Relevant Classes Research Articles

Intelligent methods for natural data analysis: application to space weather

The paper describes methods for detecting anomalies in geophysical monitoring data. This work studies a highly relevant class of problems in this area, aimed at creating methods for space weather forecasting. The negative impact of space weather anomalies on human health and practically all modern infrastructure objects requires the development of methods and the creation of effective means of detecting anomalies. Threshold wavelet filtering methods widely used for data analysis and anomaly detection allow one to obtain fairly accurate estimates using a “greed” strategy, even in the case of incomplete noise data. With this approach, the signal is estimated by isolating coherent structures. But these methods have high computational complexity, failing to provide accurate estimates when the signal-to-noise ratio is low. For such signals, we propose using adaptive probabilistic thresholds. Threshold function parameters are introduced that make it possible to estimate the variability of the process, suppress noise, and detect nonstationary features of different time-frequency structures. The paper also considers ways to combine threshold wavelet filtering with neural networks of the NARX and Autoencoder architectures. Schemes for the implementation of such approaches in the problems of detecting space weather anomalies are proposed. Using the problem of detecting ionospheric anomalies as an example, the efficiency of combining threshold wavelet filtering with the NARX network is shown. The efficiency of network sharing an Autoencoder with adaptive wavelet threshold filtering is shown in the problem of anomaly detection in cosmic ray flux intensity data.

Numerical solutions to linear transfer problems of polarized radiation

Context. A relevant class of radiative transfer problems for polarized radiation is linear, or can be linearized, and can thus be reframed as linear systems once discretized. In this context, depending on the considered physical models, there are both highly coupled and computationally expensive problems, for which state-of-the-art iterative methods struggle to converge, and lightweight ones, for which solutions can be obtained efficiently. Aims. This work aims to exploit lightweight physical models as preconditioners for iterative solution strategies to obtain accurate and fast solutions for more complex problems. Methods. We considered a highly coupled linear transfer problem for polarized radiation, which we solved iteratively using a matrix-free generalized minimal residual (GMRES) method. Different preconditioners and initial guesses, designed in a physics-based framework, are proposed and analyzed. The action of preconditioners was also computed by applying GMRES. The overall approach thus consists of two nested GMRES iterations, one for the original problem and one for its lightweight version. As a benchmark, we considered the modeling of the intensity and polarization of the solar Ca I 4227 Å line, the Sr II 4077 Å line, and the Mg II h&k lines in a semi-empirical 1D atmospheric model, accounting for partial frequency redistribution effects in scattering processes and considering a general angle-dependent treatment. Results. Numerical experiments show that using tailored preconditioners based on simplified models of the considered problem has a noticeable impact, reducing the number of iterations to convergence by a factor of 10–20. Conclusions. By designing efficient preconditioners in a physics-based context, it is possible to significantly improve the convergence of iterative processes, obtaining fast and accurate numerical solutions to the considered problems. The presented approach is general, requiring only the selection of an appropriate lightweight model, and can be applied to a larger class of radiative transfer problems in combination with arbitrary iterative procedures.

Linking antibiotic resistance gene patterns with advanced faecal pollution assessment and environmental key parameters along 2300 km of the Danube River

The global spread of antimicrobial resistance (AMR) in the environment is a growing health threat. Large rivers are of particular concern as they are highly impacted by wastewater discharge while being vital lifelines serving various human needs. A comprehensive understanding of occurrence, spread and key drivers of AMR along whole river courses is largely lacking. We provide a holistic approach by studying spatiotemporal patterns and hotspots of antibiotic resistance genes (ARGs) along 2311 km of the navigable Danube River, combining a longitudinal and temporal monitoring campaign. The integration of advanced faecal pollution diagnostics and environmental and chemical key parameters allowed linking ARG concentrations to the major pollution sources and explaining the observed patterns. Nine AMR markers, including genes conferring resistance to five different antibiotic classes of clinical and environmental relevance, and one integrase gene were determined by probe-based qPCR. All AMR targets could be quantified in Danube River water, with intI1 and sul1 being ubiquitously abundant, qnrS, tetM, blaTEM with intermediate abundance and blaOXA-48like, blaCTX−M-1 group, blaCTX−M-9 group and blaKPC genes with rare occurrence. Human faecal pollution from municipal wastewater discharges was the dominant factor shaping ARG patterns along the Danube River. Other significant correlations of specific ARGs were observed with discharge, certain metals and pesticides. In contrast, intI1 was not associated with wastewater but was already established in the water microbiome. Animal contamination was detected only sporadically and was correlated with ARGs only in the temporal sampling set. During temporal monitoring, an extraordinary hotspot was identified emphasizing the variability within natural waters. This study provides the first comprehensive baseline concentrations of ARGs in the Danube River and lays the foundation for monitoring future trends and evaluating potential reduction measures. The applided holistic approach proved to be a valuable methodological contribution towards a better understanding of the environmental occurrence of AMR.

