Diagnostic and prognostic capabilities of a biomarker and EMR-based machine learning algorithm for sepsis.

Sepsis is a major cause of mortality among hospitalized patients worldwide. Shorter time to administration of broad-spectrum antibiotics is associated with improved outcomes, but early recognition of sepsis remains a major challenge. In a two-center cohort study with prospective sample collection from 1400 adult patients in emergency departments suspected of sepsis, we sought to determine the diagnostic and prognostic capabilities of a machine-learning algorithm based on clinical data and a set of uncommonly measured biomarkers. Specifically, we demonstrate that a machine-learning model developed using this dataset outputs a score with not only diagnostic capability but also prognostic power with respect to hospital length of stay (LOS), 30-day mortality, and 3-day inpatient re-admission both in our entire testing cohort and various subpopulations. The area under the receiver operating curve (AUROC) for diagnosis of sepsis was 0.83. Predicted risk scores for patients with septic shock were higher compared with patients with sepsis but without shock (p < 0.0001). Scores for patients with infection and organ dysfunction were higher compared with those without either condition (p < 0.0001). Stratification based on predicted scores of the patients into low, medium, and high-risk groups showed significant differences in LOS (p < 0.0001), 30-day mortality (p < 0.0001), and 30-day inpatient readmission (p < 0.0001). In conclusion, a machine-learning algorithm based on electronic medical record (EMR) data and three nonroutinely measured biomarkers demonstrated good diagnostic and prognostic capability at the time of initial blood culture.

Unravelling the secret of seed-based gels in water: the nanoscale 3D network formation.

Chia (Salvia hispanica) and basil (Ocimum basilicum) seeds have the intrinsic ability to form a hydrogel concomitant with moisture-retention, slow releasing capability and proposed health benefits such as curbing diabetes and obesity by delaying digestion process. However, the underlying mode of gelation at nanoscopic level is not clearly explained or explored. The present study elucidates and corroborates the hypothesis that the gelling behavior of such seeds is due to their nanoscale 3D-network formation. The preliminary study revealed the influence of several conditions like polarity, pH and hydrophilicity/hydrophobicity on fiber extrusion from the seeds which leads to gelation. Optical microscopic analysis clearly demonstrated bundles of fibers emanating from the seed coat while in contact with water, and live growth of fibers to form 3D network. Scanning electron microscope (SEM) and transmission electron microscope (TEM) studies confirmed 3D network formation with fiber diameters ranging from 20 to 50 nm.

Mass transfer in the biomolecular binding of a target against probe molecules on the surface of microbeads sequestered in wells in a microfluidic cell.

Diagnostic tools which screen the binding interactions of a protein target against a display of biomolecular probes to identify molecules which bind the target are central to cell proteomic studies, and to diagnostic assays. Here, we study a microfluidic design for screening interactions in which the probe molecules are hosted on microbeads sequestered in wells arranged at the bottom of a microfluidic flow channel. Assays are undertaken by streaming an analyte solution with a fluorescently labelled target through the cell, and identifying the fluorescing beads. Numerical simulations are first constructed for the analyte flow over the microbeads in the well array, and the increase in the target concentration on the microbead surface. The binding profile is expressed as a function of the ratio of the convective to the diffusive transport rates (Peclet number or Pe), and the ratio of the kinetic to the diffusive rates (Damkohler number, Da). For any Pe, as Da becomes small enough, the transport is determined by the intrinsic kinetic binding rate. As Pe increases, a thin concentration boundary layer develops over the top surface of the microbead because of the convective flow, and target binds more rapidly. However, the relatively stagnant layers of liquid in the well provide a diffusion barrier which slows the target transport, and for any Da and Pe the transport is slower than equivalent patches of probes arranged on the channel wall. Experiments are also undertaken at high Pe, using the binding of fluorescently labelled NeutrAvidin as a target to probes of its binding partner, biotin, on the microbead surface. The binding profile is compared to the simulations to measure the kinetic rate constant, and this comparison shows that the transport in the cell is not kinetically limited because of the diffusion barriers created by the stagnant liquid layer in the well. Simulations and experiments on microbeads which are only partially recessed in the well demonstrate an increase in the mass transfer rate as more of the microbead surface intersects the flow and the diffusion limitation due to the stagnant layer of liquid surrounding the bottom part of the microbead is minimized.