Engineering Trifluoromethyl Groups in Porous Coordination Polymers to Enhance Stability and Regulate Pore Window for Hexane Isomers Separation.

Effective separation of hexane (C6) isomers is critical for a variety of industrial applications but conventional distillation methods are energy-intensive. Adsorptive separations based on porous coordination polymers (PCPs) offer a promising alternative due to their exceptional porosity and tunable properties. However, there is still an urgent need to develop PCPs with high stability and separation performance. This study investigates how substituting a methyl (-CH3) group with a trifluoromethyl (-CF3) group can regulate pores and hydrophobicity in PCPs. This precise adjustment aims to enhance stability and improve the kinetic separation performance of hydrophobic C6 isomers by considering the size and hydrophobicity of the trifluoromethyl group. Two isostructural PCPs with pcu topology, PCP-CH3 and PCP-CF3, were synthesized to vary pore diameters and hydrophobicity based on the presence of -CH3 or -CF3 groups. PCP-CF3 showed greater stability in water compared to PCP-CH3. While PCP-CH3 had high adsorption capacities, it lacked selectivity, whereas PCP-CF3 demonstrated improved selectivity, particularly in excluding dibranched isomers. Dynamic column separation experiments revealed that PCP-CF3 could selectively adsorb linear and monobranched isomers over dibranched isomers at room temperature. These findings highlight the potential of fluorine-modified PCPs for efficient isomer separation and underscore the importance of stability improvement strategies.

Real-world persistence and adherence of ofatumumab versus oral and injectable disease-modifying therapies in patients with multiple sclerosis.

BACKGROUND: Ofatumumab (OMB) was approved as a self-injectable disease-modifying therapy (DMT) for relapsing multiple sclerosis (MS) in August 2020. This study aimed to examine treatment persistence and adherence of OMB to oral and platform self-injectable DMTs in a real-world setting. METHODS: This was a retrospective cohort study using IQVIA Pharmetrics Plus® in adults diagnosed with MS and treated with OMB, oral DMTs, or platform self-injectable DMTs (index treatment) between August 2020 and November 2021. Patients had at least 12 months of continuous enrollment before the index date and 6 months of follow-up after the index date; the index date was defined as the date of the first pharmacy claim for an index treatment. Inverse probability of treatment weighting (IPTW) analysis was used to account for differences in baseline characteristics among patients initiating OMB versus oral DMTs and, separately, OMB versus platform self-injectable DMTs. Persistence was defined as the number of days from the index date until the earliest time of discontinuation (>60-day gap) or switch to a new DMT. Adherence was calculated based on proportion of days covered (PDC) with adherence defined as PDC ≥0.8. Persistence and adherence were compared between OMB versus oral DMTs and OMB versus platform self-injectable DMTs. Persistence was assessed via Kaplan-Meier analyses of the weighted sample, and log-rank tests were used to compare time to treatment discontinuation and treatment switch. The proportion of patients adherent at 6 and 12 months post index were compared using chi-square tests. RESULTS: A total of 11,167 patients were identified with an incident claim of OMB, an oral DMT, or a platform self-injectable DMT. After applying all inclusion and exclusion criteria and IPTW, two sets of study cohorts were created. For the oral DMT analysis, 577 patients treated with OMB and 2,468 patients treated with oral DMTs were identified. For the self-injectable DMT analysis, 574 patients treated with OMB and 578 patients treated with a platform self-injectable DMT were identified. At 6- and 12-month follow-up, the proportion of patients who were persistent on DMT was higher in the OMB cohort compared with the oral DMT cohort (6 months: 79.9% vs. 75.4 %, 12 months: 71.4% vs. 62.9 %; p < 0.05), as well as in the OMB cohort compared with the self-injectable DMT cohort (6 months: 79.3% vs. 60.4 %, 12 months: 71.5% vs. 48.7 %; p < 0.0001). A similar proportion of patients were adherent to OMB versus oral DMTs at 6- and 12-months post index, whereas a higher proportion of patients treated with OMB versus platform self-injectable DMTs were adherent at 6- and 12-months post index. CONCLUSIONS: OMB showed better persistence and adherence compared with platform self-injectable DMTs, as well as better persistence compared with oral DMTs. This real-world analysis provides additional insights into OMB utilization among patients with MS in clinical practice and demonstrates that OMB is a valuable treatment option for these patients.

