Development and international validation of custom-engineered and code-free deep-learning models for detection of plus disease in retinopathy of prematurity: a retrospective study.

BACKGROUND: Retinopathy of prematurity (ROP), a leading cause of childhood blindness, is diagnosed through interval screening by paediatric ophthalmologists. However, improved survival of premature neonates coupled with a scarcity of available experts has raised concerns about the sustainability of this approach. We aimed to develop bespoke and code-free deep learning-based classifiers for plus disease, a hallmark of ROP, in an ethnically diverse population in London, UK, and externally validate them in ethnically, geographically, and socioeconomically diverse populations in four countries and three continents. Code-free deep learning is not reliant on the availability of expertly trained data scientists, thus being of particular potential benefit for low resource health-care settings. METHODS: This retrospective cohort study used retinal images from 1370 neonates admitted to a neonatal unit at Homerton University Hospital NHS Foundation Trust, London, UK, between 2008 and 2018. Images were acquired using a Retcam Version 2 device (Natus Medical, Pleasanton, CA, USA) on all babies who were either born at less than 32 weeks gestational age or had a birthweight of less than 1501 g. Each images was graded by two junior ophthalmologists with disagreements adjudicated by a senior paediatric ophthalmologist. Bespoke and code-free deep learning models (CFDL) were developed for the discrimination of healthy, pre-plus disease, and plus disease. Performance was assessed internally on 200 images with the majority vote of three senior paediatric ophthalmologists as the reference standard. External validation was on 338 retinal images from four separate datasets from the USA, Brazil, and Egypt with images derived from Retcam and the 3nethra neo device (Forus Health, Bengaluru, India). FINDINGS: Of the 7414 retinal images in the original dataset, 6141 images were used in the final development dataset. For the discrimination of healthy versus pre-plus or plus disease, the bespoke model had an area under the curve (AUC) of 0·986 (95% CI 0·973-0·996) and the CFDL model had an AUC of 0·989 (0·979-0·997) on the internal test set. Both models generalised well to external validation test sets acquired using the Retcam for discriminating healthy from pre-plus or plus disease (bespoke range was 0·975-1·000 and CFDL range was 0·969-0·995). The CFDL model was inferior to the bespoke model on discriminating pre-plus disease from healthy or plus disease in the USA dataset (CFDL 0·808 [95% CI 0·671-0·909, bespoke 0·942 [0·892-0·982]], p=0·0070). Performance also reduced when tested on the 3nethra neo imaging device (CFDL 0·865 [0·742-0·965] and bespoke 0·891 [0·783-0·977]). INTERPRETATION: Both bespoke and CFDL models conferred similar performance to senior paediatric ophthalmologists for discriminating healthy retinal images from ones with features of pre-plus or plus disease; however, CFDL models might generalise less well when considering minority classes. Care should be taken when testing on data acquired using alternative imaging devices from that used for the development dataset. Our study justifies further validation of plus disease classifiers in ROP screening and supports a potential role for code-free approaches to help prevent blindness in vulnerable neonates. FUNDING: National Institute for Health Research Biomedical Research Centre based at Moorfields Eye Hospital NHS Foundation Trust and the University College London Institute of Ophthalmology. TRANSLATIONS: For the Portuguese and Arabic translations of the abstract see Supplementary Materials section.

Classification and analysis of text transcription from Thai depression assessment tasks among patients with depression.

Depression is a serious mental health disorder that poses a major public health concern in Thailand and have a profound impact on individuals' physical and mental health. In addition, the lack of number to mental health services and limited number of psychiatrists in Thailand make depression particularly challenging to diagnose and treat, leaving many individuals with the condition untreated. Recent studies have explored the use of natural language processing to enable access to the classification of depression, particularly with a trend toward transfer learning from pre-trained language model. In this study, we attempted to evaluate the effectiveness of using XLM-RoBERTa, a pre-trained multi-lingual language model supporting the Thai language, for the classification of depression from a limited set of text transcripts from speech responses. Twelve Thai depression assessment questions were developed to collect text transcripts of speech responses to be used with XLM-RoBERTa in transfer learning. The results of transfer learning with text transcription from speech responses of 80 participants (40 with depression and 40 normal control) showed that when only one question (Q1) of "How are you these days?" was used, the recall, precision, specificity, and accuracy were 82.5%, 84.65, 85.00, and 83.75%, respectively. When utilizing the first three questions from Thai depression assessment tasks (Q1 - Q3), the values increased to 87.50%, 92.11%, 92.50%, and 90.00%, respectively. The local interpretable model explanations were analyzed to determine which words contributed the most to the model's word cloud visualization. Our findings were consistent with previously published literature and provide similar explanation for clinical settings. It was discovered that the classification model for individuals with depression relied heavily on negative terms such as 'not,' 'sad,', 'mood', 'suicide', 'bad', and 'bore' whereas normal control participants used neutral to positive terms such as 'recently,' 'fine,', 'normally', 'work', and 'working'. The findings of the study suggest that screening for depression can be facilitated by eliciting just three questions from patients with depression, making the process more accessible and less time-consuming while reducing the already huge burden on healthcare workers.

