Emergent Research Trends on the Structural Relaxation Dynamics of Molecular Clusters: From Structure-Property Relationship to New Function Prediction.

ConspectusMolecular clusters (MCs) are monodispersed, precisely defined ensembles of atom collections featured with shape-persistent architectures that can deliver certain functions independently. Their molecular compositions and surface functionalities can be tailored feasibly in a predefined manner, and they can be applied as basic structural units to be engineered into materials with desirable hierarchical structures and enriched functions. The chemical systems also offer great opportunities for the design and fabrication of soft structural materials without the chain topologies of polymers. The bulks of MC assemblies demonstrate viscoelasticity that is used to be considered as the unique feature of polymers, while the MC systems are distinct from polymers since their elasticities are resilient even at temperatures 100 K above their glass transition temperatures. The understanding of their anomalous viscoelasticity and the extended studies of general structure-property relationships are desired for the development of new chemical systems for emergent functions and the possibilities to resolve the intrinsic trade-offs of traditional materials.Meanwhile, general macroscopic functions or properties of materials are related to the transportation of mass, momentum, and/or energy, and they are basically realized or directed by the motions of structural units at different length scales. Structural relaxation dynamics research is critical in quantifying motions ranging from fast bond deformation, bond break/formation, and diffusion of ions and particles to the cooperative motions of structure units. Due to the advancement of measurement technology for relaxation dynamics (e.g., quasi-elastic scattering and broadband dielectric spectroscopy), the structural relaxation dynamics of MC materials have been probed for the first time, and their multiple relaxation modes across several temporal scales were systematically studied to bridge the correlation between molecular structures and macroscopic functions. The fingerprint information from dynamics studies, e.g., the temperature dependence of relaxation time and certain property, e.g., ion conductivity, was proposed to quantify the structure-property relationship, and the microscopic mechanism on the mechanical properties, ion conduction, and gas absorption and separation of MC materials can be fully understood.In this Account, to elucidate the uniqueness of MC materials, especially in comparison with polymers, four topics are mainly summarized: structural features, relaxation dynamics characterization techniques, relaxation dynamics characteristics, and quantified understanding of the structure-property relationship. The capability for new function prediction from relaxation dynamics studies is also introduced, and the typical example in impact resistant materials is provided. The Account aims to prove the significance of relaxation dynamics characterization for material innovation, while it also confirms the potential of MCs for functional material fabrications.

Unique Viscoelasticity and Hierarchical Relaxation Dynamics of Molecular Granular Materials.

Resulting from the dense packing of subnanometer molecular clusters, molecular granular materials (MGMs) are shown to maintain high elasticity far above their apparent glass transition temperature (Tg*). However, our microscopic understanding of their structure-property relationship is still poor. Herein, 1 nm polyhedral oligomeric silsesquioxanes (POSSs) are appended to a backbone chain in a brush configuration with different flexible linker chains. Assemblies of these brush polymers exhibit hierarchical relaxation dynamics with the glass transition arising from the cooperative dynamics of packed POSSs. The interaction among the assemblies can be strengthened by increasing the rigidity of linkers with the MGM relaxation modes changing from colloid- to polymer chain-like behavior, rendering their tunable viscoelasticity. This finally contributes to the decoupling of mechanical and thermal properties by showing elasticity dominant mechanical properties at a temperature 150 K above the Tg*.

Supramolecular Complexation Reinforced Polymer Frustrated Packing: Controllable Dual Porosity for Improved Permselectivity of Coordination Nanocage Mixed Matrix Membranes.

