This PDF file contains the front matter associated with SPIE Proceedings Volume 13252, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

A fluorescence array based on quantum dots was built for the detection of biogenic amines, including spermine, spermidine, and putrescine. A one-step synthetic strategy is adopted for the fabrication of carbon quantum dots using N-Acetyl-L-cysteine and citric acid. The carbon quantum dots show pH sensitivity. The carbon quantum dots were combined with fluorescamine and fluorescein to form a fluorescent array sensor. On the strength of the distinct fluorescence responses between the fluorescence sensing units and biogenic amines, fluorescence spectra were collected to build a fluorescence spectral dataset. The fluorescence spectra data were utilized as features for the three amine substances. To accurately evaluate the different amines, principal component analysis (PCA) and hierarchical cluster analysis (HCA) were employed to process the spectral data. Qualitative and quantitative analysis of the three amines was successfully achieved. The quantitative range was 0.125 - 12.5 μM, with an extremely low detection limit of 0.125 μM. Furthermore, several supervised classification algorithms were applied to classify the spectral data. Some of the classification algorithms achieved accuracy rates of up to 100%. Focusing on spectral data analysis, this study provides an accurate, simple, and fast approach to amine detection.

4-Methylcytosine (4mC) represents a crucial DNA base modification, characterized by the addition of a methyl group to the cytosine base. This methylation involves normal gene expression, DNA repair, cell cycle regulation and so on. 4mC binding sites, specific regions within the DNA sequence, engage with 4mC through interactions with proteins or other molecules, influencing gene expression and other cellular functions. Accurate prediction and understanding of the 4mC binding site are of great importance for the in-depth study of the function and regulatory mechanisms of DNA methylation, and for elucidating the mechanisms of cellular processes and disease development. Traditional wet experiments for locating 4mC binding sites, despite their utility, are hindered by high costs and extensive time requirements. Therefore, various computational models have been developed to predict 4mCbinding sites. In this study, we develop a framework named “4mCMS” based on deep learning. The model uses a multi-scale feature encoding strategy that combines different kmers together so that a more comprehensive set of DNA sequence information can be captured. On the C. elegans dataset, the model has an ACC of 0.931, which exceeds the results of the comparative models. The experimental results demonstrate the effectiveness of our proposed model, as we achieve advanced results on the 4mC datasets.

Based on the principle and characteristics of surface electromyography (sEMG) signal, this paper builds a surface electromyography signal acquisition system, and collects the surface electromyography signals of four muscles of the left arm when the human body performs four hand movements. The EMG signal data of the four muscles were divided into 15 different data sets, and data filtering, effective segment extraction, feature extraction, feature normalization, and BP neural network model processing were performed. Through the network training of the EMG signal data sets of different muscle combinations, and the comprehensive evaluation of the recognition accuracy and complexity of each model, it is found that the BP neural network model trained by the two muscles of the ulnar wrist flexor and extensor digitorum is selected. In the case of high recognition accuracy, it has low complexity, and has the best effect in the process of hand gesture recognition, which solves the problem of muscle combination selection in the interactive experiment of hand gesture recognition.

Basic Tai Chi exercises serve as the foundation and core of Tai Chi practice. A deep understanding of the movement principles and essential characteristics is crucial to ensuring the effectiveness of training. However, practitioners often struggle to grasp the essentials, resulting in minimal progress or even injuries that may lead them to abandon the practice. To achieve a digital interpretation of the movement transitions and force application principles in basic Tai Chi exercises, we introduced surface electromyography (sEMG) detection into basic Tai Chi in this study. Eight Tai Chi experts were invited to construct an expert subject dataset. Then we extracted and analyzed the integral whole electric value and contribution rates of relevant muscles in basic Kai He Gong (KHG) exercise, as well as the dynamic changes on both sides of the body. An improved Teager-Kaiser Energy (TKE) algorithm was used to accurately identify the muscle force sequencing in expert-level and trainee-level basic movement. The results revealed that the cooperated relationship the left and right body sides and the crucial role of the erector spinae in the waist connecting the upper and lower limbs in the KHG exercises, demonstrating the Tai Chi principle of "controlling from the waist". In all, we explored the sEMG activity patterns in basic Tai Chi exercises in this study, enabling a digital interpretation of movement transitions and muscle force application principles. The results could provide references for the training and teaching of basic Tai Chi exercises.