Characterization of cryptic complex chromosome rearrangements in balanced chromosomal rearrangement carriers and their PGT-SR clinical outcome assessments.

Several reports have presented that balanced chromosomal rearrangements (BCRs) carriers with normal phenotypes may be carriers of complex rearrangements. However, the incidence and PGT clinical outcomes of cryptic complex chromosome rearrangements (CCCRs) in individuals with BCRs is remain unknown. We recruited a cohort of 1,264 individuals with BCR carriers from 2016 to 2021 at the Reproductive and Genetic Hospital of CITIC Xiangya. Peripheral blood was collected for karyotyping and genomic DNA extraction and the PGT-SR clinical outcomes of CCCRs carriers were analyzed and compared with those of BCR carriers. Our findings revealed that 3.6% (45/1,264) of BCR carriers had CCCRs, involving 3-25 breakpoints on 1-3 chromosomes. Furthermore, when mate-pair sequencing was employed, 63.3% (19/30) of CCCR carriers were found to have chromosome rearrangements that were different from those identified by the MicroSeq technique. And the transferable embryo rate of CCCR carriers with 3 chromosomes was significantly lower than that of CCCR carriers with only 1-2 chromosomes. In this research, we revealed that some of the BCR carriers were actually CCCR carriers, and the prognosis of PGT in CCCR carriers with one or two chromosomes is better than that of CCCR carriers with three chromosomes.

Multimodal fusion network for ICU patient outcome prediction.

Over the past decades, massive Electronic Health Records (EHRs) have been accumulated in Intensive Care Unit (ICU) and many other healthcare scenarios. The rich and comprehensive information recorded presents an exceptional opportunity for patient outcome predictions. Nevertheless, due to the diversity of data modalities, EHRs exhibit a heterogeneous characteristic, raising a difficulty to organically leverage information from various modalities. It is an urgent need to capture the underlying correlations among different modalities. In this paper, we propose a novel framework named Multimodal Fusion Network (MFNet) for ICU patient outcome prediction. First, we incorporate multiple modality-specific encoders to learn different modality representations. Notably, a graph guided encoder is designed to capture underlying global relationships among medical codes, and a text encoder with pre-fine-tuning strategy is adopted to extract appropriate text representations. Second, we propose to pairwise merge multimodal representations with a tailored hierarchical fusion mechanism. The experiments conducted on the eICU-CRD dataset validate that MFNet achieves superior performance on mortality prediction and Length of Stay (LoS) prediction compared with various representative and state-of-the-art baselines. Moreover, comprehensive ablation study demonstrates the effectiveness of each component of MFNet.

Biallelic variants in α-tubulin isotypes cause female infertility characterised as recurrent preimplantation embryo arrest.

BACKGROUND: Recurrent preimplantation embryo developmental arrest (RPEA) is the most common phenotype in assisted reproductive technology treatment failure associated with identified genetic abnormalities. Currently known maternal genetic variants explain only a limited number of cases. Variants of the ß-tubulin subunit gene, TUBB8, cause oocyte meiotic arrest and RPEA through a broad spectrum of spindle defects. In contrast, α-tubulin subunit genes are poorly studied in the context of preimplantation embryonic development. METHODS: Whole exome sequencing was performed on the PREA cohort. Functional characterisations of the identified candidate disease-causing variants were validated using Sanger sequencing, bioinformatics, in vitro functional analyses and single-cell RNA-sequencing of arrested embryos. RESULTS: Four homozygous variants were identified in the PREA cohort: two of TUBA1C (p.Gln358Ter and p.Asp444Metfs*42) and two of TUBA4A (p.Arg339Cys and p.Tyr440Ter). These variants cause varying degrees of spindle assembly defects. Additionally, we characterised changes in the human arrested embryo transcriptome carrying TUBA4A variants, with a particular focus on spindle organisation, chromosome segregation and mRNA decay. CONCLUSION: Our findings identified TUBA1C as a novel genetic marker and expanded the genetic and phenotypic spectrum of TUBA4A in female infertility and RPEA, which altogether highlighted the importance of α-tubulin isotypes in preimplantation embryonic development.

