Temporal convolutional networks and data rebalancing for clinical length of stay and mortality prediction.

It is critical for hospitals to accurately predict patient length of stay (LOS) and mortality in real-time. We evaluate temporal convolutional networks (TCNs) and data rebalancing methods to predict LOS and mortality. This is a retrospective cohort study utilizing the MIMIC-III database. The MIMIC-Extract pipeline processes 24 hour time-series clinical objective data for 23,944 unique patient records. TCN performance is compared to both baseline and state-of-the-art machine learning models including logistic regression, random forest, gated recurrent unit with decay (GRU-D). Models are evaluated for binary classification tasks (LOS > 3 days, LOS > 7 days, mortality in-hospital, and mortality in-ICU) with and without data rebalancing and analyzed for clinical runtime feasibility. Data is split temporally, and evaluations utilize tenfold cross-validation (stratified splits) followed by simulated prospective hold-out validation. In mortality tasks, TCN outperforms baselines in 6 of 8 metrics (area under receiver operating characteristic, area under precision-recall curve (AUPRC), and F-1 measure for in-hospital mortality; AUPRC, accuracy, and F-1 for in-ICU mortality). In LOS tasks, TCN performs competitively to the GRU-D (best in 6 of 8) and the random forest model (best in 2 of 8). Rebalancing improves predictive power across multiple methods and outcome ratios. The TCN offers strong performance in mortality classification and offers improved computational efficiency on GPU-enabled systems over popular RNN architectures. Dataset rebalancing can improve model predictive power in imbalanced learning. We conclude that temporal convolutional networks should be included in model searches for critical care outcome prediction systems.

On collaborative reinforcement learning to optimize the redistribution of critical medical supplies throughout the COVID-19 pandemic.

OBJECTIVE: This work investigates how reinforcement learning and deep learning models can facilitate the near-optimal redistribution of medical equipment in order to bolster public health responses to future crises similar to the COVID-19 pandemic. MATERIALS AND METHODS: The system presented is simulated with disease impact statistics from the Institute of Health Metrics, Centers for Disease Control and Prevention, and Census Bureau. We present a robust pipeline for data preprocessing, future demand inference, and a redistribution algorithm that can be adopted across broad scales and applications. RESULTS: The reinforcement learning redistribution algorithm demonstrates performance optimality ranging from 93% to 95%. Performance improves consistently with the number of random states participating in exchange, demonstrating average shortage reductions of 78.74 ± 30.8% in simulations with 5 states to 93.50 ± 0.003% with 50 states. CONCLUSIONS: These findings bolster confidence that reinforcement learning techniques can reliably guide resource allocation for future public health emergencies.

The toxin goniodomin, produced by Alexandrium spp., is identical to goniodomin A.

In 1968 Burkholder and associates (J. Antibiot. (Tokyo)1968, 21, 659-664) isolated the antifungal toxin goniodomin from an unidentified Puerto Rican dinoflagellate and partially characterized its structure. Subsequently, a metabolite of Alexandrium hiranoi was isolated by Murakami et al. from a bloom in Japan and its structure was established (Tetrahedron Lett.1988, 29, 1149-1152). The Japanese substance had strong similarities to Burkholder's but due to uncertainty as to whether it was identical or only similar, Murakami named his toxin goniodomin A. A detailed study of this question now provides compelling evidence that Burkholder's goniodomin is identical to goniodomin A. Morphological characterization of the dinoflagellate suggests that it was the genus Alexandrium but insufficient evidence is available to make a definite identification of the species. This is the only report of goniodomin in the Caribbean region.

Maternal Central Blood Pressure Is Associated with Fetal Middle Cerebral Artery Dopplers.

Cardiovascular adaptations to pregnancy involve physiological mechanisms that increase cardiac output, decrease total vascular resistance, and decrease both systolic and diastolic blood pressure (BP). These maternal hemodynamic changes modulate uteroplacental blood flow and fetal-placental Doppler indices. Our objective was to create maternal cardiac profiles of pregnant women using non-invasive measurements of central BP to identify changes in maternal-fetal hemodynamics as a surrogate to fetal status. This was a prospective cohort study of all singleton pregnancies in a perinatal referral center between January and April 2018. Central BP was measured non-invasively using the BP+ device. The BP+ device is a supra-systolic oscillometric central BP device, which measures BP waveforms peripherally and calculates central BP. We compared various BP+ values for peripheral BP with central BP and stratified by gestational age. We investigated the correlations between peripheral BP, central BP, estimated fetal weight (EFW), and the pulsatility indices (PI) of Doppler velocimetry and demonstrate that both central systolic and diastolic BP correlated to peripheral systolic and diastolic BP. Linear regression analysis confirmed that central BP predicts the middle cerebral artery (MCA) PI. The MCA PI correlated with EFW, specifically higher central systolic BP is associated with a lower MCA PI, implying a possible etiology of fetal brain shunting with poor placental perfusion. Future studies using predictors and markers of fetal outcomes from maternal cardiac parameters should consider maternal cardiovascular measurements to peripheral arterial BP.

