ABSTRACT
OBJECTIVE: Diabetic macular edema (DME) is the primary cause of vision loss among individuals with diabetes mellitus (DM). We developed, validated, and tested a deep learning (DL) system for classifying DME using images from three common commercially available optical coherence tomography (OCT) devices. RESEARCH DESIGN AND METHODS: We trained and validated two versions of a multitask convolution neural network (CNN) to classify DME (center-involved DME [CI-DME], non-CI-DME, or absence of DME) using three-dimensional (3D) volume scans and 2D B-scans, respectively. For both 3D and 2D CNNs, we used the residual network (ResNet) as the backbone. For the 3D CNN, we used a 3D version of ResNet-34 with the last fully connected layer removed as the feature extraction module. A total of 73,746 OCT images were used for training and primary validation. External testing was performed using 26,981 images across seven independent data sets from Singapore, Hong Kong, the U.S., China, and Australia. RESULTS: In classifying the presence or absence of DME, the DL system achieved area under the receiver operating characteristic curves (AUROCs) of 0.937 (95% CI 0.920-0.954), 0.958 (0.930-0.977), and 0.965 (0.948-0.977) for the primary data set obtained from CIRRUS, SPECTRALIS, and Triton OCTs, respectively, in addition to AUROCs >0.906 for the external data sets. For further classification of the CI-DME and non-CI-DME subgroups, the AUROCs were 0.968 (0.940-0.995), 0.951 (0.898-0.982), and 0.975 (0.947-0.991) for the primary data set and >0.894 for the external data sets. CONCLUSIONS: We demonstrated excellent performance with a DL system for the automated classification of DME, highlighting its potential as a promising second-line screening tool for patients with DM, which may potentially create a more effective triaging mechanism to eye clinics.
Subject(s)
Deep Learning , Diabetes Mellitus , Diabetic Retinopathy , Macular Edema , Diabetic Retinopathy/diagnostic imaging , Humans , Macular Edema/diagnostic imaging , ROC Curve , Tomography, Optical CoherenceABSTRACT
The coronavirus disease 2019 (COVID-19) pandemic has brought unprecedented challenges to ophthalmology. At least 16 ophthalmologists worldwide have succumbed to COVID-19. It reflects the susceptibility of ophthalmologists to COVID-19 infection as they are in close proximity to patients. This chapter provides an overview of the ocular manifestations of COVID-19, risks of COVID-19 to ophthalmologists and patients, clinical service adjustments due to COVID-19, and infection control measures to minimize the transmission of COVID-19 in ophthalmic practice.
Subject(s)
COVID-19 , Ophthalmology , Humans , Infection Control , Pandemics , SARS-CoV-2ABSTRACT
PURPOSE: In order to minimize the risk of transmission of coronavirus disease 2019 (COVID-19), many clinic appointments were postponed to lower patient attendance. Actively calling patients to postpone appointments is a labour-intensive process. We were the first ophthalmic clinics in Hong Kong to use short message service (SMS) to dispatch messages simultaneously to a large number of patients to offer postponement of appointments. The aim of this study is to evaluate whether SMS is an effective method to reduce outpatient attendance during the COVID-19 pandemic. METHODS: This is an observational study reviewing data on SMS messages sent to all patients attending ophthalmology clinics of a tertiary eye centre in Hong Kong. All SMS were sent at least 5 days before the scheduled appointments. The text message included an enquiry hotline for postponement of appointments and offered drug refill. The study included data from February to April 2020. Two hundred patients were invited to take part in a questionnaire on satisfaction level and reason(s) for appointment rescheduling. RESULTS: During the study period, a total of 17,028 SMS were sent. The overall response rate was 23.6%. 14.3% postponed their appointments. This led to an overall 13.9% reduction of clinic attendance. The overall satisfaction was high (96%). The main reason for postponing appointment was worries about infection risk (93.1%). CONCLUSION: SMS was an efficient and cost-effective flow-control method which was well accepted by patients and can reduce outpatient attendance. The time saved can potentially allow healthcare workers to conduct other infection control measures during the COVID-19 pandemic.