As a weakly acidic organelle, the Golgi apparatus plays an important role in cell proliferation, enzyme activity regulation and ions transportation. When the pH of the Golgi apparatus changes, it will affect the normal function of the Golgi apparatus. Herein, a coumarin-based Golgi targeted fluorescent probe (Gol-Cou) is proposed, which shows significant pH dependent fluorescence at 447 nm and a linear relationship in the pH 5.7-7.1 range. The pKa value of Gol-Cou was measured at 6.39. Besides, we successfully applied this probe to visualize the change of Golgi pH during different conditions in HeLa cells.

The paper introduces a composite filter that combines Gaussian and edge filters, designed specifically for image processing tasks. This filter offers advantages such as multi-scale feature extraction, edge enhancement, noise suppression, stability, and flexibility. Experimental results demonstrate that the composite filter performs excellently in improving image quality, highlighting details, and reducing noise. Particularly in the context of medical CT imaging, it proves highly effective and can be widely applied in feature extraction, edge detection, and noise reduction tasks. The approach balances smoothness with edge preservation by first using a Gaussian filter to smooth the image and reduce noise, followed by convolution with the edge filter to enhance edges, ultimately producing a filter result that is both smooth and edge-preserving. Therefore, the composite filter holds significant potential for extensive applications in the field of image processing.

The electrocardiogram (ECG) signal plays a crucial role in biological signal processing, particularly in the field of medical diagnostics. This study is dedicated to exploring the utilization of passive components in ECG signal processing to devise an economical device aimed at amplifying physiological signals. In the hardware setup, we employed both pre- and post-amplifiers, bandpass filters, and notch filters to achieve specific objectives, including signal amplification, white noise reduction, and removal of power frequency interference. Complementing the hardware components, our software approach involves the implementation of an R peak extraction method to accurately determine heart rate following rigorous denoising preprocessing procedures. The primary objective of this research endeavor is to offer a practical and cost-effective solution for ECG acquisition, particularly beneficial for regions characterized by limited access to electricity and constrained economic resources. Through this study, we anticipate contributing to the advancement of biomedical engineering by facilitating enhanced ECG signal acquisition methodologies in resource-limited settings.

The purpose of this paper is to propose an automated deep neural network (DNN) based method for the assessment of skin volume in mice with laser injury. By constructing a laser-damaged mouse skin model, images of the damaged area were acquired using optical coherence tomography (OCT). The necessary dataset for deep learning was created by preprocessing these images. On this basis, U-Net and DeepLabV3+ models were used to segment OCT images of laser-damaged mouse skin. Algorithms were developed to quantify the volume of skin damage. The MIoU values of U-Net in the damage area (DA), (epi-)dermis layer (EDL), subcutaneous fat layer (SFL) and muscle fascia layer (FML) were 92.03%, 89.39%, 84.86% and 82.05%, respectively. The volume evaluation errors in the DA, EDL, SFL and FML were 8.926×10^-2 mm³, 4.086×10^-2 mm³, 3.947×10-2 mm³ and 5.475×10^-2 mm³, respectively.

CT images of liver reflect the internal tissue structure of liver through density distribution imaging. In CT images of healthy liver, the tissues of adjacent layers are highly similar, and this similarity is the basis for the feature description of CT images based on healthy tissues. The research shows that SIFT feature can be used as the basic form of such similarity description. In this paper, a method of similarity matching between liver images based on improved SIFT algorithm is proposed. Taking the internal centroid of liver image as the basic reference point, SIFT key points are constrained by centroid space Angle and then feature descriptors are constructed. It can solve the problem of cross-region mismatching in SIFT feature matching, ensure the advantages of invariant affine transform and invariant scale transform of SIFT feature in CT image, and realize the similarity matching processing of adjacent layers in the tomography image. The experimental results show that the proposed algorithm can effectively avoid the occurrence of cross-region mismatching within homogeneous organs, and also verify the feasibility of feature matching based on similarity description of adjacent CT images. It provides an effective method for further characterization and semantic research of healthy tissue.

