Refine
Has Fulltext
- no (2)
Year of publication
- 2020 (2) (remove)
Document Type
- Article (2)
Language
- English (2)
Is part of the Bibliography
- yes (2)
Keywords
- COVID-19 (1)
- EHR (1)
- TRIPOD (1)
- aging (1)
- autonomic nervous system (1)
- clinical (1)
- cohort (1)
- cortical thickness (1)
- electronic health record (1)
- heart (1)
Institute
Background:
COVID-19 has infected millions of people worldwide and is responsible for several hundred thousand fatalities. The COVID-19 pandemic has necessitated thoughtful resource allocation and early identification of high-risk patients. However, effective methods to meet these needs are lacking.
Objective:
The aims of this study were to analyze the electronic health records (EHRs) of patients who tested positive for COVID-19 and were admitted to hospitals in the Mount Sinai Health System in New York City; to develop machine learning models for making predictions about the hospital course of the patients over clinically meaningful time horizons based on patient characteristics at admission; and to assess the performance of these models at multiple hospitals and time points.
Methods:
We used Extreme Gradient Boosting (XGBoost) and baseline comparator models to predict in-hospital mortality and critical events at time windows of 3, 5, 7, and 10 days from admission. Our study population included harmonized EHR data from five hospitals in New York City for 4098 COVID-19-positive patients admitted from March 15 to May 22, 2020. The models were first trained on patients from a single hospital (n=1514) before or on May 1, externally validated on patients from four other hospitals (n=2201) before or on May 1, and prospectively validated on all patients after May 1 (n=383). Finally, we established model interpretability to identify and rank variables that drive model predictions.
Results:
Upon cross-validation, the XGBoost classifier outperformed baseline models, with an area under the receiver operating characteristic curve (AUC-ROC) for mortality of 0.89 at 3 days, 0.85 at 5 and 7 days, and 0.84 at 10 days. XGBoost also performed well for critical event prediction, with an AUC-ROC of 0.80 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. In external validation, XGBoost achieved an AUC-ROC of 0.88 at 3 days, 0.86 at 5 days, 0.86 at 7 days, and 0.84 at 10 days for mortality prediction. Similarly, the unimputed XGBoost model achieved an AUC-ROC of 0.78 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. Trends in performance on prospective validation sets were similar. At 7 days, acute kidney injury on admission, elevated LDH, tachypnea, and hyperglycemia were the strongest drivers of critical event prediction, while higher age, anion gap, and C-reactive protein were the strongest drivers of mortality prediction.
Conclusions:
We externally and prospectively trained and validated machine learning models for mortality and critical events for patients with COVID-19 at different time horizons. These models identified at-risk patients and uncovered underlying relationships that predicted outcomes.
Understanding the association between autonomic nervous system [ANS] function and brain morphology across the lifespan provides important insights into neurovisceral mechanisms underlying health and disease. Resting-state ANS activity, indexed by measures of heart rate [HR] and its variability [HRV] has been associated with brain morphology, particularly cortical thickness [CT]. While findings have been mixed regarding the anatomical distribution and direction of the associations, these inconsistencies may be due to sex and age differences in HR/HRV and CT. Previous studies have been limited by small sample sizes, which impede the assessment of sex differences and aging effects on the association between ANS function and CT. To overcome these limitations, 20 groups worldwide contributed data collected under similar protocols of CT assessment and HR/HRV recording to be pooled in a mega-analysis (N = 1,218 (50.5% female), mean age 36.7 years (range: 12-87)). Findings suggest a decline in HRV as well as CT with increasing age. CT, particularly in the orbitofrontal cortex, explained additional variance in HRV, beyond the effects of aging. This pattern of results may suggest that the decline in HRV with increasing age is related to a decline in orbitofrontal CT. These effects were independent of sex and specific to HRV; with no significant association between CT and HR. Greater CT across the adult lifespan may be vital for the maintenance of healthy cardiac regulation via the ANS-or greater cardiac vagal activity as indirectly reflected in HRV may slow brain atrophy. Findings reveal an important association between CT and cardiac parasympathetic activity with implications for healthy aging and longevity that should be studied further in longitudinal research.