Eat healthy, feel better: Are differences in employees' longitudinal healthy-eating trajectories reflected in better psychological well-being?

Eating healthily in terms of fruit and vegetable consumption has beneficial effects for employees and their organisations. Yet, we know little about how employees' eating behaviour develops over longer periods of time (trajectories) as well as about how subgroups of employees in these trajectories differ (trajectory classes). Gaining such insights is critical to understand how employees address healthy eating recommendations over time as well as to develop individualised interventions that also consider the development of healthy eating (i.e. improvement versus impairment beyond mean levels). We analysed panel data (Longitudinal Internet Studies for the Social Sciences) from 1054 employees by means of growth mixture modelling. Our analyses revealed three relevant classes of healthy-eating trajectories: a favourable trajectory class, an unfavourable trajectory class and a strongly improving trajectory class. Furthermore, unfavourable healthy-eating trajectories were especially critical with respect to impaired psychological well-being. Specifically, we found robust results for impaired positive and negative affects, but not for self-esteem, in the unfavourable trajectory class. We discuss limitations and implications of these findings, thereby encouraging research and practice to further consider such fine-grained approaches (i.e. focusing on subgroups within a larger population) when addressing healthy-eating promotion over time.

Digital image analysis and machine learning-assisted prediction of neoadjuvant chemotherapy response in triple-negative breast cancer

BackgroundPathological complete response (pCR) is associated with favorable prognosis in patients with triple-negative breast cancer (TNBC). However, only 30–40% of TNBC patients treated with neoadjuvant chemotherapy (NAC) show pCR, while the remaining 60–70% show residual disease (RD). The role of the tumor microenvironment in NAC response in patients with TNBC remains unclear. In this study, we developed a machine learning-based two-step pipeline to distinguish between various histological components in hematoxylin and eosin (H&E)-stained whole slide images (WSIs) of TNBC tissue biopsies and to identify histological features that can predict NAC response.MethodsH&E-stained WSIs of treatment-naïve biopsies from 85 patients (51 with pCR and 34 with RD) of the model development cohort and 79 patients (41 with pCR and 38 with RD) of the validation cohort were separated through a stratified eightfold cross-validation strategy for the first step and leave-one-out cross-validation strategy for the second step. A tile-level histology label prediction pipeline and four machine-learning classifiers were used to analyze 468,043 tiles of WSIs. The best-trained classifier used 55 texture features from each tile to produce a probability profile during testing. The predicted histology classes were used to generate a histology classification map of the spatial distributions of different tissue regions. A patient-level NAC response prediction pipeline was trained with features derived from paired histology classification maps. The top graph-based features capturing the relevant spatial information across the different histological classes were provided to the radial basis function kernel support vector machine (rbfSVM) classifier for NAC treatment response prediction.ResultsThe tile-level prediction pipeline achieved 86.72% accuracy for histology class classification, while the patient-level pipeline achieved 83.53% NAC response (pCR vs. RD) prediction accuracy of the model development cohort. The model was validated with an independent cohort with tile histology validation accuracy of 83.59% and NAC prediction accuracy of 81.01%. The histological class pairs with the strongest NAC response predictive ability were tumor and tumor tumor-infiltrating lymphocytes for pCR and microvessel density and polyploid giant cancer cells for RD.ConclusionOur machine learning pipeline can robustly identify clinically relevant histological classes that predict NAC response in TNBC patients and may help guide patient selection for NAC treatment.

