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.
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.
INTRODUCTION/HYPOTHESIS: Delirium following an acute stroke event is associated with increased morbidity and can lead to longer lengths of hospital stay and increased mortality. These patients may experience additional negative functional outcomes if delirium is not appropriately assessed and managed. Evaluation of the assessment and treatment of stroke patients may lead to antipsychotic stewardship opportunities for pharmacists. The goal of this study was to determine the prevalence of bCAM assessments and correlation with treatment in delirious patients within the stroke population. The primary outcome of this study is the number of patients that were bCAM negative and received antipsychotics. Secondary outcomes include number of documented bCAM assessments per day, duration in days of positive bCAM, number of antipsychotics used during admission, the frequency of use for each antipsychotic, and the frequency of antipsychotics prescribed at discharge.
METHODS:A retrospective chart review was conducted to identify stroke patients with delirium who had documented bCAM assessments during their hospitalization. Data from 159 patients admitted to the non-ICU neurology units at Cleveland Clinic Main Campus between May 1, 2016 and July 31, 2019 were collected. Patient demographics were collected and bCAM assessments and antipsychotic agents used during the stay were analyzed.
RESULTS:Sixty-five patients (median age 74 years) with a stroke diagnosis were included in this study. Patients had a median National Institutes of Health Stroke Scale (NIHSS) score of 7.5. Ten patients had underlying psychiatric disorders including dementia, schizophrenia, depression, and bipolar disorder. Twenty-four patients (44.4%) received antipsychotics when they were bCAM negative. Fifty-one (78.5%) patients received at least one antipsychotic during admission. The most frequently used antipsychotic agent was quetiapine (55.4%). Twenty-three patients (35.4%) were discharged on antipsychotic medications.
CONCLUSIONS:The findings of this study identify important opportunities for pharmacists to help steward the use of antipsychotic agents in stroke patients with delirium, including identifying patients who may not need antipsychotics on discharge and recommending taper plans to prevent longterm or unnecessary antipsychotic use.
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.