Large-Kernel Attention for 3D Medical Image Segmentation.

Automated segmentation of multiple organs and tumors from 3D medical images such as magnetic resonance imaging (MRI) and computed tomography (CT) scans using deep learning methods can aid in diagnosing and treating cancer. However, organs often overlap and are complexly connected, characterized by extensive anatomical variation and low contrast. In addition, the diversity of tumor shape, location, and appearance, coupled with the dominance of background voxels, makes accurate 3D medical image segmentation difficult. In this paper, a novel 3D large-kernel (LK) attention module is proposed to address these problems to achieve accurate multi-organ segmentation and tumor segmentation. The advantages of biologically inspired self-attention and convolution are combined in the proposed LK attention module, including local contextual information, long-range dependencies, and channel adaptation. The module also decomposes the LK convolution to optimize the computational cost and can be easily incorporated into CNNs such as U-Net. Comprehensive ablation experiments demonstrated the feasibility of convolutional decomposition and explored the most efficient and effective network design. Among them, the best Mid-type 3D LK attention-based U-Net network was evaluated on CT-ORG and BraTS 2020 datasets, achieving state-of-the-art segmentation performance when compared to avant-garde CNN and Transformer-based methods for medical image segmentation. The performance improvement due to the proposed 3D LK attention module was statistically validated.

Hunting imaging biomarkers in pulmonary fibrosis: Benchmarks of the AIIB23 challenge.

Airway-related quantitative imaging biomarkers are crucial for examination, diagnosis, and prognosis in pulmonary diseases. However, the manual delineation of airway structures remains prohibitively time-consuming. While significant efforts have been made towards enhancing automatic airway modelling, current public-available datasets predominantly concentrate on lung diseases with moderate morphological variations. The intricate honeycombing patterns present in the lung tissues of fibrotic lung disease patients exacerbate the challenges, often leading to various prediction errors. To address this issue, the 'Airway-Informed Quantitative CT Imaging Biomarker for Fibrotic Lung Disease 2023' (AIIB23) competition was organized in conjunction with the official 2023 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI). The airway structures were meticulously annotated by three experienced radiologists. Competitors were encouraged to develop automatic airway segmentation models with high robustness and generalization abilities, followed by exploring the most correlated QIB of mortality prediction. A training set of 120 high-resolution computerised tomography (HRCT) scans were publicly released with expert annotations and mortality status. The online validation set incorporated 52 HRCT scans from patients with fibrotic lung disease and the offline test set included 140 cases from fibrosis and COVID-19 patients. The results have shown that the capacity of extracting airway trees from patients with fibrotic lung disease could be enhanced by introducing voxel-wise weighted general union loss and continuity loss. In addition to the competitive image biomarkers for mortality prediction, a strong airway-derived biomarker (Hazard ratio>1.5, p < 0.0001) was revealed for survival prognostication compared with existing clinical measurements, clinician assessment and AI-based biomarkers.

Using artificial intelligence and predictive modelling to enable learning healthcare systems (LHS) for pandemic preparedness.

In anticipation of potential future pandemics, we examined the challenges and opportunities presented by the COVID-19 outbreak. This analysis highlights how artificial intelligence (AI) and predictive models can support both patients and clinicians in managing subsequent infectious diseases, and how legislators and policymakers could support these efforts, to bring learning healthcare system (LHS) from guidelines to real-world implementation. This report chronicles the trajectory of the COVID-19 pandemic, emphasizing the diverse data sets generated throughout its course. We propose strategies for harnessing this data via AI and predictive modelling to enhance the functioning of LHS. The challenges faced by patients and healthcare systems around the world during this unprecedented crisis could have been mitigated with an informed and timely adoption of the three pillars of the LHS: Knowledge, Data and Practice. By harnessing AI and predictive analytics, we can develop tools that not only detect potential pandemic-prone diseases early on but also assist in patient management, provide decision support, offer treatment recommendations, deliver patient outcome triage, predict post-recovery long-term disease impacts, monitor viral mutations and variant emergence, and assess vaccine and treatment efficacy in real-time. A patient-centric approach remains paramount, ensuring patients are both informed and actively involved in disease mitigation strategies.

