Residual learning-based robotic image analysis model for low-voltage distributed photovoltaic fault identification and positioning.

With the fast development of large-scale Photovoltaic (PV) plants, the automatic PV fault identification and positioning have become an important task for the PV intelligent systems, aiming to guarantee the safety, reliability, and productivity of large-scale PV plants. In this paper, we propose a residual learning-based robotic (UAV) image analysis model for low-voltage distributed PV fault identification and positioning. In our target scenario, the unmanned aerial vehicles (UAVs) are deployed to acquire moving images of low-voltage distributed PV power plants. To get desired robustness and accuracy of PV image detection, we integrate residual learning with attention mechanism into the UAV image analysis model based on you only look once v4 (YOLOv4) network. Then, we design the sophisticated multi-scale spatial pyramid fusion and use it to optimize the YOLOv4 network for the nuanced task of fault localization within PV arrays, where the Complete-IOU loss is incorporated in the predictive modeling phase, significantly enhancing the accuracy and efficiency of fault detection. A series of experimental comparisons in terms of the accuracy of fault positioning are conducted, and the experimental results verify the feasibility and effectiveness of the proposed model in dealing with the safety and reliability maintenance of low-voltage distributed PV systems.

Multiscale Anion-Hybrid in Atomic Ni Sites for High-Rate Water Electrolysis: Insights into the Charge Accumulation Mechanism.

Single-atom catalysts, characterized by transition metal-(N/O)4 units on nanocarbon (M-(N/O)4-C), have emerged as efficient performers in water electrolysis. However, there are few guiding principles for accurately controlling the ligand fields of single atoms to further stimulate the catalyst activities. Herein, using the Ni-(N/O)4-C unit as a model, we develop a further modification of the P anion on the outer shells to modulate the morphology of the ligand. The catalyst thus prepared possesses high activity and excellent long-term durability, surpassing commercial Pt/C, RuO2, and currently reported single-atom catalysts. Notably, mechanistic studies demonstrated that the pseudocapacitive feature of multiscale anion-hybrid nanocarbon is considerable at accumulating enough positive charge [Q], contributing to the high oxygen evolution reaction (OER) order (ß) through the rate formula. DFT calculations also indicate that the catalytic activity is decided by the suitable barrier energy of the intermediates due to charge accumulation. This work reveals the activity origin of single atoms on multihybrid nanocarbon, providing a clear experiential formula for designing the electronic configuration of single-atom catalysts to boost electrocatalytic performance.

Interpretation of SPECT wall motion with deep learning.

OBJECTIVES: We sought to develop a novel deep learning (DL) workflow to interpret single-photon emission computed tomography (SPECT) wall motion. BACKGROUND: Wall motion assessment with SPECT is limited by image temporal and spatial resolution. Visual interpretation of wall motion can be subjective and prone to error. Artificial intelligence (AI) may improve accuracy of wall motion assessment. METHODS: A total of 1038 patients undergoing rest electrocardiogram (ECG)-gated SPECT and echocardiography were included. Using echocardiography as truth, a DL-model (DL-model 1) was trained to predict the probability of abnormal wall motion. Of the 1038 patients, 317 were used to train a DL-model (DL-model 2) to assess regional wall motion. A 10-fold cross-validation was adopted. Diagnostic performance of DL was compared with human readers and quantitative parameters. RESULTS: The area under the receiver operating characteristic curve (AUC) and accuracy (ACC) of DL model (AUC: .82 [95% CI: .79-.85]; ACC: .88) were higher than human (AUC: .77 [95% CI: .73-.81]; ACC: .82; P < .001) and quantitative parameter (AUC: .74 [95% CI: .66-.81]; ACC: .78; P < .05). The net reclassification index (NRI) was 7.7%. The AUC and accuracy of DL model for per-segment and per-vessel territory diagnosis were also higher than human reader. The DL model generated results within 30 seconds with operable guided user interface (GUI) and therefore could provide preliminary interpretation. CONCLUSIONS: DL can be used to improve interpretation of rest SPECT wall motion as compared with current human readers and quantitative parameter diagnosis.

Current potential therapeutics of amyotrophic lateral sclerosis.

