Machine and deep learning methods for clinical outcome prediction based on physiological data of COVID-19 patients: a scoping review.

INTRODUCTION: Since the beginning of the COVID-19 pandemic, numerous machine and deep learning (MDL) methods have been proposed in the literature to analyze patient physiological data. The objective of this review is to summarize various aspects of these methods and assess their practical utility for predicting various clinical outcomes. METHODS: We searched PubMed, Scopus, and Cochrane Library, screened and selected the studies matching the inclusion criteria. The clinical analysis focused on the characteristics of the patient cohorts in the studies included in this review, the specific tasks in the context of the COVID-19 pandemic that machine and deep learning methods were used for, and their practical limitations. The technical analysis focused on the details of specific MDL methods and their performance. RESULTS: Analysis of the 48 selected studies revealed that the majority (∼54 %) of them examined the application of MDL methods for the prediction of survival/mortality-related patient outcomes, while a smaller fraction (∼13 %) of studies also examined applications to the prediction of patients' physiological outcomes and hospital resource utilization. 21 % of the studies examined the application of MDL methods to multiple clinical tasks. Machine and deep learning methods have been shown to be effective at predicting several outcomes of COVID-19 patients, such as disease severity, complications, intensive care unit (ICU) transfer, and mortality. MDL methods also achieved high accuracy in predicting the required number of ICU beds and ventilators. CONCLUSION: Machine and deep learning methods have been shown to be valuable tools for predicting disease severity, organ dysfunction and failure, patient outcomes, and hospital resource utilization during the COVID-19 pandemic. The discovered knowledge and our conclusions and recommendations can also be useful to healthcare professionals and artificial intelligence researchers in managing future pandemics.

Dataset for Automatic Region-based Coronary Artery Disease Diagnostics Using X-Ray Angiography Images.

X-ray coronary angiography is the most common tool for the diagnosis and treatment of coronary artery disease. It involves the injection of contrast agents into coronary vessels using a catheter to highlight the coronary vessel structure. Typically, multiple 2D X-ray projections are recorded from different angles to improve visualization. Recent advances in the development of deep-learning-based tools promise significant improvement in diagnosing and treating coronary artery disease. However, the limited public availability of annotated X-ray coronary angiography image datasets presents a challenge for objective assessment and comparison of existing tools and the development of novel methods. To address this challenge, we introduce a novel ARCADE dataset with 2 objectives: coronary vessel classification and stenosis detection. Each objective contains 1500 expert-labeled X-ray coronary angiography images representing: i) coronary artery segments; and ii) the locations of stenotic plaques. These datasets will serve as a benchmark for developing new methods and assessing existing approaches for the automated diagnosis and risk assessment of coronary artery disease.

Structural Characterization of La0.6Sr0.4CoO3-δ Thin Films Grown on (100)-, (110)-, and (111)-Oriented La0.95Sr0.05Ga0.95Mg0.05O3-δ.

In this study, a detailed structural characterization of epitaxial La0.6Sr0.4CoO3-δ (LSC) films grown in (100), (110), and (111) orientations was conducted. LSC is a model air electrode material in solid oxide fuel and electrolysis cells and understanding the correlation of bulk structure and catalytic activity is essential for the design of future electrode materials. Thin films were grown on single crystals of the perovskite material La0.95Sr0.05Ga0.95Mg0.05O3-δ cut in three different directions. This enabled an examination of structural details at the atomic scale for a realistic material combination in solid oxide cells. The investigation involved the application of atomic force microscopy, X-ray diffraction, and high-resolution transmission electron microscopy to explore the distinct properties of these thin films. Interestingly, ordering phenomena in both cationic as well as anionic sublattices were found, despite the fact that the thin films were never at higher temperatures than 600 °C. Cationic ordering was found in spherical precipitates, whereas the ordering of oxygen vacancies led to the partial transition to brownmillerite in all three orientations. Our results indicate a very high oxygen vacancy concentration in all three thin films. Lattice strains in-plane and out-of-plane was measured, and its implications for the structural modifications are discussed.

Ca Solubility in a BiFeO3-Based System with a Secondary Bi2O3 Phase on a Nanoscale.

