Fast three-dimensional image generation for healthy brain aging using diffeomorphic registration.

Predicting brain aging can help in the early detection and prognosis of neurodegenerative diseases. Longitudinal cohorts of healthy subjects scanned through magnetic resonance imaging (MRI) have been essential to understand the structural brain changes due to aging. However, these cohorts suffer from missing data due to logistic issues in the recruitment of subjects. This paper proposes a methodology for filling up missing data in longitudinal cohorts with anatomically plausible images that capture the subject-specific aging process. The proposed methodology is developed within the framework of diffeomorphic registration. First, two novel modules are introduced within Synthmorph, a fast, state-of-the-art deep learning-based diffeomorphic registration method, to simulate the aging process between the first and last available MRI scan for each subject in three-dimensional (3D). The use of image registration also makes the generated images plausible by construction. Second, we used six image similarity measurements to rearrange the generated images to the specific age range. Finally, we estimated the age of every generated image by using the assumption of linear brain decay in healthy subjects. The methodology was evaluated on 2662 T1-weighted MRI scans from 796 healthy participants from 3 different longitudinal cohorts: Alzheimer's Disease Neuroimaging Initiative, Open Access Series of Imaging Studies-3, and Group of Neuropsychological Studies of the Canary Islands (GENIC). In total, we generated 7548 images to simulate the access of a scan per subject every 6 months in these cohorts. We evaluated the quality of the synthetic images using six quantitative measurements and a qualitative assessment by an experienced neuroradiologist with state-of-the-art results. The assumption of linear brain decay was accurate in these cohorts (R2  ∈ [.924, .940]). The experimental results show that the proposed methodology can produce anatomically plausible aging predictions that can be used to enhance longitudinal datasets. Compared to deep learning-based generative methods, diffeomorphic registration is more likely to preserve the anatomy of the different structures of the brain, which makes it more appropriate for its use in clinical applications. The proposed methodology is able to efficiently simulate anatomically plausible 3D MRI scans of brain aging of healthy subjects from two images scanned at two different time points.

Crystalline Porous Organic Salt for Ultrarapid Adsorption/Desorption-Based Atmospheric Water Harvesting by Dual Hydrogen Bond System.

In recent years, the porous sorbent-assisted atmospheric water harvesting (AWH) method has emerged as an effective approach for solving water crises without geographical restrictions. However, there is a limited array of porous adsorbent materials that can be used for AWH, which are inadequate to meet the needs under different climatic conditions. In light of this, herein, we synthesize a new crystalline porous organic salt (CPOS; denoted as CPOS-6) possessing a dual hydrogen bond system and verify its applicability toward AWH for the first time. Unlike other reported CPOSs, CPOS-6 displays an S-shaped water sorption isotherm owing to the presence of the dual hydrogen bond system. Under simulated drought conditions in Xinjiang Uygur Autonomous Region, CPOS-6 exhibits long-term water adsorption-desorption cycling stability, low water desorption temperature, and ultrarapid adsorption-desorption kinetics. The results confirm that CPOS-6 is an effective sorbent material for AWH.

Fabrication of COF-MOF Composite Membranes and Their Highly Selective Separation of H2/CO2.

The search for new types of membrane materials has been of continuous interest in both academia and industry, given their importance in a plethora of applications, particularly for energy-efficient separation technology. In this contribution, we demonstrate for the first time that a metal-organic framework (MOF) can be grown on the covalent-organic framework (COF) membrane to fabricate COF-MOF composite membranes. The resultant COF-MOF composite membranes demonstrate higher separation selectivity of H2/CO2 gas mixtures than the individual COF and MOF membranes. A sound proof for the synergy between two porous materials is the fact that the COF-MOF composite membranes surpass the Robeson upper bound of polymer membranes for mixture separation of a H2/CO2 gas pair and are among the best gas separation MOF membranes reported thus far.

Hsa_circ_0009096/miR-370-3p modulates hepatic stellate cell proliferation and fibrosis during biliary atresia pathogenesis.

