Developing a privacy-preserving deep learning model for glaucoma detection: a multicentre study with federated learning.

BACKGROUND: Deep learning (DL) is promising to detect glaucoma. However, patients' privacy and data security are major concerns when pooling all data for model development. We developed a privacy-preserving DL model using the federated learning (FL) paradigm to detect glaucoma from optical coherence tomography (OCT) images. METHODS: This is a multicentre study. The FL paradigm consisted of a 'central server' and seven eye centres in Hong Kong, the USA and Singapore. Each centre first trained a model locally with its own OCT optic disc volumetric dataset and then uploaded its model parameters to the central server. The central server used FedProx algorithm to aggregate all centres' model parameters. Subsequently, the aggregated parameters are redistributed to each centre for its local model optimisation. We experimented with three three-dimensional (3D) networks to evaluate the stabilities of the FL paradigm. Lastly, we tested the FL model on two prospectively collected unseen datasets. RESULTS: We used 9326 volumetric OCT scans from 2785 subjects. The FL model performed consistently well with different networks in 7 centres (accuracies 78.3%-98.5%, 75.9%-97.0%, and 78.3%-97.5%, respectively) and stably in the 2 unseen datasets (accuracies 84.8%-87.7%, 81.3%-84.8%, and 86.0%-87.8%, respectively). The FL model achieved non-inferior performance in classifying glaucoma compared with the traditional model and significantly outperformed the individual models. CONCLUSION: The 3D FL model could leverage all the datasets and achieve generalisable performance, without data exchange across centres. This study demonstrated an OCT-based FL paradigm for glaucoma identification with ensured patient privacy and data security, charting another course toward the real-world transition of artificial intelligence in ophthalmology.

New Therapeutic Targets for the Prevention and Treatment of Venous Thromboembolism With a Focus on Factor XI Inhibitors.

FXI (factor XI) and FXII (factor XII) have emerged as targets for new anticoagulants that have the potential to be both more efficacious and safer than the currently available direct oral anticoagulants for the prevention and treatment of venous thromboembolism. In this review, we discuss the role of FXI and FXII in the pathogenesis of venous thromboembolism, explain why FXI is a better target, and explain why FXI inhibitors have potential advantages over currently available anticoagulants. Finally, we describe the FXI inhibitors under development and discuss their potential to address unmet needs in venous thromboembolism management.

Clinical Studies with Anticoagulants that Have Changed Clinical Practice.

Anticoagulant therapy is the cornerstone of treatment and prevention of arterial and venous thromboembolism. Taking a historical perspective, starting in the 1960s, and progressing through to 2022, we discuss key clinical trials of anticoagulants that have changed clinical practice, and examine obstacles encountered in bringing these anticoagulants to the clinic. The design of some of the early studies that shaped clinical practice was poor by current standards, but their results were influential because nothing better was available. Both heparin and vitamin K antagonists had been in clinical use for several decades before well-designed trials in the 1980s optimized their dosing and enhanced their safety and efficacy. Low-molecular-weight heparin then replaced unfractionated heparin because it had a more predictable dose-response and a longer half-life, thereby allowing it to be used conveniently in out-of-hospital settings. More recently, direct oral anticoagulants became the oral anticoagulants of choice for most indications because they were shown to be at least as safe and effective as vitamin K antagonists when used in fixed doses without the need for laboratory monitoring. The design of the trials that led to the approval of the direct oral anticoagulants was excellent, but further studies are required to optimize their dosing in selected patients who were underrepresented in these trials.

Risk of Normal Tension Glaucoma Progression From Automated Baseline Retinal-Vessel Caliber Analysis: A Prospective Cohort Study.

PURPOSE: To determine the relationship between baseline retinal-vessel calibers computed by a deep-learning system and the risk of normal tension glaucoma (NTG) progression. DESIGN: Prospective cohort study. METHODS: Three hundred and ninety eyes from 197 patients with NTG were followed up for at least 24 months. Retinal-vessel calibers (central retinal arteriolar equivalent [CRAE] and central retinal venular equivalent [CRVE]) were computed from fundus photographs at baseline using a previously validated deep-learning system. Retinal nerve fiber layer (RNFL) thickness and visual field (VF) were evaluated semiannually. The Cox proportional-hazards model was used to evaluate the relationship of baseline retinal-vessel calibers to the risk of glaucoma progression. RESULTS: Over a mean follow-up period of 34.36 ± 5.88 months, 69 NTG eyes (17.69%) developed progressive RNFL thinning and 22 eyes (5.64%) developed VF deterioration. In the multivariable Cox regression analysis adjusting for age, gender, intraocular pressure, mean ocular perfusion pressure, systolic blood pressure, axial length, standard automated perimetry mean deviation, and RNFL thickness, narrower baseline CRAE (hazard ratio per SD decrease [95% confidence interval], 1.36 [1.01-1.82]) and CRVE (1.35 [1.01-1.80]) were associated with progressive RNFL thinning and narrower baseline CRAE (1.98 [1.17-3.35]) was associated with VF deterioration. CONCLUSION: In this study, each SD decrease in the baseline CRAE or CRVE was associated with a more than 30% increase in the risk of progressive RNFL thinning and a more than 90% increase in the risk of VF deterioration during the follow-up period. Baseline attenuation of retinal vasculature in NTG eyes was associated with subsequent glaucoma progression. High-throughput deep-learning-based retinal vasculature analysis demonstrated its clinical utility for NTG risk assessment.

