Many clinical applications based on deep learning and pertaining to radiology have been proposed and studied in radiology for classification, risk assessment, segmentation tasks, diagnosis, prognosis, and even prediction of therapy responses. There are many other innovative applications of AI in various technical aspects of medical imaging, particularly applied to the acquisition of images, ranging from removing image artifacts, normalizing/harmonizing images, improving image quality, lowering radiation and contrast dose, and shortening the duration of imaging studies. This article will address this topic and will seek to present an overview of deep learning applied to neuroimaging techniques.
Study design: Retrospective Nationwide Inpatient Sample (NIS) study. Objectives: To determine national trends in prevalence, risk factors and mortality for vertebral column fracture (VCF) and spinal cord injury (SCI) patients with and without acute respiratory distress syndrome/acute lung injury (ARDS/ALI). Setting: United States of America, 1988 to 2008. Methods: The NIS was utilized to select 284 612 admissions for VCF with and without acute SCI from 1988 to 2008 based on ICD-9-CM. The data were stratified for in-hospital complications of ARDS/ALI. Results: Patients with SCI were more likely to develop ARDS/ALI compared with those without (odds ratio (OR): 4.9, 95% confidence interval (CI) 4.7-5.2, Po0.001). Compared with patients with lumbar fractures, those with cervical, thoracic and sacral fractures were more likely to develop ARDS/ALI (Po0.001). ARDS/ALI was statistically more prevalent (Po0.01) in VCF/SCI patients with epilepsy, sepsis, cardiac arrest, congestive heart failure (CHF), hypertension, chronic obstructive pulmonary disease and metabolic disorders. Patients with female gender, surgery at rural practice setting, and coronary artery disease and diabetes were less likely to develop ARDS/ALI (Po0.001). VCF/SCI patients who developed ARDS/ALI were more likely to die in-hospital than those without ARDS/ALI (OR 6.5, 95% CI 6.0-7.1, Po0.001). Predictors of in-hospital mortality after VCF/SCI include: older age, male sex, epilepsy, sepsis, hypertension, CHF, chronic obstructive pulmonary disease and liver disease. Patients who developed ARDS/ALI stayed a mean of 25 hospital days (30-440 days) while patients without ARDS/ALI stayed a mean of 6 days (7-868 days, Po0.001). Conclusion: Our analysis demonstrates that SCI patients are more at risk for ARDS/ALI, which carries a significantly higher risk of mortality.
Background and Purpose:
Focal cerebral arteriopathy (FCA), a common cause of arterial ischemic stroke (AIS) in previously healthy children, often progresses over days to weeks, increasing risk of recurrent stroke. We developed a novel severity scoring system designed to quantify FCA progression and correlate with clinical outcomes.
Methods:
The Vascular effects of Infection in Pediatric Stroke (VIPS) study prospectively enrolled 355 children with AIS (2010–2014), including 41 with centrally confirmed FCA. Two neuroradiologists independently reviewed FCA cerebrovascular imaging, assigning a graded severity score of zero (no involvement) to four (occlusion) to individual arterial segments. The FCA severity score (FCASS) was the unweighted sum. In an iterative process, we modeled scores derived from different combinations of arterial segments to identify the model that optimized correlation with clinical outcome, simplicity and reliability.
Results:
The optimal FCASS summed scores from five arterial segments: supraclinoid internal carotid artery, A1, A2, M1, and M2. The median (interquartile range [IQR]) baseline FCASS was 4 (2, 6). Of 33 children with follow-up imaging, the maximum FCASS (at any time point) was 7 (5, 9). Twenty-four (73%) had FCA progression on follow-up with their maximum FCASS at a median of 8 (5, 35.5) days post-stroke; their median FCASS increase was 4 (2.5, 6). FCASS did not correlate with recurrent AIS. Maximum (but not baseline) FCASS correlated with 1-year Pediatric Stroke Outcome Measures (p=0.037).
Conclusions:
Our novel scoring system for FCA severity correlates with neurologic outcomes in the VIPS cohort, and provides a tool for FCA treatment trials under development.
Background and Purpose:
To assess whether initial imaging characteristics independently predict 1-year neurological outcomes in childhood arterial ischemic stroke patients.
Methods:
We used prospectively collected demographic and clinical data, imaging data, and 1-year outcomes from the VIPS study (Vascular Effects of Infection in Pediatric Stroke). In 288 patients with first-time stroke, we measured infarct volume and location on the acute magnetic resonance imaging studies and hemorrhagic transformation on brain imaging studies during the acute presentation. Neurological outcome was assessed with the Pediatric Stroke Outcome Measure. We used univariate and multivariable ordinal logistic regression models to test the association between imaging characteristics and outcome.
Results:
Univariate analysis demonstrated that infarcts involving uncinate fasciculus, angular gyrus, insular cortex, or that extended from cortex to the subcortical nuclei were significantly associated with poorer outcomes with odds ratios ranging from 1.95 to 3.95. All locations except the insular cortex remained significant predictors of poor outcome on multivariable analysis. When infarct volume was added to the model, the locations did not remain significant. Larger infarct volumes and younger age at stroke onset were significantly associated with poorer outcome, but the strength of the relationships was weak. Hemorrhagic transformation did not predict outcome.
Conclusions:
In the largest pediatric arterial ischemic stroke cohort collected to date, we showed that larger infarct volume and younger age at stroke were associated with poorer outcomes. We made the novel observation that the strength of these associations was modest and limits the ability to use these characteristics to predict outcome in children. Infarcts affecting specific locations were significantly associated with poorer outcomes in univariate and multivariable analyses but lost significance when adjusted for infarct volume. Our findings suggest that infarcts that disrupt critical networks have a disproportionate impact upon outcome after childhood arterial ischemic stroke.
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.