Very severe oligozoospermia with AZFc microdeletion patients may affect intracytoplasmic sperm injection clinical outcomes: A propensity score matching analysis.

Purpose: To explore whether spermatozoa from AZFc microdeletion patients affect their outcomes of intracytoplasmic sperm injection (ICSI). Methods: Eighty-five patients with AZFc microdeletion were recruited. A control group of one hundred and forty patients with severe oligozoospermia but without AZF microdeletion was selected using propensity score matching analysis with a 1:2 nearest neighbor algorithm ratio. The ICSI outcomes of the two groups were compared. Results: AZFc microdeletion had lower rates of normal fertilization (73% vs. 80%, p = 0.17) and high-quality embryos (44% vs. 58%, p = 0.07) than the control group. There was no significant difference in the clinical pregnancy rate, miscarriage rate, and live birth rate between the two groups. When the sperm concentration was <1 million/mL, the AZFc microdeletion group exhibited lower rates of fertilization (71% vs. 80%, p = 0.03), high-quality embryo (44% vs. 58%, p = 0.02), clinical pregnancy (57% vs. 76%, p = 0.02), and live birth (49% vs. 72%, p = 0.01) than the control group. However, if sperm concentration was ≥1 million/mL, no significant differences were found. Conclusion: If the sperm concentration is <1 million/mL, AZFc microdeletion do have a detrimental effect on most outcomes of ICSI.

A generalized AI system for human embryo selection covering the entire IVF cycle via multi-modal contrastive learning.

In vitro fertilization (IVF) has revolutionized infertility treatment, benefiting millions of couples worldwide. However, current clinical practices for embryo selection rely heavily on visual inspection of morphology, which is highly variable and experience dependent. Here, we propose a comprehensive artificial intelligence (AI) system that can interpret embryo-developmental knowledge encoded in vast unlabeled multi-modal datasets and provide personalized embryo selection. This AI platform consists of a transformer-based network backbone named IVFormer and a self-supervised learning framework, VTCLR (visual-temporal contrastive learning of representations), for training multi-modal embryo representations pre-trained on large and unlabeled data. When evaluated on clinical scenarios covering the entire IVF cycle, our pre-trained AI model demonstrates accurate and reliable performance on euploidy ranking and live-birth occurrence prediction. For AI vs. physician for euploidy ranking, our model achieved superior performance across all score categories. The results demonstrate the potential of the AI system as a non-invasive, efficient, and cost-effective tool to improve embryo selection and IVF outcomes.

FLT1 activation in cancer cells promotes PARP-inhibitor resistance in breast cancer.

Acquired resistance to PARP inhibitors (PARPi) remains a treatment challenge for BRCA1/2-mutant breast cancer that drastically shortens patient survival. Although several resistance mechanisms have been identified, none have been successfully targeted in the clinic. Using new PARPi-resistance models of Brca1- and Bard1-mutant breast cancer generated in-vivo, we identified FLT1 (VEGFR1) as a driver of resistance. Unlike the known role of VEGF signaling in angiogenesis, we demonstrate a novel, non-canonical role for FLT1 signaling that protects cancer cells from PARPi in-vivo through a combination of cell-intrinsic and cell-extrinsic pathways. We demonstrate that FLT1 blockade suppresses AKT activation, increases tumor infiltration of CD8+ T cells, and causes dramatic regression of PARPi-resistant breast tumors in a T-cell-dependent manner. Moreover, PARPi-resistant tumor cells can be readily re-sensitized to PARPi by targeting Flt1 either genetically (Flt1-suppression) or pharmacologically (axitinib). Importantly, a retrospective series of breast cancer patients treated with PARPi demonstrated shorter progression-free survival in cases with FLT1 activation at pre-treatment. Our study therefore identifies FLT1 as a potential therapeutic target in PARPi-resistant, BRCA1/2-mutant breast cancer.