The developments of mixed matrix membranes (MMMs) are severely hindered by the complex inter-phase interaction and the resulting poor utilization of inorganics' microporosity. Herein, a dual porosity framework is constructed in MMMs to enhance the accessibility of inorganics' microporosity to external gas molecules for the effective application of microporosity for gas separation. Nanocomposite organogels are first prepared from the supramolecular complexation of rigid polymers and 2 nm microporous coordination nanocages (CNCs). The network structures can be maintained with microporous features after solvent removal originated from the rigid nature of polymers, and the strong coordination and hydrogen bond between the two components. Moreover, the strong supramolecular attraction reinforces the frustrated packing of the rigid polymers on CNC surface, leading to polymer networks' extrinsic pores and the interconnection of CNCs' micro-cavities for the fast gas transportation. The gas permeabilities of the MMMs are 869 times for H2 and 1099 times for CO2 higher than those of pure polymers. The open metal sites from nanocage also contribute to the enhanced gas selectivity and the overall performance surpasses 2008 H2/CO2 Robeson upper bound. The supramolecular complexation reinforced packing frustration strategy offers a simple and practical solution to achieve improved gas permselectivity in MMMs.

Causal Associations of Genetically Determined Tinnitus With Neuroimaging Traits: Evidence From a Mendelian Randomization Study.

OBJECTIVES: Potential reverse causality and unmeasured confounding factors are common biases in most neuroimaging studies on tinnitus and central correlates. The causal association of tinnitus with neuroimaging features also remains unclear. This study aimed to investigate the causal relationship of tinnitus with neuroplastic alterations using Mendelian randomization. DESIGN: Summary-level data from a genome-wide association study of tinnitus were derived from UK Biobank (n = 117,882). The genome-wide association study summary statistics for 4 global-brain tissue and 14 sub-brain gray matter volumetric traits were also obtained (n = up to 33,224). A bidirectional Mendelian randomization analysis was conducted to explore the causal relationship between tinnitus and neuroanatomical features at global-brain and sub-brain levels. RESULTS: Genetic susceptibility to tinnitus was causally associated with increased white matter volume (odds ratio [OR] = 2.361, 95% confidence interval [CI], 1.033 to 5.393) and total brain volume (OR = 2.391, 95% CI, 1.047 to 5.463) but inversely associated with cerebrospinal fluid volume (OR = 0.362, 95% CI, 0.158 to 0.826). A smaller gray matter volume in the left Heschl's gyrus and right insular cortex and larger gray matter volume in the posterior division of the left parahippocampal gyrus may lead to an increased risk for tinnitus (OR = 0.978, 95% CI, 0.961 to 0.996; OR = 0.987, 95% CI, 0.976 to 0.998; and OR = 1.015, 95% CI, 1.001 to 1.028, respectively). CONCLUSIONS: Genetic susceptibility to tinnitus was causally associated with increased white matter volume and total brain volume. Volume alteration in several cortical regions may indicate a higher tinnitus risk, and further research is recommended for causality inference at the level of sub-brain regions. Our findings provide genetic evidence for elucidating the underlying pathophysiological mechanisms of tinnitus-related neuroanatomical abnormalities.

Supramolecular Assembly and Microscopic Dynamics Modulation of Nanoscale Inorganic Cryptand and Polymer Complex for Versatile Design of Flexible Single-Ion Conductors.

The popular design of solid-state electrolytes (SSEs) from the chain relaxation of polymers faces the trade-offs among ion conductivity, stability, and processability. Herein, 2 nm inorganic cryptand molecules with the capability to carry different types of cations, including Ag+, Na+, K+, and Ca2+, are complexed with cationic polymers via ionic interaction, respectively, and the hybrid materials further phase separate into lamellar or hexagonal columnar structures. The successful establishment of ordered structures with ion channels from the packing of inorganic cryptands confers SSEs' excellent ionic conductivity to versatile types of cations. Meanwhile, suggested from the combination of broad dielectric spectroscopy, rheology, and thermal analysis, the fast chain relaxation can activate the dynamics of inorganic cryptand molecules and facilitate the ion hopping process in ion channels. The supramolecular interaction in the complex enables the highly flexible physical appearance for defect-free contact with electrodes as well as cost-effective processability and recyclability.

Dobutamine-induced alternations in cerebral blood flow of healthy adults: a 3D pseudocontinuous arterial spin labeling study.

