta-analysis of weighted mean difference and standard deviation was conducted.

A total of 10 studies containing 377 eyes of PDR patients were selected. The analysis of the included studies revealed no significant effect of PRP on average retinal nerve fiber layer thickness using OCT.

Investigators wrote that their meta-analysis revealed no significant effect of PRP on RNFL thickness and that the impact of PRP could vary. They added that the measurement of peripapillary RNFL thickness may yield erroneous and unpredictable results in this subgroup of patients, further confounding the evaluation of nerve fiber layer damage and its progression.