Liquid Metal Interlayer for Ultrastable Solid-State Sodium Metal Battery.

Solid-state sodium metal batteries have attracted wide attention due to their high energy density, remarkable safety, and abundant sodium resources. However, the growth of Na dendrites and poor interfacial contact between Na metal anode and Na3Zr2Si2PO12 (NZSP) solid-state electrolytes severely limit their practical application. Herein, a wettable liquid metal (GaIn) interlayer significantly reduces the interfacial resistance and avoids the formation of voids at the Na/NZSP interface. Moreover, the Ga4Na and NaIn alloys at the interface caused by the spontaneous reaction of GaIn with Na metal enhance the bond of NZSP with Na anode, which provides a continuous Na+ diffusion pathway and homogeneous Na+ flux to suppress Na dendrite growth. The symmetric cell can cycle stably for over 6500 h at 0.05 mA cm-2 and over 3000 h at 0.1 mA cm-2, with a critical current density of 0.8 mA cm-2 at 25 °C, and the interfacial resistance is significantly reduced to 21.6 Ω from 1095.1 Ω. The full cell coupled with NaNi1/3Fe1/3Mn1/3O2 also shows outstanding cycling performance, maintaining 85.1% capacity after 100 cycles at 0.5 C. This work demonstrates that the liquid metal interlayer has a large potential for the practical application of solid-state metal batteries.

Low Driving Voltage Electroluminescence Device for Integrated Visual Strain Sensing.

As a pivotal component in human-machine interactions, display devices have undergone rapid development in modern life. Displays such as alternative current electroluminescence|alternative current electroluminescent (ACEL) devices with high flexibility and long operational lifetimes are essential for wearable electronics. However, ACEL devices are constrained by their inherent high driving voltage and complex fabrication processes. Our work presents an easy blade-coating method for fabricating flexible ACEL display devices based on an all-solution process. By dispersing BaTiO3 and ZnS/Cu powder into waterborne polyurethane, we successfully combined dielectric and fluorescence functionalities within a single layer, significantly reducing the device's driving voltage. Additionally, the ionic conducting hydrogel was chosen as a transparent electrode to achieve good electrical contact and strong interfacial adhesion through in situ polymerization. Owing to the unique method, our ACEL device exhibits high flexibility, low driving voltage (20-100 V), high brightness (300+ cd/m2 at 60 V), and environmental friendliness. Furthermore, by repurposing the hydrogel electrode, we integrated strain visualization capabilities within a single device, highlighting its potential for applications such as wearable healthcare monitoring.

A single-cell transcriptome atlas of human euploid and aneuploid blastocysts.

Aneuploidy is frequently detected in early human embryos as a major cause of early pregnancy failure. However, how aneuploidy affects cellular function remains elusive. Here, we profiled the transcriptomes of 14,908 single cells from 203 human euploid and aneuploid blastocysts involving autosomal and sex chromosomes. Nearly all of the blastocysts contained four lineages. In aneuploid chromosomes, 19.5% ± 1.2% of the expressed genes showed a dosage effect, and 90 dosage-sensitive domains were identified. Aneuploidy leads to prevalent genome-wide transcriptome alterations. Common effects, including apoptosis, were identified, especially in monosomies, partially explaining the lower cell numbers in autosomal monosomies. We further identified lineage-specific effects causing unstable epiblast development in aneuploidies, which was accompanied by the downregulation of TGF-ß and FGF signaling, which resulted in insufficient trophectoderm maturation. Our work provides crucial insights into the molecular basis of human aneuploid blastocysts and may shed light on the cellular interaction during blastocyst development.

Testing the generalizability and effectiveness of deep learning models among clinics: sperm detection as a pilot study.

