A transformer-guided cross-modality adaptive feature fusion framework for esophageal gross tumor volume segmentation.

BACKGROUND AND OBJECTIVE: Accurate segmentation of esophageal gross tumor volume (GTV) indirectly enhances the efficacy of radiotherapy for patients with esophagus cancer. In this domain, learning-based methods have been employed to fuse cross-modality positron emission tomography (PET) and computed tomography (CT) images, aiming to improve segmentation accuracy. This fusion is essential as it combines functional metabolic information from PET with anatomical information from CT, providing complementary information. While the existing three-dimensional (3D) segmentation method has achieved state-of-the-art (SOTA) performance, it typically relies on pure-convolution architectures, limiting its ability to capture long-range spatial dependencies due to convolution's confinement to a local receptive field. To address this limitation and further enhance esophageal GTV segmentation performance, this work proposes a transformer-guided cross-modality adaptive feature fusion network, referred to as TransAttPSNN, which is based on cross-modality PET/CT scans. METHODS: Specifically, we establish an attention progressive semantically-nested network (AttPSNN) by incorporating the convolutional attention mechanism into the progressive semantically-nested network (PSNN). Subsequently, we devise a plug-and-play transformer-guided cross-modality adaptive feature fusion model, which is inserted between the multi-scale feature counterparts of a two-stream AttPSNN backbone (one for the PET modality flow and another for the CT modality flow), resulting in the proposed TransAttPSNN architecture. RESULTS: Through extensive four-fold cross-validation experiments on the clinical PET/CT cohort. The proposed approach acquires a Dice similarity coefficient (DSC) of 0.76 ± 0.13, a Hausdorff distance (HD) of 9.38 ± 8.76 mm, and a Mean surface distance (MSD) of 1.13 ± 0.94 mm, outperforming the SOTA competing methods. The qualitative results show a satisfying consistency with the lesion areas. CONCLUSIONS: The devised transformer-guided cross-modality adaptive feature fusion module integrates the strengths of PET and CT, effectively enhancing the segmentation performance of esophageal GTV. The proposed TransAttPSNN has further advanced the research of esophageal GTV segmentation.

ULM-MbCNRT: In vivo Ultrafast Ultrasound Localization Microscopy by Combining Multi-branch CNN and Recursive Transformer.

Ultrasound localization microscopy (ULM) overcomes the acoustic diffraction limit by localizing tiny microbubbles (MBs), thus enabling the microvascular to be rendered at sub-wavelength resolution. Nevertheless, to obtain such superior spatial resolution, it is necessary to spend tens of seconds gathering numerous ultrasound (US) frames to accumulate MB events required, resulting in ULM imaging still suffering from trade-offs between imaging quality, data acquisition time and data processing speed. In this paper, we present a new deep learning (DL) framework combining multi-branch CNN and recursive Transformer, termed as ULM-MbCNRT, that is capable of reconstructing a super-resolution image directly from a temporal mean low-resolution image generated by averaging much fewer raw US frames, i.e., implement an ultrafast ULM imaging. To evaluate the performance of ULM-MbCNRT, a series of numerical simulations and in vivo experiments are carried out. Numerical simulation results indicate that ULM-MbCNRT achieves high-quality ULM imaging with ~10-fold reduction in data acquisition time and ~130-fold reduction in computation time compared to the previous DL method (e.g., the modified sub-pixel convolutional neural network, ULM-mSPCN). For the in vivo experiments, when comparing to the ULM-mSPCN, ULM-MbCNRT allows ~37-fold reduction in data acquisition time (~0.8 s) and ~2134-fold reduction in computation time (~0.87 s) without sacrificing spatial resolution. It implies that ultrafast ULM imaging holds promise for observing rapid biological activity in vivo, potentially improving the diagnosis and monitoring of clinical conditions.

Condition control training-based ConVMLP-ResU-Net for semantic segmentation of esophageal cancer in 18F-FDG PET/CT images.

