-
Automated Dynamic Bayesian Networks for Predicting Acute Kidney Injury Before Onset
Authors:
David Gordon,
Panayiotis Petousis,
Anders O. Garlid,
Keith Norris,
Katherine Tuttle,
Susanne B. Nicholas,
Alex A. T. Bui
Abstract:
Several algorithms for learning the structure of dynamic Bayesian networks (DBNs) require an a priori ordering of variables, which influences the determined graph topology. However, it is often unclear how to determine this order if feature importance is unknown, especially as an exhaustive search is usually impractical. In this paper, we introduce Ranking Approaches for Unknown Structures (RAUS),…
▽ More
Several algorithms for learning the structure of dynamic Bayesian networks (DBNs) require an a priori ordering of variables, which influences the determined graph topology. However, it is often unclear how to determine this order if feature importance is unknown, especially as an exhaustive search is usually impractical. In this paper, we introduce Ranking Approaches for Unknown Structures (RAUS), an automated framework to systematically inform variable ordering and learn networks end-to-end. RAUS leverages existing statistical methods (Cramers V, chi-squared test, and information gain) to compare variable ordering, resultant generated network topologies, and DBN performance. RAUS enables end-users with limited DBN expertise to implement models via command line interface. We evaluate RAUS on the task of predicting impending acute kidney injury (AKI) from inpatient clinical laboratory data. Longitudinal observations from 67,460 patients were collected from our electronic health record (EHR) and Kidney Disease Improving Global Outcomes (KDIGO) criteria were then applied to define AKI events. RAUS learns multiple DBNs simultaneously to predict a future AKI event at different time points (i.e., 24-, 48-, 72-hours in advance of AKI). We also compared the results of the learned AKI prediction models and variable orderings to baseline techniques (logistic regression, random forests, and extreme gradient boosting). The DBNs generated by RAUS achieved 73-83% area under the receiver operating characteristic curve (AUCROC) within 24-hours before AKI; and 71-79% AUCROC within 48-hours before AKI of any stage in a 7-day observation window. Insights from this automated framework can help efficiently implement and interpret DBNs for clinical decision support. The source code for RAUS is available in GitHub at https://github.com/dgrdn08/RAUS .
△ Less
Submitted 20 April, 2023;
originally announced April 2023.
-
Pushed fronts in a Fisher-KPP-Burgers system using geometric desingularization
Authors:
Matt Holzer,
Matthew Kearney,
Samuel Molseed,
Katie Tuttle,
David Wigginton
Abstract:
We study traveling fronts in a system of one dimensional reaction-diffusion-advection equations motivated by problems in reactive flows. In the limit as a parameter tends to infinity, we construct the approximate front profile and determine the leading order expansion for the selected wavespeed. Such fronts are often constructed as transverse intersections of stable and unstable manifolds of the t…
▽ More
We study traveling fronts in a system of one dimensional reaction-diffusion-advection equations motivated by problems in reactive flows. In the limit as a parameter tends to infinity, we construct the approximate front profile and determine the leading order expansion for the selected wavespeed. Such fronts are often constructed as transverse intersections of stable and unstable manifolds of the traveling wave differential equation. However, a re-scaling of the dependent variable leads to a lack of hyperbolicity for one of the end states making the definition of one such manifold unclear. We use geometric blow-up techniques to recover hyperbolicity and following an analysis of the blown up vector field are able to show the existence of a traveling front with a leading order expansion of its speed.
△ Less
Submitted 14 February, 2024; v1 submitted 13 December, 2022;
originally announced December 2022.
-
Arecibo Multi-Epoch HI Absorption Measurements Against Pulsars: Tiny-Scale Atomic Structure
Authors:
S. Stanimirovic,
J. M. Weisberg,
Z. Pei,
K. Tuttle,
J. T. Green
Abstract:
We present results from multi-epoch neutral hydrogen (HI) absorption observations of six bright pulsars with the Arecibo telescope. Moving through the interstellar medium (ISM) with transverse velocities of 10--150 AU/yr, these pulsars have swept across 1--200 AU over the course of our experiment, allowing us to probe the existence and properties of the tiny scale atomic structure (TSAS) in the co…
▽ More
We present results from multi-epoch neutral hydrogen (HI) absorption observations of six bright pulsars with the Arecibo telescope. Moving through the interstellar medium (ISM) with transverse velocities of 10--150 AU/yr, these pulsars have swept across 1--200 AU over the course of our experiment, allowing us to probe the existence and properties of the tiny scale atomic structure (TSAS) in the cold neutral medium (CNM). While most of the observed pulsars show no significant change in their HI absorption spectra, we have identified at least two clear TSAS-induced opacity variations in the direction of B1929+10. These observations require strong spatial inhomogeneities in either the TSAS clouds' physical properties themselves or else in the clouds' galactic distribution. While TSAS is occasionally detected on spatial scales down to 10 AU, it is too rare to be characterized by a spectrum of turbulent CNM fluctuations on scales of 10-1000 AU, as previously suggested by some work. In the direction of B1929+10, an apparent correlation between TSAS and interstellar clouds inside the warm Local Bubble (LB) indicates that TSAS may be tracing the fragmentation of the LB wall via hydrodynamic instabilities. While similar fragmentation events occur frequently throughout the ISM, the warm medium surrounding these cold cloudlets induces a natural selection effect wherein small TSAS clouds evaporate quickly and are rare, while large clouds survive longer and become a general property of the ISM.
△ Less
Submitted 7 July, 2010;
originally announced July 2010.