BACKGROUND: It is unclear whether dobutamine, commonly used clinically in echocardiography and short-term congestive heart failure treatment for promoting increased myocardial contractility, affects brain microcirculatory behavior. Cerebral microcirculation plays an important role in ensuring adequate oxygen transport. Therefore, we investigated the effects of dobutamine on cerebral hemodynamics. METHODS: Forty-eight healthy volunteers without cardiovascular or cerebrovascular disease underwent MRI to obtain cerebral blood flow (CBF) maps using 3D pseudocontinuous arterial spin labeling before and during the dobutamine stress test. Additionally, cerebrovascular morphology was obtained based on 3D-time-off-light (3D-TOF) magnetic resonance angiography (MRA). Electrocardiogram, heart rate (HR), respiration rate (RR), blood pressure, and blood oxygen were simultaneously recorded before and during dobutamine injection and during recovery (not during MRI). The anatomic features of the circle of Willis and the basilar artery (BA) diameter were assessed on MRA images by two radiologists with extensive neuroimaging experience. Binary logistic regression was used to test for the independent determinants of CBF changes. RESULTS: HR, RR, systolic (SBP), and diastolic blood pressure (DBP) significantly increased after dobutamine infusion. Blood oxygen levels remained similar. Compared to the CBF in the resting state, the CBF values exhibited significantly lower CBF levels in both grey matter and white matter. Furthermore, compared with the CBF in the resting state, that in the stress state was decreased in the anterior circulation, mainly in the frontal lobe (voxel level P < 0.001, pixel level P < 0.05). Logistic regression showed that body mass index (BMI; odds ratio [OR] 5.80, 95% confidence interval [CI] 1.60-21.01, P = 0.008], resting SBP (OR 0.64, 95% CI 0.45-0.92, P = 0.014), and BA diameter (OR 11.04, 95% CI 1.05-116.53, P = 0.046) were significantly associated with frontal lobe CBF changes. CONCLUSIONS: Dobutamine-induced stress significantly decreased CBF in the frontal lobe anterior circulation. Individuals with a high BMI and low SBP during the dobutamine stress test are more likely to have a stress-induced CBF decrease. Thus, attention should be paid to blood pressure, BMI, and cerebrovascular morphology of patients undergoing dobutamine stress echocardiography or those receiving intensive care or anesthesia.

Integrating domain knowledge for biomedical text analysis into deep learning: A survey.

The past decade has witnessed an explosion of textual information in the biomedical field. Biomedical texts provide a basis for healthcare delivery, knowledge discovery, and decision-making. Over the same period, deep learning has achieved remarkable performance in biomedical natural language processing, however, its development has been limited by well-annotated datasets and interpretability. To solve this, researchers have considered combining domain knowledge (such as biomedical knowledge graph) with biomedical data, which has become a promising means of introducing more information into biomedical datasets and following evidence-based medicine. This paper comprehensively reviews more than 150 recent literature studies on incorporating domain knowledge into deep learning models to facilitate typical biomedical text analysis tasks, including information extraction, text classification, and text generation. We eventually discuss various challenges and future directions.

Dynamics and Proton Conduction of Heterogeneously Confined Imidazole in Porous Coordination Polymers.

The nanoconfinement of proton carrier molecules may contribute to the lowing of their proton dissociation energy. However, the free proton transportation does not occur as easily as in liquid due to the restricted molecular motion from surface attraction. To resolve the puzzle, herein, imidazole is confined in the channels of porous coordination polymers with tunable geometries, and their electric/structural relaxations are quantified. Imidazole confined in a square-shape channels exhibits dynamics heterogeneity of core-shell-cylinder model. The core and shell layer possess faster and slower structural dynamics, respectively, when compared to the bulk imidazole. The dimensions and geometry of the nanochannels play an important role in both the shielding of the blocking effect from attractive surfaces and the frustration filling of the internal proton carrier molecules, ultimately contributing to the fast dynamics and enhanced proton conductivity.

