A Machine Learning Methodology for Identification and Triage of Heart Failure Exacerbations.

Inadequate at-home management and self-awareness of heart failure (HF) exacerbations are known to be leading causes of the greater than 1 million estimated HF-related hospitalizations in the USA alone. Most current at-home HF management protocols include paper guidelines or exploratory health applications that lack rigor and validation at the level of the individual patient. We report on a novel triage methodology that uses machine learning predictions for real-time detection and assessment of exacerbations. Medical specialist opinions on statistically and clinically comprehensive, simulated patient cases were used to train and validate prediction algorithms. Model performance was assessed by comparison to physician panel consensus in a representative, out-of-sample validation set of 100 vignettes. Algorithm prediction accuracy and safety indicators surpassed all individual specialists in identifying consensus opinion on existence/severity of exacerbations and appropriate treatment response. The algorithms also scored the highest sensitivity, specificity, and PPV when assessing the need for emergency care. Here we develop a machine-learning approach for providing real-time decision support to adults diagnosed with congestive heart failure. The algorithm achieves higher exacerbation and triage classification performance than any individual physician when compared to physician consensus opinion.

Utilization of the Signature Method to Identify the Early Onset of Sepsis From Multivariate Physiological Time Series in Critical Care Monitoring.

OBJECTIVES: Patients in an ICU are particularly vulnerable to sepsis. It is therefore important to detect its onset as early as possible. This study focuses on the development and validation of a new signature-based regression model, augmented with a particular choice of the handcrafted features, to identify a patient's risk of sepsis based on physiologic data streams. The model makes a positive or negative prediction of sepsis for every time interval since admission to the ICU. DESIGN: The data were sourced from the PhysioNet/Computing in Cardiology Challenge 2019 on the "Early Prediction of Sepsis from Clinical Data." It consisted of ICU patient data from three separate hospital systems. Algorithms were scored against a specially designed utility function that rewards early predictions in the most clinically relevant region around sepsis onset and penalizes late predictions and false positives. SETTING: The work was completed as part of the PhysioNet 2019 Challenge alongside 104 other teams. PATIENTS: PhysioNet sourced over 60,000 ICU patients with up to 40 clinical variables for each hour of a patient's ICU stay. The Sepsis-3 criteria was used to define the onset of sepsis. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The algorithm yielded a utility function score which was the first placed entry in the official phase of the challenge.

A machine learning approach to triaging patients with chronic obstructive pulmonary disease.

COPD patients are burdened with a daily risk of acute exacerbation and loss of control, which could be mitigated by effective, on-demand decision support tools. In this study, we present a machine learning-based strategy for early detection of exacerbations and subsequent triage. Our application uses physician opinion in a statistically and clinically comprehensive set of patient cases to train a supervised prediction algorithm. The accuracy of the model is assessed against a panel of physicians each triaging identical cases in a representative patient validation set. Our results show that algorithm accuracy and safety indicators surpass all individual pulmonologists in both identifying exacerbations and predicting the consensus triage in a 101 case validation set. The algorithm is also the top performer in sensitivity, specificity, and ppv when predicting a patient's need for emergency care.

Model for self-polarization and motility of keratocyte fragments.

Computational modelling of cell motility on substrates is a formidable challenge; regulatory pathways are intertwined and forces that influence cell motion are not fully quantified. Additional challenges arise from the need to describe a moving deformable cell boundary. Here, we present a simple mathematical model coupling cell shape dynamics, treated by the phase-field approach, to a vector field describing the mean orientation (polarization) of the actin filament network. The model successfully reproduces the primary phenomenology of cell motility: discontinuous onset of motion, diversity of cell shapes and shape oscillations. The results are in qualitative agreement with recent experiments on motility of keratocyte cells and cell fragments. The asymmetry of the shapes is captured to a large extent in this simple model, which may prove useful for the interpretation of experiments.

Dynamic internal gradients control and direct electric currents within nanostructured materials.

Switchable nanomaterials--materials that can change their properties and/or function in response to external stimuli-have potential applications in electronics, sensing and catalysis. Previous efforts to develop such materials have predominately used molecular switches that can modulate their properties by means of conformational changes. Here, we show that electrical conductance through films of gold nanoparticles coated with a monolayer of charged ligands can be controlled by dynamic, long-range gradients of both mobile counterions surrounding the nanoparticles and conduction electrons on the nanoparticle cores. The internal gradients and the electric fields they create are easily reconfigurable, and can be set up in such a way that electric currents through the nanoparticles can be modulated, blocked or even deflected so that they only pass through select regions of the material. The nanoion/counterion hybrids combine the properties of electronic conductors with those of ionic gels/polymers, are easy to process by solution-casting and, by controlling the internal gradients, can be reconfigured into different electronic elements (current rectifiers, switches and diodes).

Conformation and mechanical properties of closed diblock fibers.

We analyze the conformations and mechanical properties of closed diblock fibers. In our model, the length fraction of each component and the total fiber length are controlled by tunable chemical potentials. Our formalism can describe fibers in which one block is a bare polymer while the other is an adsorbed protein-filament complex; these blocks maintain different bending rigidities and spontaneous curvatures. We analytically calculate the shape of two-component polymers for all values of the material parameters and chemical potentials. Our results yield a complete analytical description of all possible two-component polymer conformations, a phase portrait detailing the parameter spaces in which these shapes occur, and the identification of spontaneous transitions between shapes driven by environmental changes.

Motor-mediated alignment of microtubules in semidilute mixtures.

We propose and study a model of molecular motor-induced ordering in a cytoskeletal filament solution for the semidilute case. Motors attach to a pair of filaments and walk along the pair bringing them into closer alignment. In the semidilute regime multiple motors can bind a filament to several others and, for a critical motor density, induce a transition to an ordered phase with a nonzero mean orientation. The motors, on the one hand, cause closer filament alignment, and, on the other hand, induce fluctuations that are dependent on the relative orientation of the filaments to which the motors are attached. We develop a spatially homogenous, mean-field theory that explicitly accounts for a force-dependent detachment rate of motors, which in turn affects the mean and the fluctuations of the net force acting on a filament. This model considers each filament to be in motor contact with all other filaments in the solution. We show that the transition to the oriented state changes from second order to first order when the force-dependent detachment becomes important.

Bundle dynamics of interacting polar rods.

We use a probabilistic model of microtubule interaction via molecular motors to study microtubule bundle interaction. Our model indicates that initially disordered systems of interacting polar rods exhibit an orientational instability resulting in spontaneous ordering. We study the existence and dynamic interaction of microtubule bundles analytically and numerically. Our results show a long term attraction and coalescing of bundles indicating a clear coarsening in the system; Microtubule bundles concentrate into fewer orientations on a slow logarithmic time scale.