In BiFeO3 (BFO), Bi2O3 (BO) is a known secondary phase, which can appear under certain growth conditions. However, BO is not just an unwanted parasitic phase but can be used to create the super-tetragonal BFO phase in films on substrates, which would otherwise grow in the regular rhombohedral phase (R-phase). The super-tetragonal BFO phase has the advantage of a much larger ferroelectric polarization of 130-150 µC/cm2, which is around 1.5 times the value of the rhombohedral phase with 80-100 µC/cm2. Here, we report that the solubility of Ca, which is a common dopant of bismuth ferrite materials to tune their properties, is significantly lower in the secondary BO phase than in the observed R-phase BFO. Starting from the film growth, this leads to completely different Ca concentrations in the two phases. We show this with advanced analytical transmission electron microscopy techniques and confirm the experimental results with density functional theory (DFT) calculations. At the film's fabrication temperature, caused by different solubilities, about 50 times higher Ca concentration is expected in the BFO phase than in the secondary one. Depending on the cooling rate after fabrication, this can further increase since a larger Ca concentration difference is expected at lower temperatures. When fabricating functional devices using Ca doping and the secondary BO phase, the difference in solubility must be considered because, depending on the ratio of the BO phase, the Ca concentration in the BFO phase can become much higher than intended. This can be critical for the intended device functionality because the Ca concentration strongly influences and modifies the BFO properties.

Recent Advances in Ammonia Gas Sensors Based on Carbon Nanomaterials.

This review paper is devoted to an extended analysis of ammonia gas sensors based on carbon nanomaterials. It provides a detailed comparison of various types of active materials used for the detection of ammonia, e.g., carbon nanotubes, carbon nanofibers, graphene, graphene oxide, and related materials. Different parameters that can affect the performance of chemiresistive gas sensors are discussed. The paper also gives a comparison of the sensing characteristics (response, response time, recovery time, operating temperature) of gas sensors based on carbon nanomaterials. The results of our tests on ammonia gas sensors using various techniques are analyzed. The problems related to the recovery of sensors using various approaches are also considered. Finally, the impact of relative humidity on the sensing behavior of carbon nanomaterials of various different natures was estimated.

Study on Ca Segregation toward an Epitaxial Interface between Bismuth Ferrite and Strontium Titanate.

Segregation is a crucial phenomenon, which has to be considered in functional material design. Segregation processes in perovskite oxides have been the subject of ongoing scientific interest, since they can lead to a modification of properties and a loss of functionality. Many studies in oxide thin films have focused on segregation toward the surface using a variety of surface-sensitive analysis techniques. In contrast, here we report a Ca segregation toward an in-plane compressively strained heterostructure interface in a Ca- and Mn-codoped bismuth ferrite film. We are using advanced transmission electron microscopy techniques, X-ray photoelectron spectroscopy, and density functional theory (DFT) calculations. Ca segregation is found to trigger atomic and electronic structure changes at the interface. This includes the reduction of the interface strain according to the Ca concentration gradient, interplanar spacing variations, and oxygen vacancies at the interface. The experimental results are supported by DFT calculations, which explore two segregation scenarios, i.e., one without oxygen vacancies and Fe oxidation from 3+ to 4+ and one with vacancies for charge compensation. Comparison with electron energy loss spectroscopy (EELS) measurements confirms the second segregation scenario with vacancy formation. The findings contribute to the understanding of segregation and indicate promising effects of a Ca-rich buffer layer in this heterostructure system.

Modeling of Diffusion and Incorporation of Interstitial Oxygen Ions at the TiN/SiO2 Interface.

Silica-based resistive random access memory devices have become an active research area due to complementary metal-oxide-semiconductor compatibility and recent dramatic increases in their performance and endurance. In spite of both experimental and theoretical insights gained into the electroforming process, many atomistic aspects of the set and reset operation of these devices are still poorly understood. Recently a mechanism of electroforming process based on the formation of neutral oxygen vacancies (VO0) and interstitial O ions (Oi2-) facilitated by electron injection into the oxide has been proposed. In this work, we extend the description of the bulk (Oi2-) migration to the interface of amorphous SiO2 with the polycrystaline TiN electrode, using density functional theory simulations. The results demonstrate a strong kinetic and thermodynamic drive for the movement of Oi2- to the interface, with dramatically reduced incorporation energies and migration barriers close to the interface. The arrival of Oi2- at the interface is accompanied by preferential oxidation of undercoordinated Ti sites at the interface, forming a Ti-O layer. We investigate how O ions incorporate into a perfect and defective ∑5(012)[100] grain boundary (GB) in TiN oriented perpendicular to the interface. Our simulations demonstrate the preferential incorporation of Oi at defects within the TiN GB and their fast diffusion along a passivated grain boundary. They explain how, as a result of electroforming, the system undergoes very significant structural changes with the oxide being significantly reduced, interface being oxidized, and part of the oxygen leaving the system.