Background: Hepatic stellate cell (HSC) activation and hepatic fibrosis mediated biliary atresia (BA) development, but the underlying molecular mechanisms are poorly understood. This study aimed to investigate the roles of circRNA hsa_circ_0009096 in the regulation of HSC proliferation and hepatic fibrosis. Methods: A cellular hepatic fibrosis model was established by treating LX-2 cells with transforming growth factor ß (TGF-ß1). RNaseR and actinomycin D assays were performed to detect hsa_circ_0009096 stability. Expression of hsa_circ_0009096, miR-370-3p, and target genes was detected using reverse transcription-qPCR. Direct binding of hsa_circ_0009096 to miR-370-3p was validated using dual luciferase reporter assay. Cell cycle progression and apoptosis of LX-2 cells were assessed using flow cytometry. The alpha-smooth muscle actin (α-SMA), collagen 1A1 (COL1A1), and TGF beta receptor 2 (TGFBR2) protein levels in LX-2 cells were analyzed using immunocytochemistry and western blotting. Results: Hsa_circ_0009096 exhibited more resistance to RNase R and actinomycinD digestion than UTRN mRNA. Hsa_circ_0009096 expression increased significantly in LX-2 cells treated with TGF-ß1, accompanied by elevated α-SMA and COL1A1 expression. Hsa_circ_0009096 siRNAs effectively promoted miR-370-3p and suppressed TGFBR2 expression in LX-2 cells, mediated by direct association of hsa_circ_0009096 with miR-370-3p. Hsa_circ_0009096 siRNA interfered with the cell cycle progression, promoted apoptosis, and reduced α-SMA and COL1A1 expression in LX-2 cells treated with TGF-ß1. MiR-370-3p inhibitors mitigated the alterations in cell cycle progression, apoptosis, and α-SMA, COL1A1, and TGFBR2 expression in LX-2 cells caused by hsa_circ_0009096 siRNA. In conclusion, hsa_circ_0009096 promoted HSC proliferation and hepatic fibrosis during BA pathogenesis by accelerating TGFBR2 expression by sponging miR-370-3p.

Investigating the Impact of Prompt Engineering on the Performance of Large Language Models for Standardizing Obstetric Diagnosis Text: Comparative Study.

BACKGROUND: The accumulation of vast electronic medical records (EMRs) through medical informatization creates significant research value, particularly in obstetrics. Diagnostic standardization across different health care institutions and regions is vital for medical data analysis. Large language models (LLMs) have been extensively used for various medical tasks. Prompt engineering is key to use LLMs effectively. OBJECTIVE: This study aims to evaluate and compare the performance of LLMs with various prompt engineering techniques on the task of standardizing obstetric diagnostic terminology using real-world obstetric data. METHODS: The paper describes a 4-step approach used for mapping diagnoses in electronic medical records to the International Classification of Diseases, 10th revision, observation domain. First, similarity measures were used for mapping the diagnoses. Second, candidate mapping terms were collected based on similarity scores above a threshold, to be used as the training data set. For generating optimal mapping terms, we used two LLMs (ChatGLM2 and Qwen-14B-Chat [QWEN]) for zero-shot learning in step 3. Finally, a performance comparison was conducted by using 3 pretrained bidirectional encoder representations from transformers (BERTs), including BERT, whole word masking BERT, and momentum contrastive learning with BERT (MC-BERT), for unsupervised optimal mapping term generation in the fourth step. RESULTS: LLMs and BERT demonstrated comparable performance at their respective optimal levels. LLMs showed clear advantages in terms of performance and efficiency in unsupervised settings. Interestingly, the performance of the LLMs varied significantly across different prompt engineering setups. For instance, when applying the self-consistency approach in QWEN, the F1-score improved by 5%, with precision increasing by 7.9%, outperforming the zero-shot method. Likewise, ChatGLM2 delivered similar rates of accurately generated responses. During the analysis, the BERT series served as a comparative model with comparable results. Among the 3 models, MC-BERT demonstrated the highest level of performance. However, the differences among the versions of BERT in this study were relatively insignificant. CONCLUSIONS: After applying LLMs to standardize diagnoses and designing 4 different prompts, we compared the results to those generated by the BERT model. Our findings indicate that QWEN prompts largely outperformed the other prompts, with precision comparable to that of the BERT model. These results demonstrate the potential of unsupervised approaches in improving the efficiency of aligning diagnostic terms in daily research and uncovering hidden information values in patient data.