The precise delineation of esophageal gross tumor volume (GTV) on medical images can promote the radiotherapy effect of esophagus cancer. This work is intended to explore effective learning-based methods to tackle the challenging auto-segmentation problem of esophageal GTV. By employing the progressive hierarchical reasoning mechanism (PHRM), we devised a simple yet effective two-stage deep framework, ConVMLP-ResU-Net. Thereinto, the front-end ConVMLP integrates convolution (ConV) and multi-layer perceptrons (MLP) to capture localized and long-range spatial information, thus making ConVMLP excel in the location and coarse shape prediction of esophageal GTV. According to the PHRM, the front-end ConVMLP should have a strong generalization ability to ensure that the back-end ResU-Net has correct and valid reasoning. Therefore, a condition control training algorithm was proposed to control the training process of ConVMLP for a robust front end. Afterward, the back-end ResU-Net benefits from the yielded mask by ConVMLP to conduct a finer expansive segmentation to output the final result. Extensive experiments were carried out on a clinical cohort, which included 1138 pairs of 18F-FDG positron emission tomography/computed tomography (PET/CT) images. We report the Dice similarity coefficient, Hausdorff distance, and Mean surface distance as 0.82 ± 0.13, 4.31 ± 7.91 mm, and 1.42 ± 3.69 mm, respectively. The predicted contours visually have good agreements with the ground truths. The devised ConVMLP is apt at locating the esophageal GTV with correct initial shape prediction and hence facilitates the finer segmentation of the back-end ResU-Net. Both the qualitative and quantitative results validate the effectiveness of the proposed method.

Epigallocatechin gallate delays age-related cataract development via the RASSF2/AKT pathway.

Age-related cataract (ARC) is a common eye disease, the main cause of which is oxidative stress-mediated apoptosis of lens epithelial cells (LECs). Epigallocatechin gallate (EGCG) is the most potent antioxidant in green tea. Our results demonstrated that EGCG could effectively reduce apoptosis of LECs and retard lens clouding in aged mice. By comparing transcriptome sequencing results of three groups of mice (young control, untreated aged, and EGCG-treated) and screening using GO and KEGG analyses, we selected RASSF2 as the effector gene of EGCG for mechanistic exploration. We verified that the differential expression of RASSF2 was associated with the occurrence of ARC in clinical samples and mouse tissues by immunohistochemistry and western blotting, respectively. We showed that high RASSF2 expression plays a crucial role in the oxidative induction of apoptosis in LECs, as revealed by overexpression and interference experiments. Further studies showed that RASSF2 mediates the inhibitory effect of EGCG on apoptosis and ARCogenesis in LECs by regulating AKT (Ser473) phosphorylation. In this study, we found for the first time the retarding effect of EGCG on lens clouding in mice and revealed the mechanism of action of RASSF2/AKT in it, which provides a theoretical basis for the targeted treatment of EGCG.

Automatic segmentation of esophageal gross tumor volume in 18F-FDG PET/CT images via GloD-LoATUNet.

BACKGROUND AND OBJECTIVE: For esophageal squamous cell carcinoma, radiotherapy is one of the primary treatments. During the planning before radiotherapy, the intractable task is to precisely delineate the esophageal gross tumor volume (GTV) on medical images. In current clinical practice, the manual delineation suffers from high intra- and inter-rater variability, while also exhausting the oncologists on a treadmill. There is an urgent demand for effective computer-aided automatic segmentation methods. To this end, we designed a novel deep network, dubbed as GloD-LoATUNet. METHODS: GloD-LoATUNet follows the effective U-shape structure. On the contractile path, the global deformable dense attention transformer (GloDAT), local attention transformer (LoAT), and convolution blocks are integrated to model long-range dependencies and localized information. On the center bridge and the expanding path, convolution blocks are adopted to upsample the extracted representations for pixel-wise semantic prediction. Between the peer-to-peer counterparts, enhanced skip connections are built to compensate for the lost spatial information and dependencies. By exploiting complementary strengths of the GloDAT, LoAT, and convolution, GloD-LoATUNet has remarkable representation learning capabilities, performing well in the prediction of the small and variable esophageal GTV. RESULTS: The proposed approach was validated in the clinical positron emission tomography/computed tomography (PET/CT) cohort. For 4 different data partitions, we report the Dice similarity coefficient (DSC), Hausdorff distance (HD), and Mean surface distance (MSD) as: 0.83±0.13, 4.88±9.16 mm, and 1.40±4.11 mm; 0.84±0.12, 6.89±12.04 mm, and 1.18±3.02 mm; 0.84±0.13, 3.89±7.64 mm, and 1.28±3.68 mm; 0.86±0.09, 3.71±4.79 mm, and 0.90±0.37 mm; respectively. The predicted contours present a desirable consistency with the ground truth. CONCLUSIONS: The inspiring results confirm the accuracy and generalizability of the proposed model, demonstrating the potential for automatic segmentation of esophageal GTV in clinical practice.