First-principles studies on the process of electron transfer between hydrophobic liquids and water.

Using the density functional theory, we conducted a study on the electrification upon contact between hydrophobic liquid molecules and water molecules, revealing localized characteristics of contact-electrification. These "localized features" refer to the specific microscale characteristics where electron transfer predominantly occurs at the contact regions, influenced by factors such as atomic distances and molecular orientations. Although the electrostatic potential and the highest occupied molecular orbital-lowest unoccupied molecular orbital gap offer substantial predictive insights for electron transfer across polymer interfaces, they fall short in capturing the complexities associated with the interaction between hydrophobic liquids and water molecules. The electronegativity of elements at the interface and the localization of molecular orbitals play a decisive role in electron transfer. Simultaneously, for liquid molecules with irregular structures, there is no correlation between the "contact area" and the amount of electron transfer. The "contact area" refers to the surface region where two different liquid molecules come into close proximity. It is defined by the surface area of atoms with interatomic distances smaller than the van der Waals radius. This study challenges traditional assumptions about contact-electrification, particularly in liquid-liquid interfaces, providing new insights into the localized nature of this phenomenon.

FDA Approval Summary: Fruquintinib for the Treatment of Refractory Metastatic Colorectal Cancer.

On November 8, 2023, the FDA approved fruquintinib, an inhibitor of vascular endothelial growth factor receptors (VEGFR)-1, -2, and -3, for the treatment of patients with metastatic colorectal cancer (mCRC) who have been previously treated with fluoropyrimidine­, oxaliplatin­, and irinotecan­based chemotherapy, an anti­VEGF therapy, and, if RAS wild­type and medically appropriate, an anti EGFR therapy. Approval was based on Study FRESCO-2, a globally-conducted, double-blind, placebo-controlled randomized trial. The primary endpoint was overall survival (OS). The key secondary endpoint was progression-free survival (PFS). A total of 691 patients were randomized (461 and 230 into the fruquintinib and placebo arms, respectively). Fruquintinib provided a statistically significant improvement in OS with a hazard ratio (HR) of 0.66 (95% CI: 0.55, 0.80; p<0.001). The median OS was 7.4 months (95% CI: 6.7, 8.2) in the fruquintinib arm and 4.8 months (95% CI: 4.0, 5.8) for the placebo arm. Adverse events observed were generally consistent with the known safety profile associated with inhibition of the VEGFR. The results of FRESCO-2 were supported by the FRESCO study, a double-blind, single country, placebo-controlled, randomized trial in patients with refractory mCRC who have been previously treated with fluoropyrimidine­, oxaliplatin­, and irinotecan­based chemotherapy. In FRESCO, the OS HR was 0.65 (95% CI: 0.51, 0.83; p<0.001). FDA concluded that the totality of the evidence from FRESCO-2 and FRESCO supported an indication for patients with mCRC with prior treatment with fluoropyrimidine, oxaliplatin-, and irinotecan-based chemotherapy, an anti-VEGF biological therapy, and if RAS wild­type and medically appropriate, an anti-EGFR therapy.

Constructing Chainmail-Structured CoP/C Nanospheres as Highly Active Anodic Electrocatalysts for Oxygen Evolution Reaction.

Constructing highly active and noble metal-free electrocatalysts is significant for the anodic oxygen evolution reaction (OER). Herein, uniform carbon-coated CoP nanospheres (CoP/C) are developed by a direct impregnation coupling phosphorization approach. Importantly, CoP/C only takes a small overpotential of 230 mV at the current density of 10 mA cm-2 and displays a Tafel slope of 56.87 mV dec-1. Furthermore, the intrinsic activity of CoP/C is 21.44 times better than that of commercial RuO2 under an overpotential of 260 mV. In situ Raman spectroscopy studies revealed that a large number of generated Co-O and Co-OH species could facilitate the *OH adsorption, effectively accelerating the reaction kinetics. Meanwhile, the carbon shell with a large number of mesoporous pores acts as the chainmail of CoP, which could improve the active surface area of the catalyst and prevent the Co sites from oxidative dissolution. This work provides a facile and effective reference for the development of highly active and stable OER catalysts.

