A deep learning model for clinical outcome prediction using longitudinal inpatient electronic health records.

Objective: Recent advances in deep learning show significant potential in analyzing continuous monitoring electronic health records (EHR) data for clinical outcome prediction. We aim to develop a Transformer-based, Encounter-level Clinical Outcome (TECO) model to predict mortality in the intensive care unit (ICU) using inpatient EHR data.

Materials And Methods: TECO was developed using multiple baseline and time-dependent clinical variables from 2579 hospitalized COVID-19 patients to predict ICU mortality, and was validated externally in an ARDS cohort (n=2799) and a sepsis cohort (n=6622) from the Medical Information Mart for Intensive Care (MIMIC)-IV.

Case Series Evaluating the Relationship of SGLT2 Inhibition to Pulmonary Artery Pressure and Non-Invasive Cardiopulmonary Parameters in HFpEF/HFmrEF Patients-A Pilot Study.

Unlabelled: The initiation of sodium-glucose cotransporter 2 (SGLT2) inhibitor treatment was shown to reduce pulmonary artery pressure (PAP) in New York Heart Association (NYHA) class III heart failure (HF) patients with an implanted PAP sensor. We aimed to investigate the impact of SGLT2-I initiation on pulmonary vascular resistance (PVR), pulmonary capillary wedge pressure (PCWP), pulmonary arterial capacitance (PAC), and right ventricle (RV) to PA (RV-PA) coupling in a pilot cohort of HF with preserved/mildly reduced ejection fraction (HFpEF/HFmrEF) patients and whether PVR and PCWP can be serially calculated non-invasively using PAP sensor data during follow-up.

Methods: Right heart catheterization parameters (PVR, PCWP, and PAC) were obtained at sensor implantation and echocardiographic assessments (E/E', RV-PA coupling, and RV cardiac output) were made at baseline and every 3 months.

Synthesis and Characterization of Solvated and Base-Free Cerium(III) Mixed-Sandwich Complexes.

A series of solvated complexes: [(CH)Ln(CMeR)(DME)] Ln = La, R = -Me; () Ln = Ce, R = -Me, -SiMe, -H; (, , ), [(dbCOT)Ln(CMe)(DME)] (), (dbCOT = dibenzocyclooctadienide; Ln = Ce, La), [(dbCOT)Ce(CMeH)(DME)] () and [(hdcCOT)Ce(CMe)(DME)] () (hdcCOT = hexahydrodicyclopentacyclooctatetraenide) and base-free mixed-sandwich complexes [(CH)Ce(CMe)] (), [(CH)Ce(CMeH) () and [(hdcCOT)Ce(CMe)] () of the early lanthanide metals cerium and lanthanum comprising variable cyclopentadienide (Cp) and cyclooctatetraenide (COT) ligands is described. To evaluate the effect of cyclopentadienide and cyclooctatetraenide ligands on the characteristics of these complexes, their solid-state structural, electrochemical, and photophysical properties were studied and accompanied by theoretical calculations. To further evaluate the effect of ligands on the topology of the complexes and the reducing properties of the complexes, syntheses of several base-free congeners were pursued, which led to isolation of the first base-free monomeric and polymeric Ce(III) mixed-sandwich compounds.

Development of bone mineral density and content in children with cerebral palsy: a retrospective, longitudinal study.

Objectives: The aim of this study was to investigate long-term bone development in children with cerebral palsy (CP) using longitudinal measurements of total body less head bone mineral content (TBLH-BMC) and bone mineral density (TBLH-BMD).

Methods: A retrospective longitudinal analysis was performed on 109 children with CP who participated in a rehabilitation programme from 2006 to 2018. Dual-energy X-ray absorptiometry (DXA) scans were performed at the beginning of the programme and repeated as clinically indicated.

Localized Reconstruction of Multimodal Distance Distribution from DEER Data of Biopolymers.

Pulsed Dipolar ESR Spectroscopy (PDS) is a uniquely powerful technique to characterize the structural property of intrinsically disordered proteins (IDPs) and polymers and the conformational evolution of IDPs and polymers, e.g. during assembly, by offering the probability distribution of segment end-to-end distances.