Amyotrophic lateral sclerosis (ALS) is a debilitating motor neurological disorder for which there is still no cure. The disease seriously jeopardizes the health and lifespan of adult populations. The authors extensively retrieved the current literature about clinical and experimental ALS treatments. Based on them, this review primarily focused on summarizing the current potential clinical usage and trialing therapeutics of ALS. Currently, the clinical ALS treatments have focused primarily on relieving symptoms to improve the quality of life yet. There are a number of therapeutic approaches such as medicine, gene therapy, neuron protectants, combination therapy and stem cells. Among them, Stem cells including embryonic stem cells, mesenchymal stem cells, neural stem cells, and many other types of stem cells have been used in ALS treatment, and although the short-term efficacy is good, it is worth exploring whether this improved efficacy leads to prolonged patient survival. In addition, the supportive treatments also exert an important effect on improving the quality of life and prolong the survival of ALS patients in absence of effectively care for stopping or reversing the progression of ALS.

Associating Knee Osteoarthritis Progression with Temporal-Regional Graph Convolutional Network Analysis on MR Images.

BACKGROUND: Artificial intelligence shows promise in assessing knee osteoarthritis (OA) progression on MR images, but faces challenges in accuracy and interpretability. PURPOSE: To introduce a temporal-regional graph convolutional network (TRGCN) on MR images to study the association between knee OA progression status and network outcome. STUDY TYPE: Retrospective. POPULATION: 194 OA progressors (mean age, 62 ± 9 years) and 406 controls (mean age, 61 ± 9 years) from the OA Initiative were randomly divided into training (80%) and testing (20%) cohorts. FIELD STRENGTH/SEQUENCE: Sagittal 2D IW-TSE-FS (IW) and 3D-DESS-WE (DESS) at 3T. ASSESSMENT: Anatomical subregions of cartilage, subchondral bone, meniscus, and the infrapatellar fat pad at baseline, 12-month, and 24-month were automatically segmented and served as inputs to form compartment-based graphs for a TRGCN model, which containing both regional and temporal information. The performance of models based on (i) clinical variables alone, (ii) radiologist score alone, (iii) combined features (containing i and ii), (iv) composite TRGCN (combining TRGCN, i and ii), (v) radiomics features, (vi) convolutional neural network based on Densenet-169 were compared. STATISTICAL TESTS: DeLong test was performed to compare the areas under the ROC curve (AUC) of all models. Additionally, interpretability analysis was done to evaluate the contributions of individual regions. A P value <0.05 was considered significant. RESULTS: The composite TRGCN outperformed all other models with AUCs of 0.841 (DESS) and 0.856 (IW) in the testing cohort (all P < 0.05). Interpretability analysis highlighted cartilage's importance over other structures (42%-45%), tibiofemoral joint's (TFJ) dominance over patellofemoral joint (PFJ) (58%-67% vs. 12%-37%), and importance scores changes in compartments over time (TFJ vs. PFJ: baseline: 44% vs. 43%, 12-month: 52% vs. 39%, 24-month: 31% vs. 48%). DATA CONCLUSION: The composite TRGCN, capturing temporal and regional information, demonstrated superior discriminative ability compared with other methods, providing interpretable insights for identifying knee OA progression. TECHNICAL EFFICACY: Stage 2.

BAF-Net: bidirectional attention-aware fluid pyramid feature integrated multimodal fusion network for diagnosis and prognosis.