Three-Dimensional Multi-Task Deep Learning Model to Detect Glaucomatous Optic Neuropathy and Myopic Features From Optical Coherence Tomography Scans: A Retrospective Multi-Centre Study.

Purpose: We aim to develop a multi-task three-dimensional (3D) deep learning (DL) model to detect glaucomatous optic neuropathy (GON) and myopic features (MF) simultaneously from spectral-domain optical coherence tomography (SDOCT) volumetric scans. Methods: Each volumetric scan was labelled as GON according to the criteria of retinal nerve fibre layer (RNFL) thinning, with a structural defect that correlated in position with the visual field defect (i.e., reference standard). MF were graded by the SDOCT en face images, defined as presence of peripapillary atrophy (PPA), optic disc tilting, or fundus tessellation. The multi-task DL model was developed by ResNet with output of Yes/No GON and Yes/No MF. SDOCT scans were collected in a tertiary eye hospital (Hong Kong SAR, China) for training (80%), tuning (10%), and internal validation (10%). External testing was performed on five independent datasets from eye centres in Hong Kong, the United States, and Singapore, respectively. For GON detection, we compared the model to the average RNFL thickness measurement generated from the SDOCT device. To investigate whether MF can affect the model's performance on GON detection, we conducted subgroup analyses in groups stratified by Yes/No MF. The area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and accuracy were reported. Results: A total of 8,151 SDOCT volumetric scans from 3,609 eyes were collected. For detecting GON, in the internal validation, the proposed 3D model had significantly higher AUROC (0.949 vs. 0.913, p < 0.001) than average RNFL thickness in discriminating GON from normal. In the external testing, the two approaches had comparable performance. In the subgroup analysis, the multi-task DL model performed significantly better in the group of "no MF" (0.883 vs. 0.965, p-value < 0.001) in one external testing dataset, but no significant difference in internal validation and other external testing datasets. The multi-task DL model's performance to detect MF was also generalizable in all datasets, with the AUROC values ranging from 0.855 to 0.896. Conclusion: The proposed multi-task 3D DL model demonstrated high generalizability in all the datasets and the presence of MF did not affect the accuracy of GON detection generally.

Inflammation as a Mechanism and Therapeutic Target in Peripheral Artery Disease.

Peripheral artery disease is 1 of 3 major clinical manifestations of atherosclerosis, the other 2 being coronary artery and cerebrovascular disease. Despite progress in surgery, antithrombotic therapy and therapies that modify conventional risk factors (lipid-, blood pressure-, and glucose-lowering interventions), patients with peripheral artery disease have an unacceptably high risk of vascular complications. Additional strategies to reduce this residual risk are needed. The accumulated evidence that inflammation plays an important role in the pathogenesis of atherosclerosis has spurred recent efforts to evaluate anti-inflammatory agents as an additional therapeutic approach for atherothrombosis prevention and treatment. In this review, we examine the evidence supporting the role of inflammation in atherosclerosis, review recent trials of anti-inflammatory approaches to reduce cardiovascular complications, and offer insights into the opportunities for novel anti-inflammatory strategies to reduce the burden of cardiovascular and limb complications in patients with peripheral artery disease.

Can a Single Measurement of Apixaban Levels Identify Patients at Risk of Overexposure? A Prospective Cohort Study.

Background Patients with atrial fibrillation (AF) are frequently treated with apixaban 2.5-mg twice daily (BID) off-label, presumably to reduce the bleeding risk. However, this approach has the potential to increase the risk of ischemic stroke. If a single measurement could reliably identify patients with high drug levels, the increased stroke risk may be mitigated by confining off-label dose reduction to such patients. Objectives This study aimed to determine whether a single high apixaban level is predictive of a similarly high level when the test is repeated in 2 months. Methods In this prospective cohort study of clinic patients receiving apixaban 5-mg BID for AF or venous thromboembolism, peak and trough apixaban levels were measured using the STA-Liquid anti-Xa assay at baseline and 2 months. We calculated the proportions of patients with levels that remained in the upper quintile. Results Of 100 enrolled patients, 82 came for a second visit, 55 of whom were treated with apixaban 5-mg BID. Seven (63.6%, 95% confidence interval [CI]: 35.4-84.8%) and nine (81.8%, 95% CI: 52.3-94.9%) of 11 patients with a baseline trough and peak level in the upper quintile, respectively, had a subsequent level that remained within this range. Only one (9.1%, 95% CI: 1.6-37.7%) patient had a subsequent level that fell just lower than the median. Conclusion The trough and peak levels of apixaban in patients who have a high level on a single occasion, usually remain high when the assay is repeated in 2 months. Accordingly, the finding of a high apixaban level in patients deemed to be at high risk of bleeding, allows physicians contemplating off-label use of the 2.5-mg BID dose to limit its use to selected patients who are less likely to be exposed to an increased risk of thrombosis.

In pregnant women with suspected VTE and low/intermediate or unlikely pretest probability, D-dimer rules out VTE at 3 mo.

SOURCE CITATION: Bellesini M, Robert-Ebadi H, Combescure C, et al. D-dimer to rule out venous thromboembolism during pregnancy: a systematic review and meta-analysis. J Thromb Haemost. 2021;19:2454-67. 34161671.