Preoperative anemia: impact on short- and long-term outcomes following curative-intent resection of gastroenteropancreatic neuroendocrine tumors.

BACKGROUND: The effect of preoperative anemia on clinical outcomes of patients undergoing resection of gastroenteropancreatic neuroendocrine tumors (GEP-NETs) has not been previously investigated. This study aimed to characterize how preoperative anemia affected short- and long-term outcomes of patients undergoing curative-intent resection of GEP-NETs. METHODS: Patients who underwent curative-intent resection for GEP-NETs between January 1990 and December 2020 were identified from 8 major institutions. The last preoperative hemoglobin level was recorded; anemia was defined as <13.5 g/dL in males or <12.0 g/dL in females based on the guides of the American Society of Hematology. The effect of anemia on postoperative outcomes was assessed on uni- and multivariate analyses. RESULTS: Among 1559 patients, the median age was 58 years (IQR, 48-66), and roughly one-half of the cohort was male (796 [51.1%]). Most patients had a pancreatic tumor (1040 [66.7%]), followed by small bowel (259 [16.6%]), duodenum (103 [6.6%]), stomach (66 [4.2%]), appendix (53 [3.4%]), and other locations (38 [2.6%]). The median preoperative hemoglobin level was 13.4 g/dL (IQR, 12.2-14.5). Overall, 101 (6.7%) and 119 (8.5%) patients received an intra- or postoperative packed red blood cell (pRBC) transfusion, respectively. A total of 972 patients (44.5%) experienced a postoperative complication. Although the overall incidence of complications was no different among patients who did (anemic: 48.7%) vs patients who did not (nonanemic: 47.3%) have anemia (P = .597), patients with preoperative anemia were more likely to develop a major (Clavien-Dindo grade ≥IIIa: 48.9% [anemic] vs 38.0% [nonanemic]; P = .006) and multiple (≥3 types of complications: 32.2% [anemic] vs 19.7% [anemic]; P < .001) complications. Of note, 1-, 3-, and 5-year overall survival (OS) rates were 96.7%, 90.5%, and 86.6%, respectively. On multivariable analysis, anemia (hazard ratio, 2.0; 95% CI, 1.2-3.2; P = .006) remained associated with worse OS; postoperative pRBC transfusion was associated with an OS (5-year OS: 75.0% vs 87.7%; P = .017) and recurrence-free survival (RFS; 5-year RFS: 66.9% vs 76.5%; P = .047). CONCLUSION: Preoperative anemia was commonly identified in roughly 1 in 3 patients who underwent curative-intent resection for GEP-NETs. Preoperative anemia was strongly associated with a higher risk of postoperative morbidity and worse long-term outcomes.

Real-Time Non-Invasive Imaging and Detection of Spreading Depolarizations through EEG: An Ultra-Light Explainable Deep Learning Approach.

A core aim of neurocritical care is to prevent secondary brain injury. Spreading depolarizations (SDs) have been identified as an important independent cause of secondary brain injury. SDs are usually detected using invasive electrocorticography recorded at high sampling frequency. Recent pilot studies suggest a possible utility of scalp electrodes generated electroencephalogram (EEG) for non-invasive SD detection. However, noise and attenuation of EEG signals makes this detection task extremely challenging. Previous methods focus on detecting temporal power change of EEG over a fixed high-density map of scalp electrodes, which is not always clinically feasible. Having a specialized spectrogram as an input to the automatic SD detection model, this study is the first to transform SD identification problem from a detection task on a 1-D time-series wave to a task on a sequential 2-D rendered imaging. This study presented a novel ultra-light-weight multi-modal deep-learning network to fuse EEG spectrogram imaging and temporal power vectors to enhance SD identification accuracy over each single electrode, allowing flexible EEG map and paving the way for SD detection on ultra-low-density EEG with variable electrode positioning. Our proposed model has an ultra-fast processing speed (<0.3 sec). Compared to the conventional methods (2 hours), this is a huge advancement towards early SD detection and to facilitate instant brain injury prognosis. Seeing SDs with a new dimension - frequency on spectrograms, we demonstrated that such additional dimension could improve SD detection accuracy, providing preliminary evidence to support the hypothesis that SDs may show implicit features over the frequency profile.