BACKGROUND: Deep learning has been increasingly investigated for assisting clinical in vitro fertilization (IVF). The first technical step in many tasks is to visually detect and locate sperm, oocytes, and embryos in images. For clinical deployment of such deep learning models, different clinics use different image acquisition hardware and different sample preprocessing protocols, raising the concern over whether the reported accuracy of a deep learning model by one clinic could be reproduced in another clinic. Here we aim to investigate the effect of each imaging factor on the generalizability of object detection models, using sperm analysis as a pilot example. METHODS: Ablation studies were performed using state-of-the-art models for detecting human sperm to quantitatively assess how model precision (false-positive detection) and recall (missed detection) were affected by imaging magnification, imaging mode, and sample preprocessing protocols. The results led to the hypothesis that the richness of image acquisition conditions in a training dataset deterministically affects model generalizability. The hypothesis was tested by first enriching the training dataset with a wide range of imaging conditions, then validated through internal blind tests on new samples and external multi-center clinical validations. RESULTS: Ablation experiments revealed that removing subsets of data from the training dataset significantly reduced model precision. Removing raw sample images from the training dataset caused the largest drop in model precision, whereas removing 20x images caused the largest drop in model recall. by incorporating different imaging and sample preprocessing conditions into a rich training dataset, the model achieved an intraclass correlation coefficient (ICC) of 0.97 (95% CI: 0.94-0.99) for precision, and an ICC of 0.97 (95% CI: 0.93-0.99) for recall. Multi-center clinical validation showed no significant differences in model precision or recall across different clinics and applications. CONCLUSIONS: The results validated the hypothesis that the richness of data in the training dataset is a key factor impacting model generalizability. These findings highlight the importance of diversity in a training dataset for model evaluation and suggest that future deep learning models in andrology and reproductive medicine should incorporate comprehensive feature sets for enhanced generalizability across clinics.

Degradation of Complex Carbon Sources in Organic Fertilizers Facilitates Nitrogen Fixation.

Biological nitrogen fixation is crucial for agriculture and improving fertilizer efficiency, but organic fertilizers in enhancing this process remain debated. Here, we investigate the impact of organic fertilizers on biological nitrogen fixation through experiments and propose a new model where bacterial interactions with complex carbon sources enhance nitrogen fixation. Field experiments showed that adding organic fertilizers increased the nitrogenase activity by 57.85%. Subculture experiments revealed that organic fertilizer addition enriched genes corresponding to complex carbon and energy metabolism, as well as nifJ involved in electron transfer for nitrogenase. It also enhanced bacterial interactions and enhanced connectors associated with complex carbon degradation. Validation experiments demonstrated that combinations increased nitrogenase activity by 2.98 times compared to the single. Our findings suggest that organic fertilizers promoted nitrogen fixation by enhancing microbial cooperation, improved the degradation of complex carbon sources, and thereby provided utilizable carbon sources, energy, and electrons to N-fixers, thus increasing nitrogenase activity and nitrogen fixation.

All-natural 2D nanofluidics as highly-efficient osmotic energy generators.

Two-dimensional nanofluidics based on naturally abundant clay are good candidates for harvesting osmotic energy between the sea and river from the perspective of commercialization and environmental sustainability. However, clay-based nanofluidics outputting long-term considerable osmotic power remains extremely challenging to achieve due to the lack of surface charge and mechanical strength. Here, a two-dimensional all-natural nanofluidic (2D-NNF) is developed as a robust and highly efficient osmotic energy generator based on an interlocking configuration of stacked montmorillonite nanosheets (from natural clay) and their intercalated cellulose nanofibers (from natural wood). The generated nano-confined interlamellar channels with abundant surface and space negative charges facilitate selective and fast hopping transport of cations in the 2D-NNF. This contributes to an osmotic power output of ~8.61 W m-2 by mixing artificial seawater and river water, higher than other reported state-of-the-art 2D nanofluidics. According to detailed life cycle assessments (LCA), the 2D-NNF demonstrates great advantages in resource consumption (1/14), greenhouse gas emissions (1/9), and production costs (1/13) compared with the mainstream 2D nanofluidics, promising good sustainability for large-scale and highly-efficient osmotic power generation.

The oviposition preference and offspring performance of Aethina tumida (Coleoptera: Nitidulidae).