Objective. To go beyond the deficiencies of the three conventional multimodal fusion strategies (i.e. input-, feature- and output-level fusion), we propose a bidirectional attention-aware fluid pyramid feature integrated fusion network (BAF-Net) with cross-modal interactions for multimodal medical image diagnosis and prognosis.Approach. BAF-Net is composed of two identical branches to preserve the unimodal features and one bidirectional attention-aware distillation stream to progressively assimilate cross-modal complements and to learn supplementary features in both bottom-up and top-down processes. Fluid pyramid connections were adopted to integrate the hierarchical features at different levels of the network, and channel-wise attention modules were exploited to mitigate cross-modal cross-level incompatibility. Furthermore, depth-wise separable convolution was introduced to fuse the cross-modal cross-level features to alleviate the increase in parameters to a great extent. The generalization abilities of BAF-Net were evaluated in terms of two clinical tasks: (1) an in-house PET-CT dataset with 174 patients for differentiation between lung cancer and pulmonary tuberculosis. (2) A public multicenter PET-CT head and neck cancer dataset with 800 patients from nine centers for overall survival prediction.Main results. On the LC-PTB dataset, improved performance was found in BAF-Net (AUC = 0.7342) compared with input-level fusion model (AUC = 0.6825;p< 0.05), feature-level fusion model (AUC = 0.6968;p= 0.0547), output-level fusion model (AUC = 0.7011;p< 0.05). On the H&N cancer dataset, BAF-Net (C-index = 0.7241) outperformed the input-, feature-, and output-level fusion model, with 2.95%, 3.77%, and 1.52% increments of C-index (p= 0.3336, 0.0479 and 0.2911, respectively). The ablation experiments demonstrated the effectiveness of all the designed modules regarding all the evaluated metrics in both datasets.Significance. Extensive experiments on two datasets demonstrated better performance and robustness of BAF-Net than three conventional fusion strategies and PET or CT unimodal network in terms of diagnosis and prognosis.

Semi-supervised model based on implicit neural representation and mutual learning (SIMN) for multi-center nasopharyngeal carcinoma segmentation on MRI.

BACKGROUND: The issue of using deep learning to obtain accurate gross tumor volume (GTV) and metastatic lymph nodes (MLN) segmentation for nasopharyngeal carcinoma (NPC) on heterogeneous magnetic resonance imaging (MRI) images with limited labeling remains unsolved. METHOD: We collected 918 patients with MRI images from three hospitals to develop and validate models and proposed a semi-supervised framework for the fine delineation of multi-center NPC boundaries by integrating uncertainty-based implicit neural representations named SIMN. The framework utilizes the deep mutual learning approach with CNN and Transformer, incorporating dynamic thresholds. Additionally, domain adaptive algorithms are employed to enhance the performance. RESULTS: SIMN predictions have a high overlap ratio with the ground truth. Under the 20 % labeled cases, for the internal test cohorts, the average DSC in GTV and MLN are 0.7981 and 0.7804, respectively; for external test cohort Wu Zhou Red Cross Hospital, the average DSC in GTV and MLN are 0.7217 and 0.7581, respectively; for external test cohorts First People Hospital of Foshan, the average DSC in GTV and MLN are 0.7004 and 0.7692, respectively. No significant differences are found in DSC, HD95, ASD, and Recall for patients with different clinical categories. Moreover, SIMN outperformed existing classical semi-supervised methods. CONCLUSIONS: SIMN showed a highly accurate GTV and MLN segmentation for NPC on multi-center MRI images under Semi-Supervised Learning (SSL), which can easily transfer to other centers without fine-tuning. It suggests that it has the potential to act as a generalized delineation solution for heterogeneous MRI images with limited labels in clinical deployment.

5G NB-IoT System Integrated with High-Performance Fiber Sensor Inspired by Cirrus and Spider Structures.

Real-time telemedicine detection can solve the problem of the shortage of public medical resources caused by the coming aging society. However, the development of such an integrated monitoring system is hampered by the need for high-performance sensors and the strict-requirement of long-distance signal transmission and reproduction. Here, a bionic crack-spring fiber sensor (CSFS) inspired by spider leg and cirrus whiskers for stretchable and weavable electronics is reported. Trans-scale conductive percolation networks of multilayer graphene around the surface of outer spring-like Polyethylene terephthalate (PET) fibers and printing Ag enable a high sensitivity of 28475.6 and broad sensing range over 250%. The electromechanical changes in different stretching stages are simulated by Comsol to explain the response mechanism. The CSFS is incorporated into the fabric and realized the human-machine interactions (HMIs) for robot control. Furthermore, the 5G Narrowband Internet of Things (NB-IoT) system is developed for human healthcare data collection, transmission, and reproduction together with the integration of the CSFS, illustrating the huge potential of the approach in human-machine communication interfaces and intelligent telemedicine rehabilitation and diagnosis monitoring.

Prognostic Value of a Combined Nomogram Model Integrating 3-Dimensional Deep Learning and Radiomics for Head and Neck Cancer.