IKBKE promotes the ZEB2-mediated EMT process by phosphorylating HMGA1a in glioblastoma.

IKBKE (Inhibitor of Nuclear Factor Kappa-B Kinase Subunit Epsilon) is an important oncogenic protein in a variety of tumors, which can promote tumor growth, proliferation, invasion and drug resistance, and plays a critical regulatory role in the occurrence and progression of malignant tumors. HMGA1a (High Mobility Group AT-hook 1a) functions as a cofactor for proper transcriptional regulation and is highly expressed in multiple types of tumors. ZEB2 (Zinc finger E-box Binding homeobox 2) exerts active functions in epithelial mesenchymal transformation (EMT). In our current study, we confirmed that IKBKE can increase the proliferation, invasion and migration of glioblastoma cells. We then found that IKBKE can phosphorylate HMGA1a at Ser 36 and/or Ser 44 sites and inhibit the degradation process of HMGA1a, and regulate the nuclear translocation of HMGA1a. Crucially, we observed that HMGA1a can regulate ZEB2 gene expression by interacting with ZEB2 promoter region. Hence, HMGA1a was found to promote the ZEB2-related metastasis. Consequently, we demonstrated that IKBKE can exert its oncogenic functions via the IKBKE/HMGA1a/ZEB2 signalling axis, and IKBKE may be a prominent biomarker for the treatment of glioblastoma in the future.

Three 2,3-diketoquinoxaline alkaloids with hepatoprotective activity from Heterosmilax yunnanensis / 药学学报

Three 2,3-diketoquinoxaline alkaloids were isolated from Heterosmilax yunnanensis Gagnep. Their structures were determined through 1D and 2D NMR, HR-ESI-MS, UV, and IR as 1-[5′-(3″-hydroxy-3″-methyl) glutaryl] ribityl-2,3-diketo-1,2,3,4-tetrahydro-6,7-dimethylquinoxaline (1), 1-[2′-(3″-hydroxy-3″-methyl) glutaryl]ribityl-2,3-diketo-1,2,3,4-tetrahydro-6,7-dimethylquinoxaline (2), and 1-ribityl-2,3-diketo-1,2,3,4-tetrahydro-6,7-dimethylquinoxaline (3). Compounds 1 and 2 are novel compounds, and 3 was isolated from H. yunnanensis for the first time. The hepatoprotective activity of these three compounds was evaluated, with compound 3 showing promising hepatoprotective activity.

Research progress of dysthyroid optic neuropathy / 国际眼科杂志(Guoji Yanke Zazhi)

Dysthyroid optic neuropathy is an important secondary pathological condition of thyroid-associated ophthalmopathy, characterized clinically by several clinical manifestations, including reduced visual acuity, impairment of color vision, relative afferent pupillary defect, and optic disk edema or atrophy. Ophthalmological auxiliary examination shows abnormal vision field and visual evoked potential, etc., and imagining examination shows orbital apex crowding, which can assist diagnosis. The pathogenesis of this disease is still unclear. With previous studies proposing that it was related to optic nerve compression, stretch, and ischemia. Treatment methods include high-dose intravenous glucocorticoid, orbital decompression, orbital radiation therapy, and biological agent. This article systematically reviews the research progress on the epidemiological characteristics, pathogenesis, diagnosis, and treatment of this disease, with a view to providing useful reference for future in-depth clinical practice and scientific research.

Improvement effect and mechanism of “Layers adjusting external application”paste on synovial fibrosis in rats with knee osteoarthritis / 中国药房