Maternal Fontan procedure is a predictor of a small-for-gestational-age neonate: a 10-year retrospective study.

BACKGROUND: Women with single ventricle cardiac physiologic condition who have undergone Fontan procedures are surviving well into reproductive age and historically have been discouraged from pregnancy, despite the paucity of data regarding maternal and neonatal outcomes. OBJECTIVE: Our primary objective was to investigate, in a large cohort, the maternal and neonatal outcomes of pregnant women who have undergone the Fontan procedure and to understand maternal and neonatal sequelae of their pregnancies. STUDY DESIGN: This single-center retrospective cohort study involves pregnant women with a Fontan palliation who delivered at UCLA Medical Center over a 10-year period (2007-2017). All pregnancies were evaluated for differences in maternal and neonatal characteristics. RESULTS: We identified 37 distinct pregnancies in 24 women with a Fontan procedure. The physiologic pregnancy-related increase in cardiac output is blunted substantially in Fontan circulation. Third-trimester cardiac index positively correlated to birthweight z-score (R2=0.48; P=.038) but not to small for gestational age (R2=0.13; P=.339). The most common cardiac complications in pregnancies of >24 weeks gestation were sustained arrhythmia (37.5%) and decompensated heart failure (21%). The 37 pregnancies comprised 25 live births (67.6%), 1 fetal death (2.7%), 9 spontaneous abortions (24%), and 2 pregnancy terminations (5.4%). Of the live births, 60% were preterm at an average gestational age of 34.9±3.7 weeks. Newborn infants were delivered via cesarean in 53%, operative vaginal delivery in 28%, and spontaneous vaginal delivery in 20%. Forty percent of neonates were born small (<10th percentile) for gestational age; 44.0% of all neonates were admitted to the neonatal intensive care unit. CONCLUSION: Women with a single ventricle and Fontan circulation can have a successful pregnancy, although they are at increased risk for arrhythmias and heart failure. The decreased cardiac reserve in these pregnancies blunts the normal increase in maternal cardiac output, which is associated with preterm delivery and small-for-gestational-age neonates. Further studies are needed to determine to what extent the impaired rise in maternal cardiac output reduces uteroplacental perfusion, placental exchange, fetal growth, and onset of parturition.

NOTCH1 is a mechanosensor in adult arteries.

Endothelial cells transduce mechanical forces from blood flow into intracellular signals required for vascular homeostasis. Here we show that endothelial NOTCH1 is responsive to shear stress, and is necessary for the maintenance of junctional integrity, cell elongation, and suppression of proliferation, phenotypes induced by laminar shear stress. NOTCH1 receptor localizes downstream of flow and canonical NOTCH signaling scales with the magnitude of fluid shear stress. Reduction of NOTCH1 destabilizes cellular junctions and triggers endothelial proliferation. NOTCH1 suppression results in changes in expression of genes involved in the regulation of intracellular calcium and proliferation, and preventing the increase of calcium signaling rescues the cell-cell junctional defects. Furthermore, loss of Notch1 in adult endothelium increases hypercholesterolemia-induced atherosclerosis in the descending aorta. We propose that NOTCH1 is atheroprotective and acts as a mechanosensor in adult arteries, where it integrates responses to laminar shear stress and regulates junctional integrity through modulation of calcium signaling.

The SAR of brain penetration for a series of heteroaryl urea FAAH inhibitors.

The SAR of brain penetration for a series of heteroaryl piperazinyl- and piperadinyl-urea fatty acid amide hydrolase (FAAH) inhibitors is described. Brain/plasma (B/P) ratios ranging from >4:1 to as low as 0.02:1 were obtained through relatively simple structural changes to various regions of the heteroaryl urea scaffold. It was not possible to predict the degree of central nervous system (CNS) penetration from the volumes of distribution (Vd) obtained from pharmacokinetic (PK) experiments as very high Vds did not correlate with high B/P ratios. Similarly, calculated topological polar surface areas (TPSAs) did not consistently correlate with the degree of brain penetration. The lowest B/P ratios were observed for those compounds that were significantly ionized at physiological pH. However, as this class of compounds inhibits the FAAH enzyme through covalent modification, low B/P ratios did not preclude effective central target engagement.

