Refine
Has Fulltext
- no (2)
Document Type
- Article (2)
Language
- English (2)
Is part of the Bibliography
- yes (2)
Keywords
- Bayesian modeling (1)
- COVID-19 (1)
- EHR (1)
- East Africa (1)
- Kenya Rift (1)
- Magnetostratigraphy (1)
- Paleolimnology (1)
- Pleistocene (1)
- Radiogenic isotopes (1)
- Sedimentology (1)
The Olorgesailie Drilling Project and the related Hominin Sites and Paleolakes Drilling Project in East Africa were initiated to test hypotheses and models linking environmental change to hominin evolution by drilling lake basin sediments adjacent to important archeological and paleoanthropological sites. Drill core OL012-1A recovered 139 m of sedimentary and volcaniclastic strata from the Koora paleolake basin, southern Kenya Rift, providing the opportunity to compare paleoenvironmental influences over the past million years with the parallel record exposed at the nearby Olorgesailie archeological site. To refine our ability to link core-to-outcrop paleoenvironmental records, we institute here a methodological framework for deriving a robust age model for the complex lithostratigraphy of OL012-1A. Firstly, chronostratigraphic control points for the core were established based on 4 Ar/39Ar ages from intercalated tephra deposits and a basal trachyte flow, as well as the stratigraphic position of the Brunhes-Matuyama geomagnetic reversal. This dataset was combined with the position and duration of paleosols, and analyzed using a new Bayesian algorithm for high-resolution age-depth modeling of hiatus-bearing stratigraphic sections. This model addresses three important aspects relevant to highly dynamic, nonlinear depositional environments: 1) correcting for variable rates of deposition, 2) accommodating hiatuses, and 3) quantifying realistic age uncertainty with centimetric resolution. Our method is applicable to typical depositional systems in extensional rifts as well as to drill cores from other dynamic terrestrial or aquatic environments. We use the core age model and lithostratigraphy to examine the inter connectivity of the Koora Basin to adjacent areas and sources of volcanism. (C) 2019 Elsevier Ltd. All rights reserved.
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.