Prevalence of refractive error, visual impairment and access to eyecare for the homeless in Wales, United Kingdom.

BACKGROUND/AIMS: To investigate visual impairment and disability, refractive error, and barriers to eye care in the homeless in Cardiff, United Kingdom. METHODS: Prospective cross-sectional study carried out on participants in homeless shelters in Cardiff. We collected participants socio-demographic profile, ocular history and access to eyecare services. Quantitative data included near and distance visual acuity and a non-cycloplegic refraction. RESULTS: A total of 100 participants were studied in this study. Prevalence of myopia was 19% and hyperopia 17%. Mean SE (Spherical Equivalent) for myopia -2.42D (95% CI: -1.65 to -3.19 D), for hyperopia this was +2.22D (95% CI: + 1.66 to +2.79). The prevalence of astigmatism was 36% (mean: 1.67 D, 95% CI: -0.88 to 0.94, n = 100). The number of participants with visual acuity (VA) worse than 6/12 was 11% in comparison to 0.89% and 1.1% in the general Cardiff and Welsh population respectively (p < 0.05). Additionally, 1% of the homeless subjects were registerable as blind (visual acuity worse than 3/60 in the better eye). Barriers to eyecare services were high, with 50% not seen by an optometrist within the last 5 years. CONCLUSIONS: These findings indicate a significant disparity in ocular health, visual acuity and refractive error amongst the homeless in comparison with the general population.

Ten-year survival trends of neovascular age-related macular degeneration at first presentation.

BACKGROUND: To describe 10-year trends in visual outcomes, anatomical outcomes and treatment burden of patients receiving antivascular endothelial growth factor (anti-VEGF) therapy for neovascular age-related macular degeneration (nAMD). METHODS: Retrospective cohort study of treatment-naïve, first-affected eyes with nAMD started on ranibizumab before January 1, 2009. The primary outcome was time to best-corrected visual acuity (BCVA) falling ≤35 ETDRS letters after initiating anti-VEGF therapy. Secondary outcomes included time to BCVA reaching ≥70 letters, proportion of eyes with BCVA ≥70 and ≤35 letters in 10 years, mean trend of BCVA and central retinal thickness over 10 years, and mean number of injections. RESULTS: For our cohort of 103 patients, Kaplan-Meier analyses demonstrated median time to BCVA reaching ≤35 and ≥70 letters were 37.8 (95% CI 22.2 to 65.1) and 8.3 (95% CI 4.8 to 20.9) months after commencing anti-VEGF therapy, respectively. At the final follow-up, BCVA was ≤35 letters and ≥70 letters in 41.1% and 21%, respectively, in first-affected eyes, while this was the case for 5.4% and 48.2%, respectively, in a patient's better-seeing eye. Mean injection number was 37.0±24.2 per eye and 53.6±30.1 at patient level (63.1% of patients required injections in both eyes). CONCLUSIONS: The chronicity of nAMD disease and its management highlights the importance of long-term visual prognosis. Our analyses suggest that one in five patients will retain good vision (BCVA ≥70 ETDRS letters) in the first-affected eye at 10 years after starting anti-VEGF treatment; yet, one in two patients will have good vision in their better-seeing eye. Moreover, our data suggest that early treatment of nAMD is associated with better visual outcomes.

Can Applying a Risk Stratification System, Preoperatively, Reduce Intraoperative Complications during Phacoemulsification?