Automatic Seizure Identification from EEG Signals Based on Brain Connectivity Learning.

Epilepsy is a neurological disorder caused by brain dysfunction, which could cause uncontrolled behavior, loss of consciousness and other hazards. Electroencephalography (EEG) is an indispensable auxiliary tool for clinical diagnosis. Great progress has been made by current seizure identification methods. However, the performance of the methods on different patients varies a lot. In order to deal with this problem, we propose an automatic seizure identification method based on brain connectivity learning. The connectivity of different brain regions is modeled by a graph. Different from the manually defined graph structure, our method can extract the optimal graph structure and EEG features in an end-to-end manner. Combined with the popular graph attention neural network (GAT), this method achieves high performance and stability on different patients from the CHB-MIT dataset. The average values of accuracy, sensitivity, specificity, F1-score and AUC of the proposed model are 98.90%, 98.33%, 98.48%, 97.72% and 98.54%, respectively. The standard deviations of the above five indicators are 0.0049, 0.0125, 0.0116 and 0.0094, respectively. Compared with the existing seizure identification methods, the stability of the proposed model is improved by 78-95%.

Multi-view cross-subject seizure detection with information bottleneck attribution.

Objective.Significant progress has been witnessed in within-subject seizure detection from electroencephalography (EEG) signals. Consequently, more and more works have been shifted from within-subject seizure detection to cross-subject scenarios. However, the progress is hindered by inter-patient variations caused by gender, seizure type, etc.Approach.To tackle this problem, we propose a multi-view cross-object seizure detection model with information bottleneck attribution (IBA).Significance.Feature representations specific to seizures are learned from raw EEG data by adversarial deep learning. Combined with the manually designed discriminative features, the model can detect seizures across different subjects. In addition, we introduce IBA to provide insights into the decision-making of the adversarial learning process, thus enhancing the interpretability of the model.Main results.Extensive experiments are conducted on two benchmark datasets. The experimental results verify the efficacy of the model.

High MST2 expression regulates lens epithelial cell apoptosis in age-related cataracts through YAP1 targeting GLUT1.

Age-related cataract (ARC) is a severe visual impairment disease and its pathogenesis remains unclear. This study investigated the relevance of MST2/YAP1/GLUT1 in ARC development in vivo and in vitro, and explored the role and possible mechanisms of this pathway in oxidative damage-mediated apoptosis of lens epithelial cells (LECs). Western blot analysis and immunohistochemistry showed that MST2 and phosphorylated (p)-YAP (Ser127) protein levels were increased, and YAP1 and GLUT1 protein levels were decreased in LECs of ARC patients and aged mice. Additionally, differential expression of MST2 and YAP1 was associated with H2O2-induced apoptosis of human lens epithelial B3 (HLE-B3) cells. CCK-8 and Hoechst 33,342 apoptosis assays showed that MST2 and YAP1 were involved in H2O2-induced apoptosis of LECs. Subsequent experiments showed that, during MST2-mediated H2O2-induced apoptosis, p-YAP (Ser127) levels were elevated and immunofluorescence revealed nucleoplasmic translocation and inhibition of YAP1 protein expression. Furthermore, GLUT1 was in turn synergistically transcriptionally regulated by YAP1-TEAD1 in dual luciferase reporter assays. In conclusion, our study indicates that the MST2/YAP1/GLUT1 pathway plays a major role in the pathogenesis of ARC and LECs apoptosis, providing a new direction for future development of targeted inhibitors that block this signaling pathway to prevent, delay, or even cure ARC.

EEG-Based Seizure detection using linear graph convolution network with focal loss.

BACKGROUND AND OBJECTIVES: Epilepsy is a clinical phenomenon caused by sudden abnormal and excessive discharge of brain neurons. It affects around 70 million people all over the world. Seizure detection from Electroencephalography (EEG) has achieved rapid development. However, existing methods often extract features from single channel EEG while ignoring the spatial relationship between different EEG channels. To fill this gap, a novel seizure detection model based on linear graph convolution network (LGCN) was proposed to enhance the feature embedding of raw EEG signals during seizure and non-seizure periods. METHOD: Pearson correlation matrix of raw EEG signals was calculated to build the input graph of the graph neural network where the coefficients of the matrix models the spatial relations in EEG signals. The last softmax layer makes the final decision (seizure vs. non-seizure). In addition, focal loss was utilized to redefine the loss function of LGCN to deal with the data imbalance problem during seizure detection. RESULTS: Experiments are conducted on the CHB-MIT dataset. The seizure detection accuracy, specificity, sensitivity, F1 and Auc are 99.30%, 98.82%, 99.43%, 98.73% and 98.57% respectively. CONCLUSIONS: The proposed approach yields superior performance over the-state-of-the-art in seizure detection tasks on the CHB-MIT dataset. Our method works in an end-to-end manner and it does not need manually designed features. The ability to deal with imbalanced data is also attractive in real seizure detection scenarios where the duration of seizures is much shorter than the lasting time of non-seizure events.