OBJECTIVE: The preoperative prediction of the overall survival (OS) status of patients with head and neck cancer (HNC) is significant value for their individualized treatment and prognosis. This study aims to evaluate the impact of adding 3D deep learning features to radiomics models for predicting 5-year OS status. METHODS: Two hundred twenty cases from The Cancer Imaging Archive public dataset were included in this study; 2212 radiomics features and 304 deep features were extracted from each case. The features were selected by univariate analysis and the least absolute shrinkage and selection operator, and then grouped into a radiomics model containing Positron Emission Tomography /Computed Tomography (PET/CT) radiomics features score, a deep model containing deep features score, and a combined model containing PET/CT radiomics features score +3D deep features score. TumorStage model was also constructed using initial patient tumor node metastasis stage to compare the performance of the combined model. A nomogram was constructed to analyze the influence of deep features on the performance of the model. The 10-fold cross-validation of the average area under the receiver operating characteristic curve and calibration curve were used to evaluate performance, and Shapley Additive exPlanations (SHAP) was developed for interpretation. RESULTS: The TumorStage model, radiomics model, deep model, and the combined model achieved areas under the receiver operating characteristic curve of 0.604, 0.851, 0.840, and 0.895 on the train set and 0.571, 0.849, 0.832, and 0.900 on the test set. The combined model showed better performance of predicting the 5-year OS status of HNC patients than the radiomics model and deep model. The combined model was shown to provide a favorable fit in calibration curves and be clinically useful in decision curve analysis. SHAP summary plot and SHAP The SHAP summary plot and SHAP force plot visually interpreted the influence of deep features and radiomics features on the model results. CONCLUSIONS: In predicting 5-year OS status in patients with HNC, 3D deep features could provide richer features for combined model, which showed outperformance compared with the radiomics model and deep model.

Efficacy and safety of hyperbaric oxygen therapy in acute ischaemic stroke: a systematic review and meta-analysis.

OBJECTIVE: This study aims to evaluate the efficacy and safety of adjunctive hyperbaric oxygen therapy (HBOT) in acute ischaemic stroke (AIS) based on existing evidence. METHODS: We conducted a comprehensive search through April 15, 2023, of seven major databases for randomized controlled trials (RCTs) comparing adjunctive hyperbaric HBOT with non-HBOT (no HBOT or sham HBOT) treatments for AIS. Data extraction and assessment were independently performed by two researchers. The quality of included studies was evaluated using the tool provided by the Cochrane Collaboration. Meta-analysis was conducted using Rev Man 5.3. RESULTS: A total of 8 studies involving 493 patients were included. The meta-analysis showed no statistically significant differences between HBOT and the control group in terms of NIHSS score (MD = -1.41, 95%CI = -7.41 to 4.58), Barthel index (MD = 8.85, 95%CI = -5.84 to 23.54), TNF-α (MD = -5.78, 95%CI = -19.93 to 8.36), sICAM (MD = -308.47, 95%CI = -844.13 to 13227.19), sVCAM (MD = -122.84, 95%CI = -728.26 to 482.58), sE-selectin (MD = 0.11, 95%CI = -21.86 to 22.08), CRP (MD = -5.76, 95%CI = -15.02 to 3.51), adverse event incidence within ≤ 6 months of follow-up (OR = 0.98, 95%CI = 0.25 to 3.79). However, HBOT showed significant improvement in modified Rankin score (MD = 0.10, 95%CI = 0.03 to 0.17), and adverse event incidence at the end of treatment (OR = 0.42, 95%CI = 0.19 to 0.94) compared to the control group. CONCLUSION: While our findings do not support the routine use of HBOT for improving clinical outcomes in AIS, further research is needed to explore its potential efficacy within specific therapeutic windows and for different cerebral occlusion scenarios. Therefore, the possibility of HBOT offering clinical benefits for AIS cannot be entirely ruled out.

Transcriptional profiling of Bulinus globosus provides insights into immune gene families in snails supporting the transmission of urogenital schistosomiasis.