Porous Materials for Atmospheric Water Harvesting.

Atmospheric Water Harvesting (AWH) using porous adsorbents is emerging as a promising solution to combat water shortage. Thus, a clearer understanding of the developing trends and optimization strategies of different porous adsorbents can be extremely helpful. Therefore, in this concept, the different types of porous adsorbents and AWH devices are briefly introduced with a focus on the factors that influence the static and kinetic properties of porous adsorbents and their respective optimization strategies. In addition, the fast transport characteristics of water molecules in micropores are studied from the perspective of superfluidity as part of the analysis of the kinetic properties of porous adsorbents. Finally, the future development of porous materials for AWH and the accompanying challenges are summarized.

miR-139-3p/Wnt5A Axis Inhibits Metastasis in Hepatoblastoma.

In order to examine new potential treatment options for the treatment of hepatoblastoma (HB), we identified the differential expression of five-candidate tumor suppressor miRNAs in HB and explored possible regulatory mechanisms of target miRNA molecule. By using real-time quantitative polymerase chain reaction (qPCR), we examined the expression of miRNAs in HB tissues and cells. The effect of has-miR-139-3p mimics on the invasion and migration ability was assessed by transwell assay and scratch-wound assay in HepG2 cells. Subsequently, we analyzed the target genes of miR-139-3p and their enrichment signaling pathways through bioinformatics. qPCR, Western-blot and dual-luciferase assays were further used to assess whether has-miR-139-3p targets Wnt5A. The results showed that hsa-miR-139-3p was significantly decreased in HB cells. Upregulation of hsa-miR-139-3p inhibited the invasive and migratory ability of HepG2. Bioinformatics analysis showed that hsa-miR-139-3p may target Wnt5A to regulate the WNT pathway, which was further confirmed by Western-blot and dual-luciferase assays. Overexpression of Wnt5A can reverse the miR-139-3p mimic-induced declines in the expression of WNT pathway-related proteins and restore the invasion and migration of HepG2. These data indicated that the hsa-miR-139-3p/Wnt5A axis inhibited HB metastasis, suggesting that miR-139-3p and Wnt5A may be potential targets for the treatment of HB.

Biomimetic KcsA channels with ultra-selective K+ transport for monovalent ion sieving.

Ultra-selective and fast transport of K+ are of significance for water desalination, energy conversion, and separation processes, but current bottleneck of achieving high-efficiency and exquisite transport is attributed to the competition from ions of similar dimensions and same valence through nanochannel communities. Here, inspired by biological KcsA channels, we report biomimetic charged porous subnanometer cages that enable ultra-selective K+ transport. For nanometer to subnanometer scales, conically structured double-helix columns exhibit typical asymmetric transport behaviors and conduct rapid K+ with a transport rate of 94.4 mmol m-2 h-1, resulting in the K+/Li+ and K+/Na+ selectivity ratios of 363 and 31, respectively. Experiments and simulations indicate that these results stem from the synergistic effects of cation-π and electrostatic interactions, which impose a higher energy barrier for Li+ and Na+ and lead to selective K+ transport. Our findings provide an effective methodology for creating in vitro biomimetic devices with high-performance K+ ion sieving.

Recent advances of biomimetic nano-systems in the diagnosis and treatment of tumor.

The lack of effective methods of diagnosis and treatment presents a major barrier to combat against tumor. The biomimetic concept is an emerging field that expresses great application potential in tumor fighting. Strategy for combining nano-systems with biomimetic technology has gained increasing attention that is proved bioinspired, environmentally benign, and promising. Herein, we provide an up-to-date review of biomimetic nano-systems as well as their applications in tumor therapy. In addition, the challenges and future directions of biomimetic nano-systems to achieve clinical translation are also pointed out.

Guide Subspace Learning for Unsupervised Domain Adaptation.