Hall classes of groups

In 1958, Philip Hall (Ill J Math 2:787–801, 1958) proved that if a group G has a nilpotent normal subgroup N such that G/N′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$G/N'$$\\end{document} is nilpotent, then G is nilpotent. The scope of Hall’s nilpotency criterion is not restricted to group theory, and in fact similar statements hold for Lie algebras and more generally for algebraically coherent semiabelian categories (see Chao in Math Z 103:40–42, 1968; Gray in Adv Math 349:911–919, 2019; Stitzinger in Ill J Math 22:499–505, 1978). We say that a group class X\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\\mathfrak {X}}$$\\end{document} is a Hall class if it contains every group G admitting a nilpotent normal subgroup N such that G/N′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$G/N'$$\\end{document} belongs to X\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\\mathfrak {X}}$$\\end{document}. Thus, Hall’s nilpotency criterion just asserts that nilpotent groups form a Hall class. Many other relevant classes of groups have been proved to be Hall classes; for example, Plotkin (Sov Math Dokl 2:471–474, 1961) and Robinson (Math Z 107:225–231, 1968) proved that locally nilpotent groups and hypercentral groups form Hall classes. Note that these generalizations also hold if groups are replaced by other algebraic structures, for example Lie algebras (see Stitzinger in Ill J Math 22:499–505, 1978). The aim of this paper is to develop a general theory of Hall classes of groups, that could later be reasonably extended to Lie algebras. Among other results, we prove that many natural types of generalized nilpotent groups form Hall classes, and we give examples showing in particular that the class of groups having a finite term in the lower central series is not a Hall class, even if we restrict to the universe of linear groups.

Totally geodesic and parallel hypersurfaces of Gödel-type spacetimes

We classify parallel and totally geodesic hypersurfaces of the relevant class of Gödel-type spacetimes, with particular regard to the homogeneous examples.

DeepPyramid+: medical image segmentation using Pyramid View Fusion and Deformable Pyramid Reception

PurposeSemantic segmentation plays a pivotal role in many applications related to medical image and video analysis. However, designing a neural network architecture for medical image and surgical video segmentation is challenging due to the diverse features of relevant classes, including heterogeneity, deformability, transparency, blunt boundaries, and various distortions. We propose a network architecture, DeepPyramid+, which addresses diverse challenges encountered in medical image and surgical video segmentation.MethodsThe proposed DeepPyramid+ incorporates two major modules, namely “Pyramid View Fusion” (PVF) and “Deformable Pyramid Reception” (DPR), to address the outlined challenges. PVF replicates a deduction process within the neural network, aligning with the human visual system, thereby enhancing the representation of relative information at each pixel position. Complementarily, DPR introduces shape- and scale-adaptive feature extraction techniques using dilated deformable convolutions, enhancing accuracy and robustness in handling heterogeneous classes and deformable shapes.ResultsExtensive experiments conducted on diverse datasets, including endometriosis videos, MRI images, OCT scans, and cataract and laparoscopy videos, demonstrate the effectiveness of DeepPyramid+ in handling various challenges such as shape and scale variation, reflection, and blur degradation. DeepPyramid+ demonstrates significant improvements in segmentation performance, achieving up to a 3.65% increase in Dice coefficient for intra-domain segmentation and up to a 17% increase in Dice coefficient for cross-domain segmentation.ConclusionsDeepPyramid+ consistently outperforms state-of-the-art networks across diverse modalities considering different backbone networks, showcasing its versatility. Accordingly, DeepPyramid+ emerges as a robust and effective solution, successfully overcoming the intricate challenges associated with relevant content segmentation in medical images and surgical videos. Its consistent performance and adaptability indicate its potential to enhance precision in computerized medical image and surgical video analysis applications.

Structural classification of Ag and Cu nanocrystals with machine learning.