Automatic text classification of actionable radiology reports of tinnitus patients using bidirectional encoder representations from transformer (BERT) and in-domain pre-training (IDPT).

BACKGROUND: Given the increasing number of people suffering from tinnitus, the accurate categorization of patients with actionable reports is attractive in assisting clinical decision making. However, this process requires experienced physicians and significant human labor. Natural language processing (NLP) has shown great potential in big data analytics of medical texts; yet, its application to domain-specific analysis of radiology reports is limited. OBJECTIVE: The aim of this study is to propose a novel approach in classifying actionable radiology reports of tinnitus patients using bidirectional encoder representations from transformer BERT-based models and evaluate the benefits of in domain pre-training (IDPT) along with a sequence adaptation strategy. METHODS: A total of 5864 temporal bone computed tomography(CT) reports are labeled by two experienced radiologists as follows: (1) normal findings without notable lesions; (2) notable lesions but uncorrelated to tinnitus; and (3) at least one lesion considered as potential cause of tinnitus. We then constructed a framework consisting of deep learning (DL) neural networks and self-supervised BERT models. A tinnitus domain-specific corpus is used to pre-train the BERT model to further improve its embedding weights. In addition, we conducted an experiment to evaluate multiple groups of max sequence length settings in BERT to reduce the excessive quantity of calculations. After a comprehensive comparison of all metrics, we determined the most promising approach through the performance comparison of F1-scores and AUC values. RESULTS: In the first experiment, the BERT finetune model achieved a more promising result (AUC-0.868, F1-0.760) compared with that of the Word2Vec-based models(AUC-0.767, F1-0.733) on validation data. In the second experiment, the BERT in-domain pre-training model (AUC-0.948, F1-0.841) performed significantly better than the BERT based model(AUC-0.868, F1-0.760). Additionally, in the variants of BERT fine-tuning models, Mengzi achieved the highest AUC of 0.878 (F1-0.764). Finally, we found that the BERT max-sequence-length of 128 tokens achieved an AUC of 0.866 (F1-0.736), which is almost equal to the BERT max-sequence-length of 512 tokens (AUC-0.868,F1-0.760). CONCLUSION: In conclusion, we developed a reliable BERT-based framework for tinnitus diagnosis from Chinese radiology reports, along with a sequence adaptation strategy to reduce computational resources while maintaining accuracy. The findings could provide a reference for NLP development in Chinese radiology reports.

Modulating Polymer Dynamics via Supramolecular Interaction with Ultrasmall Nanocages for Recyclable Gas Separation Membranes with Intrinsic Microporosity.

The engineering of mixed-matrix membranes is severely hindered by the trade-off between mechanical performance and effective utilization of inorganic fillers' microporosity. Herein, we report a feasible approach for optimal gas separation membranes through the fabrication of coordination nanocages with poly(4-vinylpyridine) (P4VP) via strong supramolecular interactions, enabling the homogeneous dispersion of nanocages in polymer matrixes with long-term structural stability. Meanwhile, suggested from dynamics studies, the strong attraction between P4VP and nanocages slows down polymer dynamics and rigidifies the polymer chains, leading to frustrated packing and lowered densities of the polymer matrix. This effect allows the micropores of nanocages to be accessible to external gas molecules, contributing to the intrinsic microporosity of the nanocomposites and the simultaneous enhancement of permselectivities. The facile strategy for supramolecular synthesis and polymer dynamics attenuation paves avenues to rational design of functional hybrid membranes for gas separation applications.

Unique Dynamics of Water Confined on the Surface of Metal Oxide Clusters.