Release of Mitochondrial and Nuclear DNA During On-Pump Heart Surgery: Kinetics and Relation to Extracellular Vesicles.

During heart surgery with cardiopulmonary bypass (CPB), the release of mitochondrial (mtDNA) and nuclear DNA (nDNA) and their association to extracellular vesicles were investigated. In patients undergoing elective coronary artery bypass grafting (CABG, n = 12), blood was sampled before, during, and after surgery from peripheral artery, pulmonary artery, and the coronary sinus. Plasma was separated in three fractions: microvesicles, exosomes, and supernatant. mtDNA and nDNA were measured by qPCR. mtDNA and nDNA levels increased after start of surgery, but before CPB, and increased further during CPB. mtDNA copy number was about 1000-fold higher than nDNA. mtDNA was predominantly localized to the vesicular fractions in plasma, whereas nDNA was predominantly in the supernatant. The amount of free mtDNA increased after surgery. There was no net release or disappearance of DNAs across the pulmonary, systemic, or coronary circulation. Extracellular DNAs, in particular mtDNA, may be important contributors to the whole-body inflammation during CPB.

Which aspects of HTS are empirically correlated with downstream success?

High-throughput screening (HTS) is a well-established hit-finding approach used in the pharmaceutical industry. In this article, recent experience at Novartis with respect to factors influencing the success of HTS campaigns is discussed. An inherent measure of HTS quality could be defined by the assay Z and Z' factors, the number of hits and their biological potencies; however, such measures of quality do not always correlate with the advancement of hits to the later stages of drug discovery. Also, for many target classes, such as kinases, it is easy to identify hits, but, as a result of selectivity, intellectual property and other issues, the projects do not result in lead declarations. In this article, HTS success is defined as the fraction of HTS campaigns that advance into the later stages of drug discovery, and the major influencing factors are examined. Interestingly, screening compounds in individual wells or in mixtures did not have a major impact on the HTS success and, equally interesting, there was no difference in the progression rates of biochemical and cell-based assays. Particular target types, assay technologies, structure-activity relationships and powder availability had a much greater impact on success as defined above. In addition, significant mutual dependencies can be observed - while one assay format works well with one target type, this situation might be completely reversed for a combination of the same readout technology with a different target type. The results and opinions presented here should be regarded as groundwork, and a plethora of factors that influence the fate of a project, such as biophysical measurements, chemical attractiveness of the hits, strategic reasons and safety pharmacology, are not covered here. Nonetheless, it is hoped that this information will be used industry-wide to improve success rates in terms of hits progressing into exploratory chemistry and beyond. The support that can be obtained from new in silico approaches to phase transitions are also described, along with the gaps they are designed to fill.

Stability of screening compounds in wet DMSO.

The impact of storage conditions on compound stability and compound solubility has been debated intensely over the past 5 years. At Novartis, the authors decided to opt for a storage concept that can be considered controversial because they are using a DMSO/water (90/10) mixture as standard solvent. To assess the effect of water in DMSO stocks on compound stability, the authors monitored the purity of a subset of 1404 compounds from ongoing medicinal chemistry projects over several months. The study demonstrated that 85% of the compounds were stable in wet DMSO over a 2-year period at 4 degrees C. This result validates the storage concept developed at Novartis as a pragmatic approach that takes advantage of the benefits of DMSO/water mixtures while mediating the disadvantages. In addition, the authors describe how purity data collected over the course of the chemical validation of high-throughput screening actives are used to improve the analytical quality of the Novartis screening deck.

Compound Hub: efficiency gains through innovative sample management processes.

While significant investments are made across the industry and increasingly also in academia to enhance or build a compound file, the efficient sourcing of compounds from in-house medical chemistry is frequently seen as a challenge. This article introduces the Compound Hub strategy developed at the Novartis Compound Archive. Central Compound Hubs in Basel and Cambridge were combined with web-based ordering of compounds and assays, providing assay-ready, solubilized samples to labs anywhere in the global research organization. Relieving scientists from time-intensive sample preparation tasks, error rates could be reduced through electronic processing and tracking of compounds/assays and the capture of medicinal chemistry compounds for the compound library could be increased by 75%.

Heparinase treatment of heparin-contaminated plasma from coronary artery bypass grafting patients enables reliable quantification of microRNAs.