OBJECTIVE To investigate the improvement effect and potential mechanism of “Layers adjusting external application” paste on synovial fibrosis （SF） in rats with knee osteoarthritis （KOA）. METHODS Male SD rats were randomly divided into sham operation group， KOA group and Layers adjusting external application group， with 8 rats in each group. KOA model was induced by the anterior cruciate ligament disruption method in KOA group and Layers adjusting external application group. Fourteen days after modeling， the Layers adjusting external application group was given “Layers adjusting external application” paste [Sanse powder （8 g for every 100 cm2）， Compound sanhuang ointment （5 g for every 100 cm2）] on the knee joint， 8 h every day， for 28 d in total. After the last administration， the degree of synovitis and fibrosis in rats was observed， and Krenn scoring was performed in each group. The expressions of collagen Ⅰ， high mobility group protein B1 （HMGB1） and phosphorylated nuclear factor-κB p65 （p-NF-κB p65） were detected in the synovial membrane； the contents of interleukin-1β （IL- 1β）， IL-6 and tumor necrosis factor-α （TNF-α） in serum as well as the expressions of fibrosis-related and HMGB1/Toll-like receptor 4 （TLR4）/NF-κB signaling pathway-related proteins and mRNA were detected in synovial tissue. RESULTS Compared with the sham operation group， the synovial lining cells in the KOA group showed significant proliferation and disordered arrangement， the inflammatory cell infiltration and collagen fiber deposition were obvious； the positive expressing cells of collagen Ⅰ， HMGB1 and p-NF-κB p65 were increased significantly； the contents of IL-1β， IL-6 and TNF-α in serum， the expressions of fibrosis-related protein （transforming growth factor-β， collagen Ⅰ， tissue inhibitor of metalloproteinase 1， α-smooth muscle actin） and their mRNA as well as theexpressions of HMGB1， TLR4 protein and their mRNA， the expressions of p-NF-κB p65 protein and NF-κB p65 mRNA were all increased significantly in synovial tissues of rats （P＜0.01）. Compared with the KOA group， the pathological changes in the synovial tissue of rats in Layers adjusting external application group were significantly improved， and the above quantitative indicators were significantly reversed （P＜0.05 or P＜0.01）. CONCLUSIONS “Layers adjusting external application” paste could significantly improve SF in KOA rats， the mechanism of which may be associated with the inhibition of the activation of HMGB1/ TLR4/NF-κB signaling pathway.

Lipid profiling reveals Leymus Chinensis root insensitivity to Ca limitation.

BACKGROUND: Leymus chinensis (L. chinensis) is a perennial native forage grass widely distributed in the steppe of Inner Mongolia as the dominant species. Calcium (Ca) is an essential mineral element important for plant adaptation to the growth environment. Ca limitation was previously shown to strongly inhibit Arabidopsis (Arabidopsis thaliana) seedling growth and disrupt plasma membrane stability and selectivity, increasing fluid-phase-based endocytosis and contents of all major membrane lipids. RESULTS: In this study, we investigated the significance of Ca for L. chinensis growth and membrane stability relative to Arabidopsis. Our results showed that Ca limitation did not affect L. chinensis seedling growth and endocytosis in roots. Moreover, the plasma membrane maintained high selectivity. The lipid phosphatidylcholine (PC): phosphatidylethanolamine (PE) ratio, an indicator of the membrane stability, was five times higher in L. chinensis than in Arabidopsis. Furthermore, in L. chinensis, Ca limitation did not affect the content of any major lipid types, decreased malondialdehyde (MDA) content, and increased superoxide dismutase (SOD) activity, showing an opposite pattern to that in Arabidopsis. L. chinensis roots accumulated higher contents of PC, phosphatidylinositol (PI), monogalactosyldiacylglycerol (MGDG), phosphatidylglycerol (PG), cardiolipin (CL), digalactosyldiacylglycerol (DGDG), and lysophosphatidylcholine (LPC) but less phosphatidylethanolamine (PE), diacylglycerol (DAG), triacylglycerolv (TAG), phosphatidylserine (PS), lysobisphosphatidic acids (LPAs), lysophosphatidylethanolamine (LPE), and lysophosphatidylserine (LPS) than Arabidopsis roots. Moreover, we detected 31 and 66 unique lipids in L. chinensis and Arabidopsis, respectively. CONCLUSIONS: This study revealed that L. chinensis roots have unique membrane lipid composition that was not sensitive to Ca limitation, which might contribute to the wider natural distribution of this species.

Region-based evidential deep learning to quantify uncertainty and improve robustness of brain tumor segmentation.