Water confined on metal oxide surface plays significant roles in heterogeneous catalysis. Heteropolyacid, a 1.2 nm-metal oxide cluster with well-defined structure, is applied as a model to understand the dynamics of water on its surface. The surface water strongly associates with heteropolyacid cluster and form the so-called 'pseudoliquid phase' where catalytic reactions are conducted. Broadband dielectric spectroscopy and differential scanning calorimetry have been applied to probe the dynamics of water in this pseudoliquid phase. A supercooling phase transition of water below its normal melting temperature and a dipolar glassy relaxation behaviour due to the hindered dynamics of water have been observed. The rich dynamic behavior on the surface of such well-defined metal clusters provide new perspectives to understand the properties of surface water and their relation to catalytic performance of heteropolyacid.

Network structure of swollen iodine-doped poly(vinyl alcohol) amorphous domain as characterized by low field NMR.

The network structure in the amorphous domain of swollen iodine-doped poly(vinyl alcohol) (PVA) was systematically investigated by low-field (LF) NMR techniques to reveal the PVA-iodine complex formation mechanism. Three PVA-iodine complexes were obtained under different iodine concentrations (ciodine) of KI/I2 solution: (i) ciodine < 0.1 M: PVA-I3-/I5- complex only exists in the non-crystalline region, (ii) 0.1 M < ciodine < 1 M: formation of PVA-I3- complex I, and (iii) ciodine > 1 M: formation of PVA-I3- complex II. It was found that there is no intermediate-magnitude chain motion of PVA under dyeing conditions to induce the substance exchange, as evidenced by the unchanged second moment M2 (∼1.2 × 104 m s-2) at elevated temperature (<380 K). The introduction of iodine ions can affect the chain mobility of the interphase and mobile regions. With increasing ciodine, the chain dynamics become more restricted, as detected by the faster decay of the T2 relaxometry results, which further accelerates the complexation process. The residual dipolar coupling strength, Dres, obtained by the more quantitative double-quantum (DQ) NMR, increases abruptly at ciodine > 1 M. This suggests more constraints form in the amorphous network for the PVA-I3- complex II system. The constant defects fraction further reveals that the complexation prefers to happen along the tie chains. These results supply a possible formation pathway for the PVA-iodine complexes.

Unexpected Elasticity in Assemblies of Glassy Supra-Nanoparticle Clusters.

Granular materials, composed of densely packed particles, are known to possess unique mechanical properties that are highly dependent on the surface structure of the particles. A microscopic understanding of the structure-property relationship in these systems remains unclear. Here, supra-nanoparticle clusters (SNPCs) with precise structures are developed as model systems to elucidate the unexpected elastic behaviors. SNPCs are prepared by coordination-driven assembly of polyhedral oligomeric silsesquioxane (POSS) with metal-organic polyhedron (MOP). Due to the disparity in sizes, the POSS-MOP assemblies, like their classic nanoparticles counterparts, ordering is suppressed, and the POSS-MOP mixtures will vitrify or jam as a function of decreasing temperature. An unexpected elasticity is observed for the SNPC assemblies with a high modulus that is maintained at temperatures far beyond the glass transition temperature. From studies on the dynamics of the hierarchical structures of SNPCs and molecular dynamic simulation, the elasticity has its origins in the interpenetration of POSS-ended arms. The physical molecular interpenetration and inter-locking phenomenon favors the convenient solution or pressing processing of the novel cluster-based elastomers.

Shaoyao decoction alleviates DSS-induced colitis by inhibiting IL-17a-mediated polarization of M1 macrophages.