A prevailing problem in many machine learning tasks is that the training (i.e., source domain) and test data (i.e., target domain) have different distribution [i.e., non-independent identical distribution (i.i.d.)]. Unsupervised domain adaptation (UDA) was proposed to learn the unlabeled target data by leveraging the labeled source data. In this article, we propose a guide subspace learning (GSL) method for UDA, in which an invariant, discriminative, and domain-agnostic subspace is learned by three guidance terms through a two-stage progressive training strategy. First, the subspace-guided term reduces the discrepancy between the domains by moving the source closer to the target subspace. Second, the data-guided term uses the coupled projections to map both domains to a unified subspace, where each target sample can be represented by the source samples with a low-rank coefficient matrix that can preserve the global structure of data. In this way, the data from both domains can be well interlaced and the domain-invariant features can be obtained. Third, for improving the discrimination of the subspaces, the label-guided term is constructed for prediction based on source labels and pseudo-target labels. To further improve the model tolerance to label noise, a label relaxation matrix is introduced. For the solver, a two-stage learning strategy with teacher teaches and student feedbacks mode is proposed to obtain the discriminative domain-agnostic subspace. In addition, for handling nonlinear domain shift, a nonlinear GSL (NGSL) framework is formulated with kernel embedding, such that the unified subspace is imposed with nonlinearity. Experiments on various cross-domain visual benchmark databases show that our methods outperform many state-of-the-art UDA methods. The source code is available at https://github.com/Fjr9516/GSL.

A potential carrier for anti-tumor targeted delivery-hyaluronic acid nanoparticles.

In recent years, biomacromolecules have been widely used in anti-tumor delivery systems due to the biocompatibility and biodegradability. However, their applications are limited due to the lack of specific targeting. Hyaluronic acid (HA) is a natural polysaccharide and presents in extracellular matrix and synovial fluid which can specifically recognize receptors over-expressed by tumor cells. In addition, they can self-assemble into nanoparticles. HA nanoparticles provide new hierarchical targeting strategies: passively targeting tumor tissue by enhanced permeability and retention effect, actively targeting tumor cells by cluster determinant 44 (CD44) receptor, and then entering cells through receptor-mediated endocytosis. In this review, the synthesis of HA nanoparticles is described in detail from several aspects and applications are also discussed for improving the delivery of hydrophobic drugs, nucleic acids and photosensitizers into the tumor cells. In addition, the modification of HA for improving the targeting and drug releasing characteristics are also discussed.

The influence of trapping agents on the antitumor efficacy of irinotecan liposomes: head-to-head comparison of ammonium sulfate, sulfobutylether-ß-cyclodextrin and sucrose octasulfate.

Remote loading technology is an outstanding achievement in liposome-based drug delivery systems. Compared with conventional passive loading, remote loading technology exhibits unique superiority in terms of high drug loading efficiency, low leakage rate and adequate drug accumulation. In the intra-liposome aqueous phase, the counterion of the trapping agent can control the state of aggregation/crystallization of the drug-counterion salt, and thereby contribute to control the efficiency of remote loading. Herein, irinotecan (CPT-11)-loaded liposomes were developed using three trapping agents: ammonium sulfate (AS), sulfobutylether-ß-cyclodextrin (SBE-ß-CD) and sucrose octasulfate (SOS). The corresponding formulations were named as AS liposomal CPT-11, TEA-SBE-ß-CD liposomal CPT-11 and TEA-SOS liposomal CPT-11, respectively. Cryo-transmission electron micrographs showed that bundles of CPT-11 fibers were gathered inside TEA-SOS liposomal CPT-11. Furthermore, compared with AS liposomal CPT-11 and TEA-SBE-ß-CD liposomal CPT-11, TEA-SOS liposomal CPT-11 demonstrated slower drug release, prolonged circulation time and significantly improved antitumor efficiency. To avoid the protection of ONIVYDE®-related patents, a number of other liposomal CPT-11 formulations are under preclinical investigation or even in clinical trials. Our study gives new insights into the impact of the trapping agent on remote loading, and provides valuable information to evaluate the development of CPT-11 loaded liposomes.