Schistosomiasis, urogenital and intestinal, afflicts 251 million people worldwide with approximately two-thirds of the patients suffering from the urogenital form of the disease. Freshwater snails of the genus Bulinus (Gastropoda: Planorbidae) serve as obligate intermediate hosts for Schistosoma haematobium, the etiologic agent of human urogenital schistosomiasis. These snails also act as vectors for the transmission of schistosomiasis in livestock and wildlife. Despite their crucial role in human and veterinary medicine, our basic understanding at the molecular level of the entire Bulinus genus, which comprises 37 recognized species, is very limited. In this study, we employed Illumina-based RNA sequencing (RNAseq) to profile the genome-wide transcriptome of Bulinus globosus, one of the most important intermediate hosts for S. haematobium in Africa. A total of 179,221 transcripts (N50 = 1,235) were assembled and the benchmarking universal single-copy orthologs (BUSCO) was estimated to be 97.7%. The analysis revealed a substantial number of transcripts encoding evolutionarily conserved immune-related proteins, particularly C-type lectin (CLECT) domain-containing proteins (n = 316), Toll/Interleukin 1-receptor (TIR)-containing proteins (n = 75), and fibrinogen related domain-containing molecules (FReD) (n = 165). Notably, none of the FReDs are fibrinogen-related proteins (FREPs) (immunoglobulin superfamily (IgSF) + fibrinogen (FBG)). This RNAseq-based transcriptional profile provides new insights into immune capabilities of Bulinus snails, helps provide a framework to explain the complex patterns of compatibility between snails and schistosomes, and improves our overall understanding of comparative immunology.

An Innovative and Efficient Diagnostic Prediction Flow for Head and Neck Cancer: A Deep Learning Approach for Multi-Modal Survival Analysis Prediction Based on Text and Multi-Center PET/CT Images.

Objective: To comprehensively capture intra-tumor heterogeneity in head and neck cancer (HNC) and maximize the use of valid information collected in the clinical field, we propose a novel multi-modal image-text fusion strategy aimed at improving prognosis. Method: We have developed a tailored diagnostic algorithm for HNC, leveraging a deep learning-based model that integrates both image and clinical text information. For the image fusion part, we used the cross-attention mechanism to fuse the image information between PET and CT, and for the fusion of text and image, we used the Q-former architecture to fuse the text and image information. We also improved the traditional prognostic model by introducing time as a variable in the construction of the model, and finally obtained the corresponding prognostic results. Result: We assessed the efficacy of our methodology through the compilation of a multicenter dataset, achieving commendable outcomes in multicenter validations. Notably, our results for metastasis-free survival (MFS), recurrence-free survival (RFS), overall survival (OS), and progression-free survival (PFS) were as follows: 0.796, 0.626, 0.641, and 0.691. Our results demonstrate a notable superiority over the utilization of CT and PET independently, and exceed the result derived without the clinical textual information. Conclusions: Our model not only validates the effectiveness of multi-modal fusion in aiding diagnosis, but also provides insights for optimizing survival analysis. The study underscores the potential of our approach in enhancing prognosis and contributing to the advancement of personalized medicine in HNC.

The genome and transcriptome of the snail Biomphalaria sudanica s.l.: immune gene diversification and highly polymorphic genomic regions in an important African vector of Schistosoma mansoni.

BACKGROUND: Control and elimination of schistosomiasis is an arduous task, with current strategies proving inadequate to break transmission. Exploration of genetic approaches to interrupt Schistosoma mansoni transmission, the causative agent for human intestinal schistosomiasis in sub-Saharan Africa and South America, has led to genomic research of the snail vector hosts of the genus Biomphalaria. Few complete genomic resources exist, with African Biomphalaria species being particularly underrepresented despite this being where the majority of S. mansoni infections occur. Here we generate and annotate the first genome assembly of Biomphalaria sudanica sensu lato, a species responsible for S. mansoni transmission in lake and marsh habitats of the African Rift Valley. Supported by whole-genome diversity data among five inbred lines, we describe orthologs of immune-relevant gene regions in the South American vector B. glabrata and present a bioinformatic pipeline to identify candidate novel pathogen recognition receptors (PRRs). RESULTS: De novo genome and transcriptome assembly of inbred B. sudanica originating from the shoreline of Lake Victoria (Kisumu, Kenya) resulted in a haploid genome size of ~ 944.2 Mb (6,728 fragments, N50 = 1.067 Mb), comprising 23,598 genes (BUSCO = 93.6% complete). The B. sudanica genome contains orthologues to all described immune genes/regions tied to protection against S. mansoni in B. glabrata, including the polymorphic transmembrane clusters (PTC1 and PTC2), RADres, and other loci. The B. sudanica PTC2 candidate immune genomic region contained many PRR-like genes across a much wider genomic region than has been shown in B. glabrata, as well as a large inversion between species. High levels of intra-species nucleotide diversity were seen in PTC2, as well as in regions linked to PTC1 and RADres orthologues. Immune related and putative PRR gene families were significantly over-represented in the sub-set of B. sudanica genes determined as hyperdiverse, including high extracellular diversity in transmembrane genes, which could be under pathogen-mediated balancing selection. However, no overall expansion in immunity related genes was seen in African compared to South American lineages. CONCLUSIONS: The B. sudanica genome and analyses presented here will facilitate future research in vector immune defense mechanisms against pathogens. This genomic/transcriptomic resource provides necessary data for the future development of molecular snail vector control/surveillance tools, facilitating schistosome transmission interruption mechanisms in Africa.