We use machine learning (ML) to classify the structures of mono-metallic Cu and Ag nanoparticles. Our datasets comprise a broad range of structures - both crystalline and amorphous - derived from parallel-tempering molecular dynamics simulations of nanoparticles in the 100-200 atom size range. We construct nanoparticle features using common neighbor analysis (CNA) signatures, and we utilize principal component analysis to reduce the dimensionality of the CNA feature set. To sort the nanoparticles into structural classes, we employed both K-means clustering and the Gaussian mixture model (GMM). We evaluated the performance of the clustering algorithms through the gap statistic and silhouette score, as well as by analysis of the CNA signatures. For Ag, we found five structural classes, with 14 detailed sub-classes, while for Cu, we found two broad classes (crystalline and amorphous), with the same five classes as for Ag, and 15 detailed sub-classes. Our results demonstrate that these ML methods are effective in identifying and categorizing nanoparticle structures to different levels of complexity, enabling us to classify nanoparticles into distinct and physically relevant structural classes with high accuracy. This capability is important for understanding nanoparticle properties and potential applications.

LINKING HEALTH ATTRIBUTES TO ACUTE-CARE USAGE IN SOUTH FLORIDA VETERANS, A PILOT SURVEY STUDY

Abstract “High-Need, High-Risk” (HNHR) Veteran patients are identified quarterly by the Department of Veterans Affairs (VA) as having the top 5% predicted probability of hospitalization or mortality. We fielded a pilot survey to 2,543 older “HNHR” Veterans receiving care at the Miami VA Medical Center between October 2017 and September 2018, where 634 replied; mean age 70.61, standard deviation, 8.98, racially/ethnically White (403, 61.24%), Black (240, 36.47%), Hispanic (537, 81.61%) and nineteen women (3%). The survey contained 42 self-reported questions covering physical, mental, and social determinants of health, and access/use of healthcare resources. Two measured prior 6-month acute-care utilization outcomes: emergency room visits (ERV) and inpatient hospital admissions (IHA). Logistic regression was used to find health attributes associated with both outcomes, and latent class analysis was used to group Veterans into clinically relevant latent classes, analyzing health attributes endorsed by each. We found that transportation access issues to receiving healthcare were associated with both outcomes (OR = 2.980, p = 0.014), underscoring the role of social determinants of health. Self-perceptions of general health (OR = 0.806, p = 0.04), and a Veteran attending the Miami VA frailty clinic (OR = 0.435, p < 0.001), were associated with ERV and IHA, respectively. We identified four latent classes distinctly grouping patients by endorsing 1) depression; 2) not having a caregiver and being unmarried; 3) functional issues, and 4) being homebound and receiving home healthcare. These latent classes will help better inform clinical recommendations and program development to better meet needs for “HNHR” Veterans.

Exploring Thiazolopyridine AV25R: Unraveling of Biological Activities, Selective Anti-Cancer Properties and In Silico Target and Binding Prediction in Hematological Neoplasms.

Thiazolopyridines are a highly relevant class of small molecules, which have previously shown a wide range of biological activities. Besides their anti-tubercular, anti-microbial and anti-viral activities, they also show anti-cancerogenic properties, and play a role as inhibitors of cancer-related proteins. Herein, the biological effects of the thiazolopyridine AV25R, a novel small molecule with unknown biological effects, were characterized. Screening of a set of lymphoma (SUP-T1, SU-DHL-4) and B- acute leukemia cell lines (RS4;11, SEM) revealed highly selective effects of AV25R. The selective anti-proliferative and metabolism-modulating effects were observed in vitro for the B-ALL cell line RS4;11. Further, we were able to detect severe morphological changes and the induction of apoptosis. Gene expression analysis identified a large number of differentially expressed genes after AV25R exposure and significant differentially regulated cancer-related signaling pathways, such as VEGFA-VEGFR2 signaling and the EGF/EGFR pathway. Structure-based pharmacophore screening approaches using in silico modeling identified potential biological AV25R targets. Our results indicate that AV25R binds with several proteins known to regulate cell proliferation and tumor progression, such as FECH, MAP11, EGFR, TGFBR1 and MDM2. The molecular docking analyses indicates that AV25R has a higher binding affinity compared to many of the experimentally validated small molecule inhibitors of these targets. Thus, here we present in vitro and in silico analyses which characterize, for the first time, the molecular acting mechanism of AV25R, including cellular and molecular biologic effects. Additionally, this predicted the target binding of the molecule, revealing a high affinity to cancer-related proteins and, thus, classified AVR25 for targeted intervention approaches.

