Browse Articles

Breaking barriers in ICD classification with a robust graph neural network for hierarchical coding

Abstract The accurate classification of International Classification of Diseases (ICD) codes is a complex and critical multi-label task in clinical documentation, involving the assignment of diagnostic codes to medical discharge summaries. Existing automated methods face challenges due to the sparsity and nuanced nature of medical text, while traditional backpropagation-based models often lack flexibility and robustness. To address these issues, we propose Labeled Graph Generation with Node Representation Grasp (LGG-NRGrasp), an advanced adversarial learning framework that models ICD coding as a labeled graph generation problem. By leveraging a hierarchical structure to refine feature learning, our approach addresses the issue of over-smoothing in deep graph neural networks. A key innovation of LGG-NRGrasp is the integration of adversarial reinforcement learning and domain adaptation techniques, which enhance its ability to generalize across heterogeneous datasets. Extensive evaluations on benchmark datasets indicate that LGG-NRGrasp markedly surpasses leading models, exhibiting enhanced performance and dependability in automated ICD coding.

Transplanting drought-Protective Bacteria to enhance clove basil’s (Ocimum gratissimum) drought tolerance

Microstructural injury to the optic nerve with vigabatrin treatment in West syndrome: A DTI study

Abstract To evaluate optic nerve injury associated with vigabatrin treatment in children with West syndrome using diffusion tensor imaging. Thirty-five children with West syndrome (aged 9 days–22 months) were retrospectively analyzed and grouped as follows: (1) vigabatrin with symmetrical thalamic abnormalities, (2) vigabatrin without thalamic abnormalities, and (3) controls on other anti-seizure medications. Fractional anisotropy and apparent diffusion coefficient values of the optic nerves were assessed. ROC curves were used to determine fractional anisotropy thresholds for optic nerve injury. fractional anisotropy values in group 1 were significantly lower than those in the control group (P < 0.05), while apparent diffusion coefficient values showed no significant differences. fractional anisotropy values increased significantly after vigabatrin discontinuation (P < 0.05). ROC analysis yielded an fractional anisotropy cut-off value of 304 with 63.6% sensitivity and 100% specificity. fractional anisotropy values are a sensitive imaging biomarker for detecting vigabatrin-related optic nerve injury in West syndrome, particularly when thalamic abnormalities are present. These changes appear reversible after stopping vigabatrin.

Association between various insulin resistance surrogates and gallstone disease based on national health and nutrition examination survey

Abstract Previous studies have reported potential associations between insulin resistance (IR) and the occurrence of gallstone disease (GSD). Recently, some novel indexes have been developed and applied for the assessment of IR. However, the association of these IR surrogates with GSD is unclear. The present study was designed to investigate the correlation between these novel IR surrogates and the risk of developing GSD and to explore the index with the best predictive value. We conducted a cross-sectional analysis using data from the National Health and Nutrition Examination Survey 2017-March 2020. Participants who self-reported a history of GSD and other necessary information were available and were enrolled. Ten IR surrogates were created based on previous literature. Univariate and multivariate logistic regression models were applied to estimate the effect of higher IR surrogates on the risk of GSD and restricted cubic splines (RCS) were used to show the dose-response relationships. Receiver operating characteristics (ROC) curves were depicted and the areas under the curves (AUC) were calculated to show the diagnostic value of ten indexes, respectively. Finally, the best index was selected and subgroup analysis was performed to further evaluate the risk level in different populations. Among the 2811 participants, 305 (11%) were diagnosed with GSD. According to results from univariate and multivariate logistic regression models, elevated levels of IR surrogates were significantly associated with an increased risk of GSD, including triglyceride glucose-body mass index (TyG-BMI), triglyceride glucose-waist circumference (TyG-WC), triglyceride glucose-waist to height ratio (TyG-WHtR), homeostasis model assessment-insulin resistance (HOMA-IR), metabolic score for insulin resistance (METS-IR), Chinese visceral adiposity index (CVAI), and lipid accumulation product (LAP). The RCS also confirmed the trend of positive correlation between the above indexes and GSD (p for overall < 0.0001). The ROC curves showed that TyG-WHtR demonstrated the strongest predictive power among these indexes, with an area under the curve of 0.6796 (95% CI 0.6513–0.7090). Subgroup analysis of the correlation between TyG-WHtR and GSD showed that the correlation was more pronounced among females, the never-married group, the nondiabetic group, and the group that did not use hypoglycemic or lipid-lowering medication. This study identified several novel IR surrogates that had significant positive correlations with the development of GSD. Among these indexes, TyG-WHtR was the strongest predictor of GSD, and the correlation was more pronounced in female groups and non-diabetic groups, which may provide significant value for screening, disease prediction, and early intervention in high-risk groups in clinical practice.