ETHNOPHARMACOLOGICAL RELEVANCE: The efficacy of Shaoyao decoction (SYD), a traditional Chinese medicine prescription, in alleviating colonic mucosal inflammation in ulcerative colitis (UC) has been established. However, the specific mechanism underlying the therapeutic effects of SYD on UC is not clear. AIM OF STUDY: In this study, we demonstrated the therapeutic effect of SYD on the polarization of M1 macrophages and elucidated the underlying mechanism. MATERIALS AND METHODS: UC mice were induced with 3% DSS for one week and subsequently treated with SYD for another week. The composition of SYD was determined by HPLC. To assess the therapeutic efficacy of SYD, key parameters, including body weight changes, DAI scores, colon length, and histological alterations in colonic tissues, were monitored. ELISA was performed to quantify inflammatory cytokines, while Western blotting and immunofluorescence analyses were performed to quantify tight junction protein expression and M1 macrophage infiltration, respectively. Functional assessments focused on lysosomal activity and glucose oxidation in primary macrophages and RAW 264.7 cells exposed to LPS, IL-17a, and SYD using dedicated kits. To elucidate the mechanisms underlying the action of SYD, the data derived from TMT-based proteomics were analyzed to screen and predict its potential targets and regulated pathways. RESULTS: After treatment for seven days, SYD significantly mitigated colitis symptoms in mice, as determined by a decrease in body weight loss, an increase in colon length, and a decrease in disease activity index (DAI) score. The results of histopathological analysis showed substantial improvements in the integrity of colonic tissue. Additionally, SYD treatment significantly decreased the levels of proinflammatory cytokines, including IL-17a, IL-6, IL-1ß, and TNF-α. This effect was accompanied by a reduction in the infiltration of CD86+ macrophages and restoration of occludin and ZO-1 levels, thus improving colonic mucosal permeability. SYD treatment also reversed the upregulation of cathepsin (CTS) E, CTSS, lysosomal-associated membrane protein (LAMP)-1, and LAMP-2 expression observed in CD86+ macrophages and RAW 264.7 cells treated with IL-17a. These changes were accompanied by an increase in lysosomal acidification and the enzymatic activities of CTSE and CTSS, suggesting that SYD can restore lysosomal function. SYD also corrected the metabolic alterations in M1 macrophages, characterized by an increase in the extracellular acidification rate (ECAR) and a decrease in oxygen consumption rate (OCR). This was confirmed by a decrease in the ADP/ATP ratio, downregulation of pyruvate dehydrogenase kinase 4 (PDK4) and lactate dehydrogenase (LDH) expression, and a decrease in the concentrations of intracellular pyruvate and lactate. These findings showed that SYD promotes glucose oxidation in macrophages. The data derived from TMT-based proteomics showed that the PPAR pathway was a key target in regulating the polarization of M1 macrophages. SYD was also found to regulate the expression of the PPAR-α, PPAR-γ, and PPAR-δ proteins. CONCLUSION: SYD inhibited the polarization of M1 macrophages induced by IL-17a. This effect occurred due to the restoration of lysosomal activity and glucose oxidation via activation of the PPAR/NF-κB pathway. Our findings provided novel insights into the mechanisms underlying the therapeutic effects of SYD in UC.

Causal relationship between multiparameter brain MRI phenotypes and age: evidence from Mendelian randomization.

To explore the causal relationship between age and brain health (cortical atrophy, white matter integrity, white matter hyperintensities and cerebral microbleeds in various brain regions) related multiparameter imaging features using two-sample Mendelian randomization. Age was determined as chronological age of the subject. Cortical volume, white matter micro-integrity, white matter hyperintensity volume and cerebral microbleeds of each brain region were included as phenotypes for brain health. Age and imaging of brain health related genetic data were analysed to determine the causal relationship using inverse-variance weighted model, validated by heterogeneity and horizontal pleiotropy variables. Age is causally related to increased volumes of white matter hyperintensities (ß = 0.151). For white matter micro-integrity, fibres of the inferior cerebellar peduncle (axial diffusivity ß = -0.128, orientation dispersion index ß = 0.173), cerebral peduncle (axial diffusivity ß = -0.136), superior fronto-occipital fasciculus (isotropic volume fraction ß = 0.163) and fibres within the limbic system were causally deteriorated. We also detected decreased cortical thickness of multiple frontal and temporal regions (P < 0.05). Microbleeds were not related with aging (P > 0.05). Aging is a threat of brain health, leading to cortical atrophy mainly in the frontal lobes, as well as the white matter degeneration especially abnormal hyperintensity and deteriorated white matter integrity around the hippocampus.

