Purpose The objective of the study is to develop deep learning models using synthetic fundus images to assess the direction (intorsion versus extorsion) and amount (physiologic versus pathologic) of static ocular torsion. Static ocular torsion assessment is an important clinical tool for classifying vertical ocular misalignment; however, current methods are time-intensive with steep learning curves for frontline providers. Methods We used a dataset ( n = 276) of right eye fundus images. The disc-foveal angle was calculated using ImageJ to generate synthetic images via image rotation. Using synthetic datasets ( n = 12,740 images per model) and transfer learning (the reuse of a pretrained deep learning model on a new task), we developed a binary classifier (intorsion versus extorsion) and a multiclass classifier (physiologic versus pathologic intorsion and extorsion). Model performance was evaluated on unseen synthetic and nonsynthetic data. Results On the synthetic dataset, the binary classifier had an accuracy and area under the receiver operating characteristic curve (AUROC) of 0.92 and 0.98, respectively, whereas the multiclass classifier had an accuracy and AUROC of 0.77 and 0.94, respectively. The binary classifier generalized well on the nonsynthetic data (accuracy = 0.94; AUROC = 1.00). Conclusions The direction of static ocular torsion can be detected from synthetic fundus images using deep learning methods, which is key to differentiate between vestibular misalignment (skew deviation) and ocular muscle misalignment (superior oblique palsies). Translational Relevance Given the robust performance of our models on real fundus images, similar strategies can be adopted for deep learning research in rare neuro-ophthalmologic diseases with limited datasets.
The critical closing pressure (PCRIT), a quantitative assessment of upper airway collapsibility, is derived from pressure flow relationship during sleep. The analytic generation of the pressure flow relationships are non-standardized due to various regression models (linear, spline, median), breath characteristics (flow limited, non-flow limited) or known covariates (sleep stage, body position). We propose a GUI based PCRIT Analysis Software (PAS) to streamline PCRIT analysis and validate its reliability and accuracy compared to current analysis procedures. Methods Seventeen subjects underwent a physiology sleep study in which the PCRIT was determined during NREM sleep. Data analysis was performed independently using three paradigms: 1) PAS (Igor Pro; median regression), 2) non-graphic statistics application (SAS; spline regression), and 3) manual spreadsheet calculations (Excel; linear regression). The reliability and accuracy of the PAS was examined through the agreement between each approach using Bland-Altman plots of the mean difference and within-individual variation using intra-class correlation (ICC). Results There was no difference in the group mean values for PCRIT using the PAS (−1.7±0.7 cm H2O) compared to spline regression (−1.6±0.7 cm H2O; p=0.69) or linear regression (−2.1±0.7 cm H2O; p=0.92). The Bland-Altman analysis did not demonstrate a systematic bias between the PAS and either approach. There was a mean difference of 0.39±0.2 cm H2O between the PAS and linear regression approaches, with upper and lower limits of agreement of 1.81 and −1.02 cm H2O, respectively. The PAS and spline analysis demonstrated an even smaller mean difference of −0.10 ± 0.1 cm H2O, with upper and lower limits of 0.90 and −1.08 cm H2O, respectively. Conclusion PAS preserves the reliability and accuracy of the original PCRIT analysis methods while vastly improving their efficiency through graphic user interface and automation of analytic processes. Providing a standardized platform for physiologic data processing offers the ability to implement quality assurance and control procedures for multicenter studies as well as cost saving by improving the efficiency of complex repetitive tasks.
There have been over 621 million cases of COVID-19 worldwide with over 6.5 million deaths. Despite the high secondary attack rate of COVID-19 in shared households, some exposed individuals do not contract the virus. In addition, little is known about whether the occurrence of COVID-19 resistance differs among people by health characteristics as stored in the electronic health records (EHR). In this retrospective analysis, we develop a statistical model to predict COVID-19 resistance in 8,536 individuals with prior COVID-19 exposure using demographics, diagnostic codes, outpatient medication orders, and count of Elixhauser comorbidities in EHR data from the COVID-19 Precision Medicine Platform Registry. Cluster analyses identified 5 patterns of diagnostic codes that distinguished resistant from non-resistant patients in our study population. In addition, our models showed modest performance in predicting COVID-19 resistance (best performing model AUROC = 0.61). Monte Carlo simulations conducted indicated that the AUROC results are statistically significant (p < 0.001) for the testing set. We hope to validate the features found to be associated with resistance/non-resistance through more advanced association studies.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.