A haplotype-like, chromosome-level assembled and annotated genome of Biomphalaria glabrata, an important intermediate host of schistosomiasis and the best studied model of schistosomiasis vector snails.

Schistosomiasis is one of the world's most devastating parasitic diseases, afflicting 251 million people globally. The Neotropical snail Biomphalaria glabrata is an important intermediate host of the human blood fluke Schistosoma mansoni and a predominant model for schistosomiasis research. To fully exploit this model snail for biomedical research, here we report a haplotype-like, chromosome-level assembled and annotated genome of the homozygous iM line of B. glabrata that we developed at the University of New Mexico. Using multiple sequencing platforms, including Illumina, PacBio, and Omni-C sequencing, 18 sequence contact matrices representing 18 haploid chromosomes (2n = 36) were generated (337x genome coverage), and 96.5% of the scaffold sequences were anchored to the 18 chromosomes. Protein-coding genes (n = 34,559), non-coding RNAs (n = 2,406), and repetitive elements (42.52% of the genome) were predicted for the whole genome, and detailed annotations for individual chromosomes were also provided. Using this genomic resource, we have investigated the genomic structure and organization of the Toll-like receptor (TLR) and fibrinogen-domain containing protein (FReD) genes, the two important immune-related gene families. Notably, TLR-like genes are scattered on 13 chromosomes. In contrast, almost all (39 of 40) fibrinogen-related genes (FREPs) (immunoglobulin superfamily (IgSF) + fibrinogen (FBG)) are clustered within a 5-million nucleotide region on chromosome 13, yielding insight into mechanisms involved in the diversification of FREPs. This is the first genome of schistosomiasis vector snails that has been assembled at the chromosome level, annotated, and analyzed. It serves as a valuable resource for a deeper understanding of the biology of vector snails, especially Biomphalaria snails.

Association between dietary patterns and cardiovascular diseases: A review.

Cardiovascular disease (CVD), especially atherosclerosis, is the primary cause of global deaths. It accounts for millions of deaths annually. Even a small reduction in CVD through preventive treatment can have a substantial impact. Dietary patterns and substances are strongly linked to chronic diseases such as atherosclerosis, hypertension, heart failure, and type 2 diabetes. An unhealthy diet could lead to traditional risk factors such as LDL levels, TG levels, diabetes, and high blood pressure while accelerating atherosclerosis progression. Recent research has shown the potential of dietary interventions to prevent and treat cardiovascular disease, particularly through healthy dietary patterns such as the Mediterranean diet or DASH. In 2016, the World Health Organization (WHO) and the US Centers for Disease Control and Prevention (CDC) launched a new initiative aimed at enhancing the prevention and control of cardiovascular disease (CVD) by improving the management of CVD in primary care, including the optimization of dietary patterns. Here, this review summarizes several large cohort researches about the effects of dietary patterns on atherosclerosis, refines dietary components, and outlines some typical anti-atherosclerosis dietary agents. Finally, this review discusses recent mechanisms by which dietary interventions affect atherosclerosis progression.

Stress-only versus rest-stress SPECT MPI in the detection and diagnosis of myocardial ischemia and infarction by machine learning.