Preclinical Characterization of the FAAH Inhibitor JNJ-42165279.

The pre-clinical characterization of the aryl piperazinyl urea inhibitor of fatty acid amide hydrolase (FAAH) JNJ-42165279 is described. JNJ-42165279 covalently inactivates the FAAH enzyme, but is highly selective with regard to other enzymes, ion channels, transporters, and receptors. JNJ-42165279 exhibited excellent ADME and pharmacodynamic properties as evidenced by its ability to block FAAH in the brain and periphery of rats and thereby cause an elevation of the concentrations of anandamide (AEA), oleoyl ethanolamide (OEA), and palmitoyl ethanolamide (PEA). The compound was also efficacious in the spinal nerve ligation (SNL) model of neuropathic pain. The combination of good physical, ADME, and PD properties of JNJ-42165279 supported it entering the clinical portfolio.

Heteroarylureas with spirocyclic diamine cores as inhibitors of fatty acid amide hydrolase.

A series of mechanism based heteroaryl urea fatty acid amide hydrolase (FAAH) inhibitors with spirocyclic diamine cores is described. A potent member of this class, (37), was found to inhibit FAAH centrally, elevate the brain levels of three fatty acid ethanolamides [FAAs: anandamide (AEA), oleoyl ethanolamide (OEA) and palmitoyl ethanolamide (PEA)], and was moderately efficacious in a rat model of neuropathic pain.

Ultrasensitive gas-phase chemical sensing based on functionalized photonic crystal nanobeam cavities.

Photonic crystal nanobeam cavities with high-quality factors are very sensitive to the changes of the dielectric properties of their surroundings. Utilizing this high sensitivity and by applying chemical functionalization, an ultrasensitive chemical sensor for gases based on a nanobeam cavity was demonstrated. A limit of detection of 1.5 parts-per-billion (ppb) in ambient conditions, determined from the noise level of the system, was achieved for nerve agent simulant methyl salicylate. The nanobeam cavity's nonlinear thermo-optical bistability is also utilized to realize a threshold detector for cumulative chemical exposure.

Heteroaryl urea inhibitors of fatty acid amide hydrolase: structure-mutagenicity relationships for arylamine metabolites.

The structure-activity relationships for a series of heteroaryl urea inhibitors of fatty acid amide hydrolase (FAAH) are described. Members of this class of inhibitors have been shown to inactivate FAAH by covalent modification of an active site serine with subsequent release of an aromatic amine from the urea electrophile. Systematic Ames II testing guided the optimization of urea substituents by defining the structure-mutagenicity relationships for the released aromatic amine metabolites. Potent FAAH inhibitors were identified having heteroaryl amine leaving groups that were non-mutagenic in the Ames II assay.

Ulcerative disease outbreak in crayfish Orconectes propinquus linked to Saprolegnia australis in big Muskellunge Lake, Wisconsin.

Crayfish populations in the area of the North Temperate Lakes Long Term Ecological Research (LTER) project, Wisconsin, USA, have been monitored for >25 yr. In 2005, native crayfish Orconectes propinquus from Big Muskellunge Lake were found with ulcerated lesions in the cuticle. In 2006, lesions occurred in 9.5% of sampled crayfish from the lake (n=3146). Ulcers generally occurred on the appendages of affected individuals but varied in location and severity. The prevalence of ulcers varied widely among sites, sample depths, and sampling dates, ranging from < 2% to >20%. The prevalence of ulcers in crayfish increased from a minimum in early June to a maximum in late July and August. In aquarium trials, healthy crayfish representing either O. propinquus or O. rusticus co-housed with ulcerated crayfish did not develop ulcers within 4 wk of exposure. Gross and histopathologic analyses of ulcerated crayfish revealed the presence of filamentous hyphae in the lesions while hemocytic infiltrates, melanotic reactions and silver-stained sections indicated that the ulcers had an oomycete etiology. Excised samples of ulcerated crayfish cuticle grown in culture developed an oomycete that was identified as Saprolegnia australis by PCR amplification and sequence analysis of 2 different DNA fragments. This is the first report of the occurrence of ulcers in wild crayfish associated with S. australis infection in the U.S.A. The advent of the outbreak and its underlying ecological causes are still under investigation.