Stereoselective β‐Arylation of Enol Ethers by a Pd‐Catalyzed Fujiwara‐Moritani Coupling of α‐Oxyacrylates

AbstractHerein, Pd(II)‐catalyzed dehydrogenative Heck coupling of 2‐oxyacrylates with arenes for the stereocontrolled production of (Z)‐β‐arylated vinyl ethers is reported. This methodology offers a straightforward and step‐economical route to the synthesis of attractive β‐arylated α‐oxy α,β‐unsaturated carboxylates family which emerged as a relevant class of building blocks with different applications.

CD8+ T Cell-Dependent Antitumor Activity InVivo of a Mass Spectrometry-Identified Neoepitope despite Undetectable CD8+ Immunogenicity InVitro.

Identification of neoepitopes that can control tumor growth invivo remains a challenge even 10 y after the first genomics-defined cancer neoepitopes were identified. In this study, we identify a neoepitope, resulting from a mutation in the junction plakoglobin (Jup) gene (chromosome 11), from the mouse colon cancer line MC38-FABF (C57BL/6). This neoepitope, Jup mutant (JupMUT), was detected during mass spectrometry of MHC class I-eluted peptides from the tumor. JupMUT has a predicted binding affinity of 564 nM for the Kb molecule and a higher predicted affinity of 82 nM for Db. However, whereas structural modeling of JupMUT and its unmutated counterpart Jup wild-type indicates that there are little conformational differences between the two epitopes bound to Db, large structural divergences are predicted between the two epitopes bound to Kb. Together with invitro binding data with RMA-S cells, these data suggest that Kb rather than Db is the relevant MHC class I molecule of JupMUT. Immunization of naive C57BL/6 mice with JupMUT elicits CD8-dependent tumor control of a MC38-FABF challenge. Despite the CD8 dependence of JupMUT-mediated tumor control invivo, CD8+ T cells from JupMUT-immunized mice do not produce higher levels of IFN-γ than do naive mice. The structural and immunological characteristics of JupMUT are substantially different from those of many other neoepitopes that have been shown to mediate tumor control.

Supermodular and directionally convex comparison results for general factor models

This paper provides comparison results for general factor models with respect to the supermodular and directionally convex order. These results extend and strengthen previous ordering results from the literature concerning certain classes of mixture models as mixtures of multivariate normals, multivariate elliptic and exchangeable models to general factor models. For the main results, we first strengthen some known orthant ordering results for the multivariate ▪ -product of the specifications, which represents the copula of the factor model, to the stronger notion of the supermodular ordering. The stronger comparison results are based on classical rearrangement results and in particular are rendered possible by some involved constructions of transfers as arising from mass transfer theory. The ordering results for ▪ -products are then extended to factor models with general conditional dependencies. As a consequence of the ordering results, we derive worst case scenarios in relevant classes of factor models allowing, in particular, interesting applications to deriving sharp bounds in financial and insurance risk models. The results and methods of this paper are a further indication of the particular effectiveness of Sklar‘s copula notion.

Feature selection based on self-information combining double-quantitative class weights and three-order approximation accuracies in neighborhood rough sets

Feature selection is related to information processing, and its measurement and algorithm use various intelligent methodologies, such as neighborhood rough sets (NRSs). At present by NRSs, the relative neighborhood self-information (Relative-NSI) introduces an information function for algebraic characterization, and its feature selection algorithm (NSI-FS) has been successfully applied. However, Relative-NSI and NSI-FS ignore the underlying recognition information of decision classes; thus, they have the advancement space. In this study, absolute rates and correlative coefficients of decision classes are double-quantitatively introduced, and three-order approximation accuracies are systematically established to induce a robust self-information measure; thus, corresponding feature selection optimally improves NSI-FS to have generalization abilities. Firstly, three-order approximation accuracies are constructed by using two-time class weights of absolute rates and Spearman's correlation coefficients, and the new measure called ClaWNSI promotes Relative-NSI in terms of class recognition. Then, the three-order approximation accuracies and subsequent ClaWNSI are evolved in matrix forms, and relevant class weights of Spearman's correlation coefficients are realized by two vectors from approximation matrices. Furthermore, ClaWNSI and its feature significance motivate a heuristic selection algorithm (called ClaWNSI-FS). Finally, data experiments on 14 datasets validate ClaWNSI and ClaWNSI-FS; ClaWNSI-FS outperforms NSI-FS and five other contrast algorithms to acquire better classification performance.