Given the rapid spread and potential harm caused by the small hive beetle, Aethina tumida (Coleoptera: Nitidulidae) in China, it has become imperative to comprehend the developmental biology of this invasive species. Currently, there is limited knowledge regarding the impact of A. tumida female oviposition site preference on larval growth and development. To examine this, we investigated the ovipositional preference of adult female A. tumida on bee pupae, beebread, banana, and honey through a free choice test. Furthermore, we assessed the impact of these food resources on offspring performance, which included larval development time, survival, wandering larvae weight, emerged adult body mass, reproduction, and juvenile hormone titer. Our results showed that A. tumida females exhibited a strong preference for ovipositing on bee pupae compared to other diets, while showing reluctance toward honey. Moreover, A. tumida larvae that were fed on bee pupae displayed accelerated growth compared to those fed on other diets. Furthermore, A. tumida fed on bee pupae exhibited higher weights for wandering larvae, and emerged adult, increased pupation rates, enhanced fecundity and fertility, as well as a larger number of unilateral ovarioles during the larval stage when compared to those fed on other diets. Overall, the results indicate that the oviposition preferences of A. tumida females are adaptive, as their choices can enhance the fitness of their offspring. This finding aligns broadly with the hypothesis of oviposition preference and larval performance. This study can provide a foundation for the development of attractants aimed at promoting the oviposition of the A. tumida adults.

CO2 utilization for methanol production: a review on the safety concerns and countermeasures.

The catalytic conversion of carbon dioxide is one of the important ways to achieve the goal of carbon neutralization, which can be further divided into electrocatalysis, thermal catalysis, and photocatalysis. Although photocatalysis and electrocatalysis have the advantages of mild reaction conditions and low energy consumption, the thermal catalytic conversion of CO2 has larger processing capacity, better reduction effect, and more complete industrial foundation, which is a promising technology in the future. During the development of new technology from laboratory to industrial application, ensuring the safety of production process is essential. In this work, safety optimization design of equipment, safety performance of catalysts, accident types, and their countermeasures in the industrial applications of CO2 to methanol are reviewed and discussed in depth. Based on that, future research demands for industrial process safety of CO2 to methanol were proposed, which provide guidance for the large-scale application of CO2 thermal catalytic conversion technology.

Hierarchical Porous Bimetallic FeMn Metal-Organic Framework Gel for Efficient Activation of Peracetic Acid in Antibiotic Degradation.

Effective techniques for eliminating antibiotics from water environments are in high demand. The peracetic acid (PAA)-based advanced oxidation process has recently drawn increasing attention for its effective antibiotic degrading capability. However, current applications of PAA-based techniques are limited and tend to have unsatisfactory performance. An additional catalyst for PAA activation could provide a promising solution to improve the performance of PAA. Bulky metal-organic framework gels (MOGs) stand out as ideal catalysts for PAA activation owing to their multiple advantages, including large surface areas, high porosity, and hierarchical pore systems. Herein, a bimetallic hierarchical porous structure, i.e., FeMn13BTC, was synthesized through a facile one-pot synthesis method and employed for PAA activation in ofloxacin (OFX) degradation. The optimized FeMn MOG/PAA system exhibited efficient catalytic performance, characterized by 81.85% OFX degradation achieved within 1 h owing to the specific hierarchical structure and synergistic effect between Fe and Mn ions, which greatly exceeded the performance of the only PAA-catalyzed system. Furthermore, the FeMn MOG/PAA system maintained >80% OFX degradation in natural water. Quenching experiments, electron spin resonance spectra, and model molecular degradation revealed that the primary reactive oxygen species responsible for the catalytic effect was R-O•, especially CH3C(=O)OO•, with minor contributions of •OH and 1O2. Overall, introduction of the MOG catalyst strategy for PAA activation achieved high antibiotic degradation performance, establishing a paradigm for the design of heterogeneous hierarchical systems to broaden the scope of catalyzed water treatment applications.

Flexible, Transparent and Conductive Metal Mesh Films with Ultra-High FoM for Stretchable Heating and Electromagnetic Interference Shielding.

