Locating Youth Exposed to Parental Justice Involvement in the Electronic Health Record: Development of a Natural Language Processing Model.

BACKGROUND: Parental justice involvement (eg, prison, jail, parole, or probation) is an unfortunately common and disruptive household adversity for many US youths, disproportionately affecting families of color and rural families. Data on this adversity has not been captured routinely in pediatric health care settings, and if it is, it is not discrete nor able to be readily analyzed for purposes of research. OBJECTIVE: In this study, we outline our process training a state-of-the-art natural language processing model using unstructured clinician notes of one large pediatric health system to identify patients who have experienced a justice-involved parent. METHODS: Using the electronic health record database of a large Midwestern pediatric hospital-based institution from 2011-2019, we located clinician notes (of any type and written by any type of provider) that were likely to contain such evidence of family justice involvement via a justice-keyword search (eg, prison and jail). To train and validate the model, we used a labeled data set of 7500 clinician notes identifying whether the patient was ever exposed to parental justice involvement. We calculated the precision and recall of the model and compared those rates to the keyword search. RESULTS: The development of the machine learning model increased the precision (positive predictive value) of locating children affected by parental justice involvement in the electronic health record from 61% (a simple keyword search) to 92%. CONCLUSIONS: The use of machine learning may be a feasible approach to addressing the gaps in our understanding of the health and health services of underrepresented youth who encounter childhood adversities not routinely captured-particularly for children of justice-involved parents.

A multiplexed ion-exchange membrane-based miRNA (MIX·miR) detection platform for rapid diagnosis of myocardial infarction.

Micro RNAs (miRNAs) have shown great potential as rapid and discriminating biomarkers for acute myocardial infarction (AMI) diagnosis. We have developed a multiplexed ion-exchange membrane-based miRNA (MIX·miR) preconcentration/sensing amplification-free platform for quantifying in parallel a panel of miRNAs, including miR-1, miR-208b, and miR-499, from the same plasma samples from: 1) reference subjects with no evident coronary artery disease (NCAD); 2) subjects with stable coronary artery disease (CAD); and 3) subjects experiencing ST-elevation myocardial infarction (STEMI) prior to (STEMI-pre) and following (STEMI-PCI) percutaneous coronary intervention. The picomolar limit of detection from raw plasma and 3-decade dynamic range of MIX·miR permits detection of the miRNA panel in untreated samples from disease patients and its precise standard curve, provided by large 0.1 to 1 V signals and eliminates individual sensor calibration. The use of molecular concentration feature reduces the assay time to less than 30 minutes and increases the detection sensitivity by bringing all targets close to the sensors. miR-1 was low for NCAD patients but more than one order of magnitude above the normal value for all samples from three categories (CAD, STEMI-pre, and STEMI-PCI) of patients with CAD. In fact, miR-1 expression levels of stable CAD, STEMI-pre and STEMI-PCI are each more than 10-fold higher than the previous class, in that order, well above the 95% confidence level of MIX·miR. Its overexpression estimate is significantly higher than the PCR benchmark. This suggests that, in contrast to protein biomarkers of myocardial injury, miR-1 appears to differentiate ischemia from both reperfusion injury and non-AMI CAD patients. The battery-operated MIX·miR can be a portable and low-cost AMI diagnostic device, particularly useful in settings where cardiac catheterization is not readily available to determine the status of coronary reperfusion.

Integrated Biophysical Characterization of Fibrillar Collagen-Based Hydrogels.

This paper describes an experimental characterization scheme of the biophysical properties of reconstituted hydrogel matrices based on indentation testing, quantification of transport via microfluidics, and confocal reflectance microscopy analysis. While methods for characterizing hydrogels exist and are widely used, they often do not measure diffusive and convective transport concurrently, determine the relationship between microstructure and transport properties, and decouple matrix mechanics and transport properties. Our integrated approach enabled independent and quantitative measurements of the structural, mechanical, and transport properties of hydrogels in a single study. We used fibrillar type I collagen as the base matrix and investigated the effects of two different matrix modifications: (1) cross-linking with human recombinant tissue transglutaminase II (hrTGII) and (2) supplementation with the nonfibrillar matrix constituent hyaluronic acid (HA). hrTGII modified the matrix structure and transport but not mechanical parameters. Furthermore, changes in the matrix structure due to hrTGII were seen to be dependent on the concentration of collagen. In contrast, supplementation of HA at different collagen concentrations altered the matrix microstructure and mechanical indentation behavior but not transport parameters. These experimental observations reveal the important relationship between extracellular matrix (ECM) composition and biophysical properties. The integrated techniques are versatile, robust, and accessible; and as matrix-cell interactions are instrumental for many biological processes, the methods and findings described here should be broadly applicable for characterizing hydrogel materials used for three-dimensional (3-D) tissue-engineered culture models.