Occurrence and evolutionary conservation analysis of α-helical cationic amphiphilic segments in the human proteome.

The existence of encrypted fragments with antimicrobial activity in human proteins has been thoroughly demonstrated in the literature. Recently, algorithms for the large-scale identification of these segments in whole proteomes were developed, and the pervasiveness of this phenomenon was stated. These algorithms typically mine encrypted cationic and amphiphilic segments of proteins, which, when synthesized as individual polypeptide sequences, exert antimicrobial activity by membrane disruption. In the present report, the human reference proteome was submitted to the software kamal for the uncovering of protein segments that correspond to putative intragenic antimicrobial peptides (IAPs). The assessment of the identity of these segments, frequency, functional classes of parent proteins, structural relevance, and evolutionary conservation of amino acid residues within their corresponding proteins was conducted in silico. Additionally, the antimicrobial and anticancer activity of six selected synthetic peptides was evaluated. Our results indicate that cationic and amphiphilic segments can be found in 2% of all human proteins, but are more common in transmembrane and peripheral membrane proteins. These segments are surface-exposed basic patches whose amino acid residues present similar conservation scores to other residues with similar solvent accessibility. Moreover, the antimicrobial and anticancer activity of the synthetic putative IAP sequences was irrespective to whether these are associated to membranes in the cellular setting. Our study discusses these findings in light of the current understanding of encrypted peptide sequences, offering some insights into the relevance of these segments to the organism in the context of their harboring proteins or as separate polypeptide sequences.

Concavity principles for nonautonomous elliptic equations and applications

In the study of concavity properties of positive solutions to nonlinear elliptic partial differential equations the diffusion and the nonlinearity are typically independent of the space variable. In this paper we obtain new results aiming to get almost concavity results for a relevant class of anisotropic semilinear elliptic problems with spatially dependent source and diffusion.

Allied health professional-physician collaboration to enhance guideline-directed medical therapy of heart failure: expanding therapeutic opportunities

Abstract Background Guideline-directed medical therapy (GDMT) is the cornerstone of treatment for heart failure patients with reduced ejection fraction (HFrEF). Optimization of GDMT at a targeted allied professional medication titration clinic represents a key opportunity to provide greater access to care and further refine patient care models, leading to improvements in clinical outcomes. Purpose To evaluate the effectiveness of an allied health practitioner optimization clinic model in HFrEF as indicated by changes in rates of triple and quadruple GDMT therapy (see associated table). Methods Qualified patients were either: proactively screened from clinic lists, having demonstrated a diagnosis of heart failure, reduced ejection fraction, and not receiving triple or quadruple therapy, or were physician-referred after initial optimization during 2021-2022. Participants subsequently underwent medication review and optimization by an allied health professional, involving introduction, up-titration, down-titration, or withdrawal of medications across relevant medication classes as tolerated. Results 52 (78.8% male) patients with a mean age of 70.3 (11.4) years and mean ejection fraction of 30.1 (6.7) percent underwent medication optimization. Along with HFrEF, all patients exhibited cardio-metabolic multi-morbidity (2-3 cardio-metabolic diagnoses), including hypertension (59.6%), dyslipidemia (71.2%), coronary artery disease (61.5%), atrial fibrillation (42.3%), and type 2 diabetes (21.2%). Pre-optimization prescription rates ranged from: 53.8% for ACEi/ARB, 30.8% for ARNI, 80.8% for beta-blockers, 59.6% for MRA, 40.4% for diuretics, and 13.5% for SGLT2i. Medication review and optimization spanned an average of 2.8 (1-7) meetings, and lead to an average of 1.9 (range 0-4) medication changes. Among 49 patients considered for ARNI optimization, there were 43 (87.7%) successful introductions and progressive dose adjustments, including 12 (27.9%) dose adjustments among those already prescribed. Additional changes involved optimizations in: beta-blockers (5 introductions, 5 dose-changes), MRA (8 introductions, 4 dose-changes, 2 withdrawals), diuretics (1 introduction, 7 dose-changes, 2 withdrawals), and SGLT2i (17 introductions). Collaboration between allied health professionals and physicians in optimization demonstrated significant improvements in rates of triple (42.3% vs 63.5%, P= 0.003) and quadruple (5.8% vs 26.9%, P= 0.001) therapy post-optimization. Discussion A Collaborative approach to medication optimization increased the use of GDMT in ambulatory patients with HFrEF. Collaboration presents both clinicians and health systems with greater opportunities to improve healthcare access, therapeutic inertia, and disease management.Pre-post Changes in GDMT Rates