Despite recent advances in the accuracy of brain tumor segmentation, the results still suffer from low reliability and robustness. Uncertainty estimation is an efficient solution to this problem, as it provides a measure of confidence in the segmentation results. The current uncertainty estimation methods based on quantile regression, Bayesian neural network, ensemble, and Monte Carlo dropout are limited by their high computational cost and inconsistency. In order to overcome these challenges, Evidential Deep Learning (EDL) was developed in recent work but primarily for natural image classification and showed inferior segmentation results. In this paper, we proposed a region-based EDL segmentation framework that can generate reliable uncertainty maps and accurate segmentation results, which is robust to noise and image corruption. We used the Theory of Evidence to interpret the output of a neural network as evidence values gathered from input features. Following Subjective Logic, evidence was parameterized as a Dirichlet distribution, and predicted probabilities were treated as subjective opinions. To evaluate the performance of our model on segmentation and uncertainty estimation, we conducted quantitative and qualitative experiments on the BraTS 2020 dataset. The results demonstrated the top performance of the proposed method in quantifying segmentation uncertainty and robustly segmenting tumors. Furthermore, our proposed new framework maintained the advantages of low computational cost and easy implementation and showed the potential for clinical application.

Multi-site, Multi-domain Airway Tree Modeling.

Open international challenges are becoming the de facto standard for assessing computer vision and image analysis algorithms. In recent years, new methods have extended the reach of pulmonary airway segmentation that is closer to the limit of image resolution. Since EXACT'09 pulmonary airway segmentation, limited effort has been directed to the quantitative comparison of newly emerged algorithms driven by the maturity of deep learning based approaches and extensive clinical efforts for resolving finer details of distal airways for early intervention of pulmonary diseases. Thus far, public annotated datasets are extremely limited, hindering the development of data-driven methods and detailed performance evaluation of new algorithms. To provide a benchmark for the medical imaging community, we organized the Multi-site, Multi-domain Airway Tree Modeling (ATM'22), which was held as an official challenge event during the MICCAI 2022 conference. ATM'22 provides large-scale CT scans with detailed pulmonary airway annotation, including 500 CT scans (300 for training, 50 for validation, and 150 for testing). The dataset was collected from different sites and it further included a portion of noisy COVID-19 CTs with ground-glass opacity and consolidation. Twenty-three teams participated in the entire phase of the challenge and the algorithms for the top ten teams are reviewed in this paper. Both quantitative and qualitative results revealed that deep learning models embedded with the topological continuity enhancement achieved superior performance in general. ATM'22 challenge holds as an open-call design, the training data and the gold standard evaluation are available upon successful registration via its homepage (https://atm22.grand-challenge.org/).

Acanthopanax senticosus Harms improves Parkinson's disease by regulating gut microbial structure and metabolic disorders.

Parkinson's disease (PD) is the second most common neurodegenerative disease, with an increasing prevalence as the population ages, posing a serious threat to human health, but the pathogenesis remains uncertain. Acanthopanax senticosus (Rupr. et Maxim.) Harms (ASH) (aqueous ethanol extract), a Chinese herbal medicine, provides obvious and noticeable therapeutic effects on PD. To further investigate the ASH's mechanism of action in treating PD, the structural and functional gut microbiota, as well as intestinal metabolite before and after ASH intervention in the PD mice model, were examined utilizing metagenomics and fecal metabolomics analysis. α-syn transgenic mice were randomly divided into a model and ASH groups, with C57BL/6 mice as a control. The ASH group was gavaged with ASH (45.5 mg/kg/d for 20d). The time of pole climbing and autonomous activity were used to assess motor ability. The gut microbiota's structure, composition, and function were evaluated using Illumina sequencing. Fecal metabolites were identified using UHPLC-MS/MS to construct intestinal metabolites. The findings of this experiment demonstrate that ASH may reduce the climbing time of PD model mice while increasing the number of autonomous movements. The results of metagenomics analysis revealed that ASH could up-regulated Firmicutes and down-regulated Actinobacteria at the phylum level, while Clostridium was up-regulated and Akkermansia was down-regulated at the genus level; it could also recall 49 species from the phylum Firmicutes, Actinobacteria, and Tenericutes. Simultaneously, metabolomics analysis revealed that alpha-Linolenic acid metabolism might be a key metabolic pathway for ASH to impact in PD. Furthermore, metagenomics function analysis and metabolic pathway enrichment analysis revealed that ASH might influence unsaturated fatty acid synthesis and purine metabolism pathways. These metabolic pathways are connected to ALA, Palmitic acid, Adenine, and 16 species of Firmicutes, Actinobacteria, and Tenericutes. Finally, these results indicate that ASH may alleviate the movement disorder of the PD model, which may be connected to the regulation of gut microbiota structure and function as well as the modulation of metabolic disorders by ASH.