Graph Attention Network with Focal Loss for Seizure Detection on Electroencephalography Signals.

Automatic seizure detection from electroencephalogram (EEG) plays a vital role in accelerating epilepsy diagnosis. Previous researches on seizure detection mainly focused on extracting time-domain and frequency-domain features from single electrodes, while paying little attention to the positional correlations between different EEG channels of the same subject. Moreover, data imbalance is common in seizure detection scenarios where the duration of nonseizure periods is much longer than the duration of seizures. To cope with the two challenges, a novel seizure detection method based on graph attention network (GAT) is presented. The approach acts on graph-structured data and takes the raw EEG data as input. The positional relationship between different EEG signals is exploited by GAT. The loss function of the GAT model is redefined using the focal loss to tackle data imbalance problem. Experiments are conducted on the CHB-MIT dataset. The accuracy, sensitivity and specificity of the proposed method are 98.89[Formula: see text], 97.10[Formula: see text] and 99.63[Formula: see text], respectively.

RNA sequencing and bioinformatics analysis of human lens epithelial cells in age-related cataract.

BACKGROUND: Age-related cataract (ARC) is the main cause of blindness in older individuals but its specific pathogenic mechanism is unclear. This study aimed to identify differentially expressed genes (DEGs) associated with ARC and to improve our understanding of the disease mechanism. METHODS: Anterior capsule samples of the human lens were collected from ARC patients and healthy controls and used for RNA sequencing to detect DEGs. Identified DEGs underwent bioinformatics analyses, including Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses. Subsequently, reverse transcription quantitative RT-qPCR was used to validate the different expression levels of selected genes. RESULTS: A total of 698 up-regulated DEGs and 414 down-regulated DEGs were identified in ARC patients compared with controls by transcriptome analysis. Through GO and KEGG bioinformatics analysis, the functions of significantly DEGs and their possible molecular mechanisms were determined. Sequencing results were verified by RT-qPCR as being accurate and reliable. CONCLUSIONS: This study identified several genes associated with ARC, which improves our knowledge of the disease mechanism.

Amino acid substitutions in VP2, VP1, and 2C attenuate a Coxsackievirus A16 in mice.

Coxsackievirus A16 (CVA16) is one of the major etiological agents of hand, foot and mouth disease (HFMD), a common acute infectious disease affecting infants and young children. Severe symptoms of the central nervous system may develop and even lead to death. Here, a plaque-purified CVA16 strain, L731-P1 (P1), was serially passaged in Vero cells for six times and passage 6 (P6) stock became highly attenuated in newborn mice. Genomic sequencing of the P1 and P6 revealed seven nucleotide substitutions at positions 1434 (C to U), 2744 (A to G), 2747 (A to G), 3161 (G to A), 3182 (A to G), 4968 (C to U), and 6064 (C to U). Six of these substitutions resulted in amino acid changes at VP2-T161 M, VP1-N102D, VP1-T103A, VP1-E241K, VP1-T248A, and 2C-S297F, respectively. P1-based infectious cDNA was generated to further investigate these virulent determinants. Independent reverse transcription-polymerase chain reaction (RT-PCR) amplifications for mutant constructions and plaque-purification of the P6 for isolation of variants were performed to determine dominant mutations and strains more related to attenuation. The virulent P1, attenuated P6, as well as a plaque purified strain (PP) and other four recombinant mutants, were inoculated into one-day-old BALB/c mice and the 50% lethal dose of each strain was determined. Comparison of virulence among these strains indicated that amino acid changes of VP1-N102D, VP1-E241K and 2C-S297F might be associated more closely with a high level attenuation of CVA16-L731-P6 than other mutations. Identification of novel residues associated with virulence may contribute to understanding of molecular basis of virulence of CVA16 and other enteroviruses.

Increased Excess Intracellular Cyclic di-AMP Levels Impair Growth and Virulence of Bacillus anthracis.