Subject(s)
COVID-19/epidemiology , Eye Diseases/epidemiology , Pandemics , Patient Compliance/statistics & numerical data , Text Messaging , Comorbidity , Female , Hong Kong/epidemiology , Humans , Male , SARS-CoV-2Subject(s)
Betacoronavirus , Pneumonia, Viral , COVID-19 , Coronavirus Infections , Eye , Humans , Pandemics , SARS-CoV-2ABSTRACT
PURPOSE: Coronavirus disease (COVID-19) has rapidly emerged as a global health threat. The purpose of this article is to share our local experience of stepping up infection control measures in ophthalmology to minimise COVID-19 infection of both healthcare workers and patients. METHODS: Infection control measures implemented in our ophthalmology clinic are discussed. The measures are based on detailed risk assessment by both local ophthalmologists and infection control experts. RESULTS: A three-level hierarchy of control measures was adopted. First, for administrative control, in order to lower patient attendance, text messages with an enquiry phone number were sent to patients to reschedule appointments or arrange drug refill. In order to minimise cross-infection of COVID-19, a triage system was set up to identify patients with fever, respiratory symptoms, acute conjunctivitis or recent travel to outbreak areas and to encourage these individuals to postpone their appointments for at least 14 days. Micro-aerosol generating procedures, such as non-contact tonometry and operations under general anaesthesia were avoided. Nasal endoscopy was avoided as it may provoke sneezing and cause generation of droplets. All elective clinical services were suspended. Infection control training was provided to all clinical staff. Second, for environmental control, to reduce droplet transmission of COVID-19, installation of protective shields on slit lamps, frequent disinfection of equipment, and provision of eye protection to staff were implemented. All staff were advised to measure their own body temperatures before work and promptly report any symptoms of upper respiratory tract infection, vomiting or diarrhoea. Third, universal masking, hand hygiene, and appropriate use of personal protective equipment (PPE) were promoted. CONCLUSION: We hope our initial experience in stepping up infection control measures for COVID-19 infection in ophthalmology can help ophthalmologists globally to prepare for the potential community outbreak or pandemic. In order to minimise transmission of COVID-19, ophthalmologists should work closely with local infection control teams to implement infection control measures that are appropriate for their own clinical settings.
Subject(s)
Coronavirus Infections/prevention & control , Disease Outbreaks , Eye Diseases , Ophthalmology , Pandemics/prevention & control , Pneumonia, Viral/prevention & control , Ambulatory Care Facilities , Betacoronavirus , COVID-19 , Coronavirus Infections/epidemiology , Coronavirus Infections/transmission , Hong Kong , Humans , Ophthalmology/instrumentation , Pneumonia, Viral/epidemiology , Pneumonia, Viral/transmission , SARS-CoV-2 , TriageABSTRACT
Purpose. To evaluate the recent trends in demographics, risk factors, and microbiological profiles of microbial keratitis at a university hospital in Hong Kong. Design. Retrospective review. Methods. The medical records of 51 patients admitted to the Prince of Wales Hospital for microbial keratitis from January 2010 to June 2012 were reviewed. Demographics, risk factors, clinical features, microbiological results, and treatment were recorded. Data was analyzed and compared to our historical sampled data collected 11 years ago. Results. The mean age of patients was 41.6 ± 20.3 years. Contact lens use was the major risk factor (45%), followed by injury (12%). The culture positive rate was 59%, of which 37% were Gram-positive organisms and 53% were Gram-negative organisms. Pseudomonas aeruginosa (50%) and coagulase-negative Staphylococcus (13%) were the most commonly isolated pathogens. No resistance to fluoroquinolones was identified. Conclusions. Our study showed that contact lens wear remained the major risk factor for microbial keratitis in Hong Kong and Pseudomonas aeruginosa remained the commonest bacterium isolated. This is comparable to our historical data and other studies conducted in East Asia.