Enhanced Charge Balance for Efficient Electroluminescence from Cesium Copper Halides.

Cesium copper halides have the advantages of high photoluminescence quantum efficiency and good stability, making them attractive for replacing toxic lead halides in the field of perovskite light-emitting diodes (LEDs). However, due to their shallow conduction band and the lack of electron transport layers compatible with it, it remains a great challenge to achieve charge balance in LED devices. This drawback manifests as the accumulation of holes at the interface between the emitting layer and electron transport layer, resulting in nonradiative recombination. Here, we demonstrate an effective approach to address this issue by suppressing hole injection, which is realized through modification of the poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) layer with polyethylenimine. This leads to cesium-copper-halide LEDs with a high external quantum efficiency of 5.6%, representing an advance in device architecture for efficient electroluminescence from cesium copper halides.

Adversarial Transformer for Repairing Human Airway Segmentation.

Automated airway segmentation models often suffer from discontinuities in peripheral bronchioles, which limits their clinical applicability. Furthermore, data heterogeneity across different centres and pathological abnormalities pose significant challenges to achieving accurate and robust segmentation in distal small airways. Accurate segmentation of airway structures is essential for the diagnosis and prognosis of lung diseases. To address these issues, we propose a patch-scale adversarial-based refinement network that takes in preliminary segmentation and original CT images and outputs a refined mask of the airway structure. Our method is validated on three datasets, including healthy cases, pulmonary fibrosis, and COVID-19 cases, and quantitatively evaluated using seven metrics. Our method achieves more than a 15% increase in the detected length ratio and detected branch ratio compared to previously proposed models, demonstrating its promising performance. The visual results show that our refinement approach, guided by a patch-scale discriminator and centreline objective functions, effectively detects discontinuities and missing bronchioles. We also demonstrate the generalizability of our refinement pipeline on three previous models, significantly improving their segmentation completeness. Our method provides a robust and accurate airway segmentation tool that can help improve diagnosis and treatment planning for lung diseases.

Fuzzy Attention Neural Network to Tackle Discontinuity in Airway Segmentation.

Airway segmentation is crucial for the examination, diagnosis, and prognosis of lung diseases, while its manual delineation is unduly burdensome. To alleviate this time-consuming and potentially subjective manual procedure, researchers have proposed methods to automatically segment airways from computerized tomography (CT) images. However, some small-sized airway branches (e.g., bronchus and terminal bronchioles) significantly aggravate the difficulty of automatic segmentation by machine learning models. In particular, the variance of voxel values and the severe data imbalance in airway branches make the computational module prone to discontinuous and false-negative predictions, especially for cohorts with different lung diseases. The attention mechanism has shown the capacity to segment complex structures, while fuzzy logic can reduce the uncertainty in feature representations. Therefore, the integration of deep attention networks and fuzzy theory, given by the fuzzy attention layer, should be an escalated solution for better generalization and robustness. This article presents an efficient method for airway segmentation, comprising a novel fuzzy attention neural network (FANN) and a comprehensive loss function to enhance the spatial continuity of airway segmentation. The deep fuzzy set is formulated by a set of voxels in the feature map and a learnable Gaussian membership function. Different from the existing attention mechanism, the proposed channel-specific fuzzy attention addresses the issue of heterogeneous features in different channels. Furthermore, a novel evaluation metric is proposed to assess both the continuity and completeness of airway structures. The efficiency, generalization, and robustness of the proposed method have been proved by training on normal lung disease while testing on datasets of lung cancer, COVID-19, and pulmonary fibrosis.