Cyclic di-AMP (c-di-AMP) is a recently identified bacterial second messenger that regulates biological processes. In this study, we found that inactivation of two c-di-AMP phosphodiesterases (PDEs), GdpP and PgpH, resulted in accumulation of 3.8-fold higher c-di-AMP levels than in the parental strain Sterne in Bacillus anthracis and inhibited bacterial growth. Moreover, excess c-di-AMP accumulation decreased bacterial toxin expression, increased sensitivity to osmotic stress and detergent, and attenuated virulence in both C57BL/6J and A/J mice. Complementation of the PDE mutant with a plasmid carrying gdpP or pgpH in trans from a Pspac promoter restored bacterial growth, virulence factor expression, and resistance to detergent. Our results indicate that c-di-AMP is a pleiotropic signaling molecule in B. anthracis that is important for host-pathogen interaction.IMPORTANCE Anthrax is an ancient and deadly disease caused by the spore-forming bacterial pathogen Bacillus anthracis Vegetative cells of this species produce anthrax toxin proteins and S-layer components during infection of mammalian hosts. So far, how the expression of these virulence factors is regulated remains largely unknown. Our results suggest that excess elevated c-di-AMP levels inhibit bacterial growth and reduce expression of S-layer components and anthracis toxins as well as reduce virulence in a mouse model of disease. These results indicate that c-di-AMP signaling plays crucial roles in B. anthracis biology and disease.

The surface-exposed loop region of norovirus GII.3 VP1 plays an essential role in binding histo-blood group antigens.

Trypsin digestion promotes disassembly of GII.3 NoV virus-like particles (VLPs) and binding of VLPs to salivary histo-blood group antigens (HBGAs), but it is not clear which specific regions or residues mediate viral attachment to HBGAs. An earlier study indicated that arginine residues in the predicted surface-exposed loop region are susceptible to trypsin digestion. Here, we introduced single or multiple substitutions of four arginine residues located in the predicted surface-exposed loop region of the GII.3 NoV capsid protein (VP1) and observed their effects on susceptibility to trypsin digestion and binding to HBGAs. All of the mutations in VP1, including single substitutions (R287G, R292G, R296G or R307G) and quadruple substitutions (R287G, R292G, R296G and R307G), permitted successful VLP assembly. After tryptic digestion, all VP1 proteins bearing single point mutations were cleaved, resulting in complete digestion or single fragments with various molecular sizes (27-35 kDa), while the VP1 protein bearing four substitutions was cleaved into two fragments (27-55 kDa). Binding assays using synthetic and salivary HBGAs showed that none of the VP1 mutants (singly or quadruply substituted) exhibited detectable binding to HBGA before or after trypsin cleavage. These results indicated that arginine residues within the predicted surface loop region of GII.3 NoV VP1 were involved directly or indirectly in binding salivary HBGAs and could potentially mediate the HBGA-GII.3 NoV interactions through which host cells become infected.

A Reverse Genetics System for Cypovirus Based on a Bacmid Expressing T7 RNA Polymerase.

Dendrolimus punctatus cypovirus (DpCPV), belonging to the genus Cypovirus within the family Reoviridae, is considered the most destructive pest of pine forests worldwide. DpCPV has a genome consisting of 10 linear double-stranded RNA segments. To establish a reverse genetics system, we cloned cDNAs encoding the 10 genomic segments of DpCPV into three reverse genetics vectors in which each segment was transcribed under the control of a T7 RNA polymerase promoter and terminator tagged with a hepatitis delta virus ribozyme sequence. We also constructed a vp80-knockout Autographa californica multiple nucleopolyhedrovirus bacmid to express a T7 RNA polymerase codon-optimized for Sf9 cells. Following transfection of Sf9 cells with the three vectors and the bacmid, occlusion bodies (OBs) with the typical morphology of cypovirus polyhedra were observed by optical microscopy. The rescue system was verified by incorporation of a HindIII restriction enzyme site null mutant of the 9th genomic segment. Furthermore, when we co-transfected Sf9 cells with the reverse genetics vectors, the bacmid, and an additional vector bearing an egfp gene flanked with the 5' and 3' untranslated regions of the 10th genomic segment, aggregated green fluorescence co-localizing with the OBs was observed. The rescued OBs were able to infect Spodopetra exigua larvae, although their infectivity was significantly lower than that of wild-type DpCPV. This reverse genetics system for DpCPV could be used to explore viral replication and pathogenesis and to facilitate the development of novel bio-insecticides and expression systems for exogenous proteins.