74 Setting Students Up for Success in the Agricultural Industry

Abstract When asked about future employment goals, many agricultural students express interest in being employed in the industry following graduation. Even though a large number of students have this career path in mind, students are often not prepared for industry roles. The agricultural industry favors employees who have good soft skills, such as communication, teamwork, and interpersonal skills, as it is often required that employees collaborate within a team or connect with clients and colleagues. Nevertheless, these soft skills are often not taught or are not emphasized enough during undergraduate or graduate education as they are often difficult to teach in a classroom setting. International students face unique challenges when seeking employment as industry professionals. Some of these challenges, such as language barriers or social and cultural differences, can make important tasks, like networking, more difficult. There are opportunities for university faculty to assist students wanting to seek employment in the agricultural industry, such as by organizing networking events and establishing industry-focused classes or lectures. Networking in a professional setting is a vital skill for students trying to enter the industry; therefore, providing students with opportunities to practice networking and professional social skills among their peers and superiors is critical. Courses or lectures focused on understanding the industry and its challenges would also be highly beneficial to students, particularly those who do not come from agricultural backgrounds and those that are international. Encouraging students to take part in student organizations and extension events, and to apply for teaching assistant positions for relevant production classes may also contribute to student preparation for industry employment. By providing students with an overall understanding of the agricultural industry and opportunities to improve skills that are relevant for industry positions, faculty members may be able to improve student preparedness for these roles and assist in making students more marketable to individual companies.

Detection and localization of caries and hypomineralization on dental photographs with a vision transformer model

Caries and molar-incisor hypomineralization (MIH) are among the most prevalent diseases worldwide and need to be reliably diagnosed. The use of dental photographs and artificial intelligence (AI) methods may potentially contribute to realizing accurate and automated diagnostic visual examinations in the future. Therefore, the present study aimed to develop an AI-based algorithm that can detect, classify and localize caries and MIH. This study included an image set of 18,179 anonymous photographs. Pixelwise image labeling was achieved by trained and calibrated annotators using the Computer Vision Annotation Tool (CVAT). All annotations were made according to standard methods and were independently checked by an experienced dentist. The entire image set was divided into training (N = 16,679), validation (N = 500) and test sets (N = 1000). The AI-based algorithm was trained and finetuned over 250 epochs by using image augmentation and adapting a vision transformer network (SegFormer-B5). Statistics included the determination of the intersection over union (IoU), average precision (AP) and accuracy (ACC). The overall diagnostic performance in terms of IoU, AP and ACC were 0.959, 0.977 and 0.978 for the finetuned model, respectively. The corresponding data for the most relevant caries classes of non-cavitations (0.630, 0.813 and 0.990) and dentin cavities (0.692, 0.830, and 0.997) were found to be high. MIH-related demarcated opacity (0.672, 0.827, and 0.993) and atypical restoration (0.829, 0.902, and 0.999) showed similar results. Here, we report that the model achieves excellent precision for pixelwise detection and localization of caries and MIH. Nevertheless, the model needs to be further improved and externally validated.