BACKGROUND: Rest-stress SPECT myocardial perfusion imaging (MPI) is widely used to evaluate coronary artery disease (CAD). We aim to evaluate stress-only versus rest-stress MPI in diagnosing CAD by machine learning (ML). METHODS: A total of 276 patients with suspected CAD were randomly divided into training (184 patients) and validation (92 patients) cohorts. Variables extracted from clinical, physiological, and rest-stress SPECT MPI were screened. Stress-only and rest-stress MPI using ML were established and compared using the training cohort. Then the diagnostic performance of two models in diagnosing myocardial ischemia and infarction was evaluated in the validation cohort. RESULTS: Six ML models based on stress-only MPI selected summed stress score, summed wall thickness score of stress%, and end-diastolic volume of stress as key variables and performed equally good as rest-stress MPI in detecting CAD [area under the curve (AUC): 0.863 versus 0.877, P  = 0.519]. Furthermore, stress-only MPI showed a reasonable prediction of reversible deficit, as shown by rest-stress MPI (AUC: 0.861). Subsequently, nomogram models using the above-stated stress-only MPI variables showed a good prediction of CAD and reversible perfusion deficit in training and validation cohorts. CONCLUSION: Stress-only MPI demonstrated similar diagnostic performance compared with rest-stress MPI using 6 ML algorithms. Stress-only MPI with ML models can diagnose CAD and predict ischemia from scar.

GCLR: A self-supervised representation learning pretext task for glomerular filtration barrier segmentation in TEM images.

Automatic segmentation of the three substructures of glomerular filtration barrier (GFB) in transmission electron microscopy (TEM) images holds immense potential for aiding pathologists in renal disease diagnosis. However, the labor-intensive nature of manual annotations limits the training data for a fully-supervised deep learning model. Addressing this, our study harnesses self-supervised representation learning (SSRL) to utilize vast unlabeled data and mitigate annotation scarcity. Our innovation, GCLR, is a hybrid pixel-level pretext task tailored for GFB segmentation, integrating two subtasks: global clustering (GC) and local restoration (LR). GC captures the overall GFB by learning global context representations, while LR refines three substructures by learning local detail representations. Experiments on 18,928 unlabeled glomerular TEM images for self-supervised pre-training and 311 labeled ones for fine-tuning demonstrate that our proposed GCLR obtains the state-of-the-art segmentation results for all three substructures of GFB with the Dice similarity coefficient of 86.56 ± 0.16%, 75.56 ± 0.36%, and 79.41 ± 0.16%, respectively, compared with other representative self-supervised pretext tasks. Our proposed GCLR also outperforms the fully-supervised pre-training methods based on the three large-scale public datasets - MitoEM, COCO, and ImageNet - with less training data and time.

Systematic review and meta-analysis of the efficacy and safety of electroacupuncture for poststroke dysphagia.

Introduction: Optimal treatment strategies for post-stroke dysphagia (PSD) remain to be explored. Electroacupuncture (EA) has attracted widespread attention due to its simplicity, cheapness, and safety. However, the efficacy of EA in the treatment of PSD lacks high-level evidence-based medical support. This study aimed to systematically evaluate the clinical value of EA in the treatment of PSD. Methods: A total of seven databases were searched for relevant literature. All randomized controlled trials (RCTs) on EA alone or EA combined with other interventions for the treatment of PSD were assessed using the modified Jadad scale. The studies with a score of ≥4 were included. The quality of the included studies was then assessed using the Cochrane Collaboration's tool. The meta-analysis was performed using Rev. Man 5.3 software. Results: Twelve studies involving 1,358 patients were included in the meta-analysis. Meta-analysis results showed that the EA group was superior to the control group in terms of clinical response rate (OR = 2.63, 95% CI = 1.97 to 3.53) and videofluoroscopic swallowing study (VFSS) score (MD = 0.73, 95% CI = 0.29 to 1.16). There was no significant difference between the two groups in the standardized swallowing assessment (SSA) score (MD = -3.11, 95% CI = -6.45 to 0.23), Rosenbek penetration-aspiration scale (PAS) score (MD = -0.68, 95% CI = -2.78 to 1.41), Swallowing Quality of Life (SWAL-QOL) score (MD = 13.24, 95% CI = -7.74 to 34.21), or incidence of adverse events (OR = 1.58, 95% CI = 0.73 to 3.38). Conclusion: This study shows that EA combined with conventional treatment or other interventions can significantly improve the clinical response rate and VFSS score in patients with PSD without increasing adverse reactions.Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=396840.