Changes in iron load in specific brain areas lead to neurodegenerative diseases of the central nervous system.

The causes of neurodegenerative diseases remain largely elusive, increasing their personal and societal impacts. To reveal the causal effects of iron load on Parkinson's disease (PD), Alzheimer's disease (AD), amyotrophic lateral sclerosis and multiple sclerosis, we used Mendelian randomisation and brain imaging data from a UK Biobank genome-wide association study of 39,691 brain imaging samples (predominantly of European origin). Using susceptibility-weighted images, which reflect iron load, we analysed genetically significant brain regions. Inverse variance weighting was used as the main estimate, while MR Egger and weighted median were used to detect heterogeneity and pleiotropy. Nine clear associations were obtained. For AD and PD, an increased iron load was causative: the right pallidum for AD and the right caudate, left caudate and right accumbens for PD. However, a reduced iron load was identified in the right and left caudate for multiple sclerosis, the bilateral hippocampus for mixed vascular dementia and the left thalamus and bilateral accumbens for subcortical vascular dementia. Thus, changes in iron load in different brain regions have causal effects on neurodegenerative diseases. Our results are crucial for understanding the pathogenesis and investigating the treatment of these diseases.

An interpretable deep learning framework for predicting liver metastases in postoperative colorectal cancer patients using natural language processing and clinical data integration.

BACKGROUND: The significance of liver metastasis (LM) in increasing the risk of death for postoperative colorectal cancer (CRC) patients necessitates innovative approaches to predict LM. AIM: Our study presents a novel and significant contribution by developing an interpretable fusion model that effectively integrates both free-text medical record data and structured laboratory data to predict LM in postoperative CRC patients. METHODS: We used a robust dataset of 1463 patients and leveraged state-of-the-art natural language processing (NLP) and machine learning techniques to construct a two-layer fusion framework that demonstrates superior predictive performance compared to single modal models. Our innovative two-tier algorithm fuses the results from different data modalities, achieving balanced prediction results on test data and significantly enhancing the predictive ability of the model. To increase interpretability, we employed Shapley additive explanations to elucidate the contributions of free-text clinical data and structured clinical data to the final model. Furthermore, we translated our findings into practical clinical applications by creating a novel NLP score-based nomogram using the top 13 valid predictors identified in our study. RESULTS: The proposed fusion models demonstrated superior predictive performance with an accuracy of 80.8%, precision of 80.3%, recall of 80.5%, and an F1 score of 80.8% in predicting LMs. CONCLUSION: This fusion model represents a notable advancement in predicting LMs for postoperative CRC patients, offering the potential to enhance patient outcomes and support clinical decision-making.

Bidirectional Mendelian randomization study of insulin-related traits and risk of ovarian cancer.

Background: It is well known that the occurrence and development of ovarian cancer are closely related to the patient's weight and various endocrine factors in the body. Aim: Mendelian randomization (MR) was used to analyze the bidirectional relationship between insulin related characteristics and ovarian cancer. Methods: The data on insulin related characteristics are from up to 5567 diabetes free patients from 10 studies, mainly including fasting insulin level, insulin secretion rate, peak insulin response, etc. For ovarian cancer, UK Biobank data just updated in 2021 was selected, of which the relevant gene data was from 199741 Europeans. Mendelian randomization method was selected, with inverse variance weighting (IVW) as the main estimation, while MR Pleiotropy, MR Egger, weighted median and other methods were used to detect the heterogeneity of data and whether there was multi validity affecting conclusions. Results: Among all insulin related indicators (fasting insulin level, insulin secretion rate, peak insulin response), the insulin secretion rate was selected to have a causal relationship with the occurrence of ovarian cancer (IVW, P < 0.05), that is, the risk of ovarian cancer increased with the decrease of insulin secretion rate. At the same time, we tested the heterogeneity and polymorphism of this indicator, and the results were non-existent, which ensured the accuracy of the analysis results. Reverse causal analysis showed that there was no causal effect between the two (P>0.05). Conclusion: The impairment of the insulin secretion rate has a causal effect on the risk of ovarian cancer, which was confirmed by Mendel randomization. This suggests that the human glucose metabolism cycle represented by insulin secretion plays an important role in the pathogenesis of ovarian cancer, which provides a new idea for preventing the release of ovarian cancer.