Dynamic monitoring of vegetation phenology on the Qinghai-Tibetan plateau from 2001 to 2020 via the MSAVI and EVI

Controlled biodegradability of polyhydroxybutyrate via surface coating with cellulose triacetate

Impact of Brownian motion on the optical soliton solutions for the three component nonlinear Schrödinger equation

In vivo ultrasonographic evidence that tensed calcaneofibular ligament consistently lifts the peroneal tendons

In vitro shear bond strength evaluation of resin cements between zirconia and titanium

Abstract This study aimed to evaluate the Shear Bond Strength (SBS) and durability of different resin cements used for cement-retained restorations between zirconia crowns and titanium abutments. A total of 36 zirconia cubes and 324 titanium cylinders were employed in this study. Accordingly, 6 groups were assigned according to the used resin cement: three self-etching types (RU, AI, GMP), two self-adhesive types (IS, GM) and one unfilled type (SB). Each group was further divided into 3 sub-group based on storage condition (n = 10): 24-hour water storage at 37 °C, 5,000 thermocycles, and 10,000 thermocycles respectively. SBS testing employed a universal testing machine to quantify the force required for material fracture at the interface, followed by failure mode analysis. Two-way ANOVA showed significant influences of storage conditions and bonding agents on SBS values (p < 0.05), except for IS and GMP (p > 0.05). The SBS values of AI and RU showed a downward trend, while SB and GM showed an upward trend after 10,000 thermocycles. The SBS values of GM and GMP were significantly different in the 24-hour water storage but not after thermocycles. Within the limitations of the current study and based on the findings, the following can be concluded: (1) SBS was significantly influenced by both bonding agent type and storage condition; (2) Bonding systems containing HEMA and 10-MDP exhibited high initial strength but declined significantly after thermocycling; (3) IS, GM, and GMP demonstrated stable bond durability.

Supporting resilience-based coral reef management using broadscale threshold approaches

Abstract Resilience-Based Management of coral reefs aims to maintain ecosystem function and maximise resilience. This requires identification of resilience indicators and clear ecological reference thresholds for reef managers to maintain or aim for. In the absence of local thresholds, managers can assess reef condition by comparing locally collected indicator data to broadscale thresholds, which account for spatial and temporal variability. This study assesses reef condition at Aldabra Atoll, a remote MPA in the western Indian Ocean, relative to broadscale thresholds for structural complexity, fish biomass, herbivore biomass, juvenile coral density, and trophic-level fish biomass. Results were synthesized into a resilience index, and sites were classified into ‘management strategies’ using a published reef management framework. Resilience scores were then compared to observe coral cover changes following the 2016 bleaching event, tracking recovery through to 2022. Findings showed that seven of the eight assessed seaward reefs at Aldabra displayed the resilience expected of a remote, well-managed marine reserve. The research station and associated human activity appeared to have minimal negative impacts on reef resilience. We recommend expanding the range of broadscale threshold categories and integrating site-specific factors to improve future assessments and management decisions.

Evaluating a novel approach to placenta accreta spectrum management: the modified Triple-P technique with cystoinflation (a randomized controlled trial)

Design, synthesis and evaluation of benzodioxole and bromofuran tethered 1,2,4-triazole hybrids as potential anti breast cancer agents with computational insights

Abstract 1,2,4-Triazole derivatives are the focus of extensive research in medicinal chemistry because of their diverse biological activities, particularly their potential as anticancer agents. In this study, we designed and synthesized six compounds featuring benzo[d][1,3]dioxole and 5-bromofuran tethered to 1,2,4-triazole hybrids. This was achieved through an efficient four-step protocol with reproducible results. The characterization of the 1,2,4-triazoles was conducted via various analytical techniques, including FTIR, 1H NMR, 13C NMR, and mass spectrometry. The in vitro anti-breast cancer activities of synthesized 1,2,4-triazols were evaluated against the MCF-7 cell line using the MTT assay. Among all the synthesized compounds, compound 12b demonstrated an excellent IC50 value of 3.54 ± 0.265 µg/mL. Additionally, in silico studies, such as molecular docking to predict the orientation of the compounds, molecular dynamics simulations to evaluate binding stability with the target protein, drug-likeness studies to evaluate Lipinski’s rule of five and Jorgensen’s rule of three, DFT analysis to determine the energy gap of the frontier molecular orbitals (FMOs) and the molecular electrostatic potential (MEP) for identifying sites of nucleophilic and electrophilic attacks, were also conducted.