Electroacupuncture at HT5 + GB20 promotes brain remodeling and significantly improves swallowing function in patients with stroke.

Background: This study compared the differences in the degree of brain activation, and swallowing function scales in patients with post-stroke dysphagia after treatment. We explored the mechanism of cortical remodeling and the improvement effect of electroacupuncture on swallowing function in patients and provided a theoretical basis for the clinical application of electroacupuncture. Methods: Fifty patients with post-stroke dysphagia were randomized to the control or electroacupuncture group. The control group underwent conventional swallowing rehabilitation for 30 min each time for 12 sessions. In the electroacupuncture group, electroacupuncture was performed based on conventional swallowing rehabilitation for 30 min each time for 12 sessions. Cortical activation tests and swallowing function assessments were performed before and after treatment. Statistical analyses were used to investigate the differences within and between the two groups to explore the treatment effects. Results: There were no statistical differences in clinical characteristics and baseline data between the two groups before treatment. Cortical activation and swallowing function were improved to different degrees in both groups after treatment compared with before treatment. After treatment, the electroacupuncture group showed higher LPM (t = 4.0780, p < 0.001) and RPM (t = 4.4026, p < 0.0001) cortical activation and tighter functional connectivity between RS1 and LM1 (t = 2.5336, p < 0.05), RM1 and LPM (t = 3.5339, p < 0.001), RPM and LM1 (t = 2.5302, p < 0.05), and LM1 and LPM (t = 2.9254, p < 0.01) compared with the control group. Correspondingly, the improvement in swallowing function was stronger in the electroacupuncture group than in the control group (p < 0.05). Conclusion: This study demonstrated that electroacupuncture based on conventional treatment activated more of the cerebral cortex associated with swallowing and promoted functional connectivity and remodeling of the brain. Accompanying the brain remodeling, patients in the electroacupuncture group also showed greater improvement in swallowing function. Clinical trial registration: ClinicalTrials.gov, ChiCTR2300067457.

Electroacupuncture at HT5 + GB20 produces stronger activation effect on swallowing cortex and muscle than single points.

Introduction: This study aimed to investigate the effects of electroacupuncture on cortical activation and swallowing muscle groups. The study examined brain activation in healthy subjects performing swallowing tasks during electroacupuncture. Additionally, the study analyzed electromyographic signals of swallowing muscle groups after electroacupuncture. Methods: Twenty-seven healthy subjects were randomly separated into three groups. They underwent electroacupuncture at HT5 acupoint (HT5 group), or GB20 acupoint (GB20 group), or HT5 + GB20 acupoint (HT5 + GB20 group) for 30 min of intervention. Subjects performed a swallowing task while receiving electroacupuncture. Functional near-infrared spectroscopy (fNIRS) was used to detect cortical activation and functional connectivity (FC). The mean amplitude values of the swallowing muscle groups after electroacupuncture were also measured. Statistical analysis was used to investigate the differences between the three groups. The protocol was registered with the China Clinical Trials Registry with the registration number ChiCTR2300067457. Results: Compared with the HT5 group, the HT5 + GB20 group showed higher cortical activation in the LM1 (t = 2.842, P < 0.05) and a tighter FC in the RM1 and LM1 (t = 2.4629, P < 0.05) with considerably increased mean amplitude values of the swallowing muscle groups (t = 5.2474, P < 0.0001). Increased FC was found in the HT5 + GB20 group compared to the GB20 group between the RM1 and RS1 (t = 2.9997, P < 0.01), RM1 and RPM (t = 2.2116, P < 0.05), RM1 and LM1 (t = 3.2078, P < 0.01), RPM and LM1 (t = 2.7440, P < 0.05). However, there were no statistically significant differences in cortical activation or mean amplitude values of swallowing muscle groups. Conclusion: This study showed that electroacupuncture at HT5 + GB20 acupoints particularly engaged the cerebral cortex related to swallowing, resulting in tighter functional connectivity and higher amplitude values of swallowing muscle groups than electroacupuncture at single acupoints. The results may reveal the mechanism of electroacupuncture for post-stroke swallowing dysphagia.