SIGNIFICANCE: Adopting a risk stratification system and appropriate listing of cases reduces the rates of intraoperative complications during phacoemulsification. Such listing would allow both safe surgery and enhance training, by ensuring an appropriately experienced surgeon is available to operate on a case or supervise and teach a junior surgeon. PURPOSE: To assess if the application of a simple, robust, validated preoperative scoring system can reduce the rates of intraoperative complications if patients are assigned to the appropriately experienced surgeon's list and surgical time is modified accordingly. METHODS: Prospective data collection. One thousand one hundred and thirty five (1135) consecutive patients undergoing phakoemulsification cataract surgery were assessed preoperatively according to weighted criteria. According to the points of risk they accumulated using this system, the patients were preoperatively allocated to one of the four risk groups. The total rate of intraoperative complications for each risk group as well as the rate of each reported complication for each risk group was calculated. RESULTS: The rate of intraoperative complications through the risk groups was 1 = 0.62%, 2 = 0.44%, 3 = 0.18%, and 4 = 0% (P = .005). Thirty-seven per cent 37% (n = 420) of all operations were performed on eyes of patients carrying at least one risk factor for intraoperative complications. The overall rate of any intraoperative complication was 1.2% (n = 14). There was a 0.4% (n = 5) rate of a posterior capsule tear with 2 of these cases with vitreous loss. CONCLUSIONS: Risk stratification, allowing adequate theatre time and appropriate surgical experience, can reduce the rates of intraoperative complications. The risk stratification system allows for better planning of surgical lists and could be used as a transition for those trainees deemed to have sufficient experience for the more 'challenging' cases under adequate supervision.

Prostaglandin F2-Alpha Eye Drops (Bimatoprost) in Graves' Orbitopathy: A Randomized Controlled Double-Masked Crossover Trial (BIMA Trial).

BACKGROUND: Previous in vitro experiments have demonstrated that prostaglandin F2-alpha (PF2α) reduced proliferation and adipogenesis in a murine cell line and human orbital fibroblasts derived from subjects with inactive Graves' orbitopathy (GO). The objective of this study was to determine if the PGF2α analogue bimatoprost is effective at reducing proptosis in this population. METHODS: A randomized controlled double-masked crossover trial was conducted in a single tertiary care academic medical center. Patients with long-standing, inactive GO but persistent proptosis (>20 mm in at least one eye) were recruited. Allowing for a 15% dropout rate, 31 patients (26 females) were randomized in order to identify a treatment effect of 2.0 mm (p = 0.05; power 0.88). Following informed consent, participants were randomized to receive bimatoprost or placebo for three months, after which they underwent a two-month washout before switching to the opposite treatment. The primary outcome was the change in exophthalmometry readings over the two three-month treatment periods. RESULTS: The mean exophthalmometer at baseline was 23.6 mm (range 20.0-30.5 mm), and the mean age of the patients was 55 years (range 28-74 years). The median duration of GO was 7.6 years (interquartile range 3.6-12.3 years). The majority were still suffering from diplopia (61.3%) with bilateral involvement (61.3%). Using multi-level modeling adjusted for baseline, period, and carry-over, bimatoprost resulted in a -0.17 mm (reduction) exophthalmometry change ([confidence interval -0.67 to +0.32]; p = 0.490). There was a mean change in intraocular pressure of -2.7 mmHg ([confidence interval -4.0 to -1.4]; p = 0.0070). One patient showed periorbital fat atrophy on treatment, which resolved on stopping treatment. Independent analysis of proptosis by photographic images (all subjects) and subgroup analysis on monocular disease (n = 12) did not show any apparent benefit. CONCLUSIONS: In inactive GO, bimatoprost treatment over a three-month period does not result in an improvement in proptosis.

Radiation retinopathy treated successfully with aflibercept.

Aflibercept (aflibercept) is a novel anti-vascular endothelial growth factor drug indicated for wet age-related macular degeneration and macular oedema secondary to retinal vein occlusion and diabetic macular oedema. While only newly introduced on the market, it is growing in popularity and over 5.5 million doses have been prescribed worldwide. Due to its versatile mechanism, it is indicated for numerous eye pathologies, and in particular, has been adapted to treat various types of retinopathy. To our knowledge, this is the first case report of solely using aflibercept to treat cystoid macular oedema in radiation retinopathy.