BACKGROUND: microRNAs have recently been identified as powerful biomarkers of human disease. Reliable polymerase chain reaction (PCR)-based quantification of nucleic acids in clinical samples contaminated with polymerase inhibitor heparin requires deheparinization. However, the effects of deheparinization procedure on quantification of nucleic acids remain largely unknown. The aim of this study was to determine whether the deheparinization procedure completely eliminates the inhibition of amplification, while maintaining RNA integrity and technical variability of the measured microRNA levels. METHODS: Heparinized plasma from 9 patients undergoing coronary artery bypass grafting (CABG) and the heparin-free plasma from 58 rats were spiked with a synthetic RNA oligonucleotide and total RNA was extracted. The RNA solutions were then treated with heparinase I to remove contaminating heparin prior to reverse transcription. Levels of synthetic spike-in RNA oligonucleotide, as well as endogenous hsa-miR-1-3p and hsa-miR-208a-3p, were measured using quantitative reverse transcription PCR (RT-qPCR). The amplification efficiency and presence of inhibitors in individual samples were directly determined using calibration curves. RESULTS: In contrast to RNA samples from rat plasma, RNA samples derived from the CABG patient plasma contained inhibitors, which were completely eliminated by treatment with heparinase. The procedure caused a decrease in the amount of detected RNA; however, the technical variability of the measured targets did not change, allowing for the quantification of circulating endogenous hsa-miR-1-3p and hsa-miR-208a-3p in the plasma of CABG patients. CONCLUSIONS: The heparinase treatment procedure enables utilization of RT-qPCR for reliable microRNA quantification in heparinized plasma.

Insulin assembly damps conformational fluctuations: Raman analysis of amide I linewidths in native states and fibrils.

The crystal structure of insulin has been investigated in a variety of dimeric and hexameric assemblies. Interest in dynamics has been stimulated by conformational variability among crystal forms and evidence suggesting that the functional monomer undergoes a conformational change on receptor binding. Here, we employ Raman spectroscopy and Raman microscopy to investigate well-defined oligomeric species: monomeric and dimeric analogs in solution, native T(6) and R(6) hexamers in solution and corresponding polycrystalline samples. Remarkably, linewidths of Raman bands associated with the polypeptide backbone (amide I) exhibit progressive narrowing with successive self-assembly. Whereas dimerization damps fluctuations at an intermolecular beta-sheet, deconvolution of the amide I band indicates that formation of hexamers stabilizes both helical and non-helical elements. Although the structure of a monomer in solution resembles a crystallographic protomer, its encagement in a native assembly damps main-chain fluctuations. Further narrowing of a beta-sheet-specific amide I band is observed on reorganization of insulin in a cross-beta fibril. Enhanced flexibility of the native insulin monomer is in accord with molecular dynamics simulations. Such conformational fluctuations may initiate formation of an amyloidogenic nucleus and enable induced fit on receptor binding.

Key aspects of the Novartis compound collection enhancement project for the compilation of a comprehensive chemogenomics drug discovery screening collection.

The NIBR (Novartis Institutes for BioMedical Research) compound collection enrichment and enhancement project integrates corporate internal combinatorial compound synthesis and external compound acquisition activities in order to build up a comprehensive screening collection for a modern drug discovery organization. The main purpose of the screening collection is to supply the Novartis drug discovery pipeline with hit-to-lead compounds for today's and the future's portfolio of drug discovery programs, and to provide tool compounds for the chemogenomics investigation of novel biological pathways and circuits. As such, it integrates designed focused and diversity-based compound sets from the synthetic and natural paradigms able to cope with druggable and currently deemed undruggable targets and molecular interaction modes. Herein, we will summarize together with new trends published in the literature, scientific challenges faced and key approaches taken at NIBR to match the chemical and biological spaces.

Recent trends and observations in the design of high-quality screening collections.

The design of a high-quality screening collection is of utmost importance for the early drug-discovery process and provides, in combination with high-quality assay systems, the foundation of future discoveries. Herein, we review recent trends and observations to successfully expand the access to bioactive chemical space, including the feedback from hit assessment interviews of high-throughput screening campaigns; recent successes with chemogenomics target family approaches, the identification of new relevant target/domain families, diversity-oriented synthesis and new emerging compound classes, and non-classical approaches, such as fragment-based screening and DNA-encoded chemical libraries. The role of in silico library design approaches are emphasized.

Sequential superparamagnetic clustering for unbiased classification of high-dimensional chemical data.

For the clustering of chemical structures that are described by the Similog, ISIS count, and ISIS binary fingerprints, we propose a sequential superparamagnetic clustering approach. To appropriately handle nonbinary feature keys, we introduce an extension of the binary Tanimoto similarity measure. In our applications, data sets composed of structures from seven chemically distinct compound classes are evaluated and correctly clustered. The comparison, with results from leading methods, indicates the superiority of our sequential superparamagnetic clustering approach.