A machine learning approach to predict cerebral perfusion status based on internal carotid artery blood flow.

BACKGROUND AND OBJECTIVE: Cerebral blood flow (CBF), or perfusion, is a prerequisite for maintaining brain metabolism and normal physiological functions. Diagnosing and evaluating cerebral perfusion status is crucial to managing brain disease. However, cerebral perfusion imaging devices are complicated to operate, should be controlled by specialized technicians, are often large, and are usually installed in fixed places such as hospitals. It is significantly difficult for clinicians to obtain the cerebral perfusion status in time. Considering that CBF is mainly supplied by the internal carotid artery (ICA), this study proposes a cerebral perfusion status prediction model that can automatically quantify the level of cerebral perfusion in patients by modeling the association between ICA blood flow and cerebral perfusion. MATERIALS AND METHODS: Forty-eight participants were enrolled in the study after screening. We collected participants' ICA ultrasound and brain magnetic resonance imaging (MRI) data before and after dobutamine injection based on a rigorous experimental paradigm and built an ICA-cerebral perfusion datasetdd. Support vector machine (SVM), k-nearest neighbor (KNN), decision tree (DT), random forest (RF), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGBOOST) were used for early prediction of cerebral perfusion status. The SHAP analysis was adopted to reveal the impact of interpretable predictions for each feature. RESULTS: The XGBOOST model demonstrated the best overall classification performance with an accuracy of 78.01%, sensitivity of 96.67%, specificity of 98.23%, F1 score of 74.57%, Matthews correlation coefficient (MCC) of 62.17%, and area under the receiver operating characteristic curve (AUC) of 87.08%. Accelerated speed, peak systolic flow velocity, and resistance index of ICA blood flow are important factors for cerebral perfusion prediction. CONCLUSIONS: The proposed method paves a new avenue for the study of predicting cerebral perfusion status automatically and providesv a noninvasive, real-time, and low-cost alternative to brain perfusion imaging. Moreover, this analysis identifies highly predictive features for the cerebral perfusion status and gives clinicians an intuitive understanding of the influence of key features. The prediction models can serve as an early warning tool that offers sufficient time for clinicians to take early intervention measures.

Double decoupled network for imbalanced obstetric intelligent diagnosis.

Electronic Medical Record (EMR) is the data basis of intelligent diagnosis. The diagnosis results of an EMR are multi-disease, including normal diagnosis, pathological diagnosis and complications, so intelligent diagnosis can be treated as multi-label classification problem. The distribution of diagnostic results in EMRs is imbalanced. And the diagnostic results in one EMR have a high coupling degree. The traditional rebalancing methods does not function effectively on highly coupled imbalanced datasets. This paper proposes Double Decoupled Network (DDN) based intelligent diagnosis model, which decouples representation learning and classifier learning. In the representation learning stage, Convolutional Neural Networks (CNN) is used to learn the original features of the data. In the classifier learning stage, a Decoupled and Rebalancing highly Imbalanced Labels (DRIL) algorithm is proposed to decouple the highly coupled diagnostic results and rebalance the datasets, and then the balanced datasets is used to train the classifier. This paper evaluates the proposed DDN using Chinese Obstetric EMR (COEMR) datasets, and verifies the effectiveness and universality of the model on two benchmark multi-label text classification datasets: Arxiv Academic Papers Datasets (AAPD) and Reuters Corpus1 (RCV1). Demonstrating the effectiveness of the proposed methods is an imbalanced obstetric EMRs. The accuracy of DDN model on COEMR, AAPD and RCV1 datasets is 84.17, 86.35 and 93.87% respectively, which is higher than the current optimal experimental results.