Despite the growing demand for transparent conductive films in smart and wearable electronics for electromagnetic interference (EMI) shielding, achieving a flexible EMI shielding film, while maintaining a high transmittance remains a significant challenge. Herein, a flexible, transparent, and conductive copper (Cu) metal mesh film for EMI shielding is fabricated by self-forming crackle template method and electroplating technique. The Cu mesh film shows an ultra-low sheet resistance (0.18 Ω â–¡-1), high transmittance (85.8%@550 nm), and ultra-high figure of merit (> 13,000). It also has satisfactory stretchability and mechanical stability, with a resistance increases of only 1.3% after 1,000 bending cycles. As a stretchable heater (ε > 30%), the saturation temperature of the film can reach over 110 °C within 60 s at 1.00 V applied voltage. Moreover, the metal mesh film exhibits outstanding average EMI shielding effectiveness of 40.4 dB in the X-band at the thickness of 2.5 µm. As a demonstration, it is used as a transparent window for shielding the wireless communication electromagnetic waves. Therefore, the flexible and transparent conductive Cu mesh film proposed in this work provides a promising candidate for the next-generation EMI shielding applications.

Prognostic nomogram for patients with tongue squamous cell carcinoma: A SEER-based study.

OBJECTIVES: In order to predict the patients' prognosis with tongue squamous cell carcinoma (SCC), this study set out to develop a clinically useful and trustworthy prognostic nomogram. SUBJECTS AND METHODS: The Surveillance, Epidemiology, and End Results (SEER) Program was used to compile clinical information on patients with tongue SCC between 2010 and 2015. The likelihood of Cancer-Specific Survival (CSS) and Overall Survival (OS) for specific patients was predicted using a prognostic nomogram created with the help of the RStudio software. The nomogram's predictive ability was evaluated using the consistency index (C-index) and decision curve analysis, and the nomogram was calibrated for 1-, 2-, 3-, 5-, and 10-year CSS and OS. RESULTS: Patients numbering 6453were enrolled in this study. The primary cohort (3895) and validation cohort (2558) were each randomly assigned. Sex, age, tumor-node-metastasis (TNM) stage, surgery, chemotherapy, and radiation were significant risk factors for OS, whereas age, TNM stage, surgery, chemotherapy, and radiotherapy were significant risk factors for CSS. Additionally, C-index and calibration curves indicated that the prognostic nomogram prediction and the actual observation in both cohorts would be very coherent. CONCLUSIONS: The predictive nomogram created in this study can offer patients with tongue SCC customized treatment and survival risk assessment.

Graph-guided deep hashing networks for similar patient retrieval.

With the rapid growth and widespread application of electronic health records (EHRs), similar patient retrieval has become an important task for downstream clinical decision support such as diagnostic reference, treatment planning, etc. However, the high dimensionality, large volume, and heterogeneity of EHRs pose challenges to the efficient and accurate retrieval of patients with similar medical conditions to the current case. Several previous studies have attempted to alleviate these issues by using hash coding techniques, improving retrieval efficiency but merely exploring underlying characteristics among instances to preserve retrieval accuracy. In this paper, drug categories of instances recorded in EHRs are regarded as the ground truth to determine the pairwise similarity, and we consider the abundant semantic information within such multi-labels and propose a novel framework named Graph-guided Deep Hashing Networks (GDHN). To capture correlation dependencies among the multi-labels, we first construct a label graph where each node represents a drug category, then a graph convolution network (GCN) is employed to derive the multi-label embedding of each instance. Thus, we can utilize the learned multi-label embeddings to guide the patient hashing process to obtain more informative and discriminative hash codes. Extensive experiments have been conducted on two datasets, including a real-world dataset concerning IgA nephropathy from Peking University First Hospital, and a publicly available dataset from MIMIC-III, compared with traditional hashing methods and state-of-the-art deep hashing methods using three evaluation metrics. The results demonstrate that GDHN outperforms the competitors at different hash code lengths, validating the superiority of our proposal.