Research on the types of environmental characteristics of traditional villages in Guizhou province, China

Isolation effects and soil properties drive genetic differentiation in Ormosia microphylla populations

Early mobilisation to enhance recovery following cardiac valvular surgery in atrial fibrillation patients: a randomised controlled trial

Exploring psychotherapists’ experiences in delivering effective treatment for major depression in Saudi Arabia: a qualitative study

Factors affecting health-related quality of life in ICU survivors

Hydroxysafflor Yellow A modulation of metabolite networks and inhibition of JAK2/STAT1 pathway in sepsis

Synchrotron-based X-ray 3D phase contrast imaging and analysis of transmural myocardial tissue from heart failure patients

Abstract Synchrotron-based X-ray phase contrast imaging (X-PCI) is a non-destructive imaging modality that can provide high resolution three-dimensional (3D) visualisation of transmural myocardial tissue, collagen matrix reconstruction, and quantification of myocyte aggregate orientation (‘myomapping’). We aimed to use X-PCI to analyse microstructural features in transmural myocardial samples from patients with advanced heart failure. Six patients were included: two receiving a left ventricular assist device (LVAD) for ischaemic (ICM) and dilated cardiomyopathy (DCM), and four undergoing heart transplantation (HTx), two for the ICM, one for DCM and one for toxic cardiomyopathy. Samples were obtained by left ventricular (LV) apical coring (LVAD group) or from the LV free wall of the explanted hearts (HTx group) and imaged by X-PCI using a multi-scale setup (maximal resolution at 0.65 µm pixel size). The 3D image datasets were analysed via two-dimensional orthogonal cuts in different layers. Visualisation and quantification of the myocyte aggregates orientation showed a disruption in epicardial-to-endocardial transition in DCM, whereas the collagen matrix reconstruction identified characteristic fibrosis patterns amongst different HF aetiologies. In conclusion, X-PCI is a 3D imaging method that can extend the amount of information available from ex-vivo tissue analysis and, as an addition to multimodal imaging protocols, potentially improve disease phenotyping.

In Silico tool for predicting, designing and scanning IL-2 inducing peptides

Abstract Interleukin-2 (IL-2) based immunotherapy has been approved for treating certain types of cancer, as IL-2 plays a crucial role in regulating the immune system. In this study, we developed a method for predicting IL-2-inducing peptides. Our method was trained, tested, and validated on a main dataset containing 6,574 experimentally validated Major histocompatibility complex (MHC) binders, including 3,429 IL-2-inducing and 3,145 non-inducing peptides. A primary analysis of IL-2 inducing and non-inducing peptides revealed that certain residues, such as alanine and leucine, are more abundant in IL-2-inducing peptides. Initially, we developed alignment-based methods, which demonstrated high precision but limited coverage. Subsequently, we developed artificial intelligence-based models, including machine learning (ML), deep learning (DL), and large language models (LLM), to predict IL-2-inducing peptides. Our Extra Tree-based model, developed using dipeptide composition and peptide length, achieved a maximum AUC of 0.82. Finally, we constructed ensemble models that combined artificial intelligence and alignment-based methods. Our best ensemble model, which integrates the Extra Tree-based model with MERCI, achieved the highest AUC of 0.84 and an MCC of 0.51 on the main dataset. One limitation of the main dataset is that both IL-2-inducing and non-inducing peptides are MHC binders. To address this limitation, we created two additional datasets: Alternate Dataset 1, consisting of 3,429 IL-2-inducing peptides and 3,429 non-inducing peptides (MHC non-binders), and Alternate Dataset 2, consisting of 3,429 IL-2-inducing peptides and 3,439 non-inducing peptides (MHC binders + MHC non-binders). Our best ensemble model achieved AUCs of 0.9 and 0.8 with MCCs of 0.61 and 0.44 on Alternate Datasets 1 and 2, respectively. To assist the scientific community, we have integrated the best models from this study into a standalone software and web server, IL2pred, which enables users to predict, scan, and design IL-2-inducing peptides ( https://webs.iiitd.edu.in/raghava/il2pred/ ).