It is important to predict microbe-disease associations, as it helps to understand the cause of diseases episodes, the prevention of diseases, among other roles. Traditionally, the study of microbe-disease association is mainly based on laboratory purified culture for research and observation, which is highly scientific but has some limitations, such as some microbes needing to be cultured in a laboratory environment and costs heavily. This study proposes an attention guided graph convolutional network d approach (AGGCN) to predict possible potential associations between microbes and diseases. In the first place, AGGCN constructs four different similarity networks of microbes and diseases, which are combined with association networks to form a microbe-disease heterogeneous network. Subsequently, it is used as an input to the graph convolutional neural network to obtain each node embedding of this heterogeneous network. In this study, a three-layer network is used, so three different types of node embeddings can be obtained, and these three different types of node embedding representations are integrated via the multichannel graph attention mechanism to obtain the final node embeddings. Finally, the node embeddings obtained above are used for association prediction of microbes-diseases in heterogeneous networks. Experimentally verified, the AUC of AGGCN is 0.8972, and the AUPR is 0.5387. Comparative experimental results indicated that AGGCN shows better than the other five mainstream methods. It provides a new insight to explore network potential feature representation and predict associations between microbe and disease.

The rising incidence of cancer underscores the significance of cancer screening and thorough investigation of cancer drug resistance mechanisms. Higher–resolution microscopy is required to uncover these mechanisms. Super–resolution fluorescence imaging (SRM) is an advanced imaging technique that can surpass the resolution limit of conventional fluorescence microscopy and is, therefore, crucial in this field. This paper discusses the principles and development of super–resolution fluorescence imaging technology, including its different classifications, such as STED, SIM, and PALM/STORM. The labelling methods used in super–resolution fluorescence imaging technology include fluorescent proteins, antibody probes, and aptamer probes. Small molecule probes have become a popular marker in super–resolution fluorescence imaging due to their small size, high stability, and good permeability. Their use can aid in revealing cancer pathogenesis and support more effective therapeutic approaches.

Lung Computed Tomography (CT) images play an important role in the diagnosis and treatment of patients with COVID- 19. However, manually identifying lesions in CT images is very time-consuming and requires specialized medical knowledge. The accuracy and integrity of the existing models are still not ideal due to the reasons such as scattered distribution, small lesions and blurred boundaries. Therefore, we propose a segmentation model based on U-Net architecture by combining Mirror-symmetry Boundary Guided (MBG) module and then adding Spatial Attention Dilation (SAD) convolution Module, called MSU-Net. The SAD module uses spatial attention mechanism and dilated convolution and multi-scale feature extraction strategy to enhance the network’s ability to recognize small lesions. The MBG module further enhances the model’s ability to capture more complex context information and boundary details, making the model more robust and able to effectively deal with the challenges of fuzzy boundaries. The proposed method has shown superior performance in terms of the Dice coefficient, Hausdorff distance, and Sensitivity on the publicly available COVID-19 CT dataset.

6-methyl-1-indanone as a photoproducts of 2,5-dimethylphenacyl esters based on the effective photoinduced intramolecular hydrogen abstraction can further sensitize the photoreaction. Self-photosensitization was discussed in details in the paper and the mechanism was deduced as the intermolecular triplet-triplet energy transfer. Importantly, photosensitization can increase the rate of photodeprotective reaction, but also present a new approach to wavelength-dependent deprotection.

In order to enhance the transmission and storage efficiency of medical images, this paper proposes a lossless compression algorithm designed for medical imaging. The algorithm processes input images using an improved weighted template matching predictor, yielding error calculation results and symbol matrices. These outcomes are then entropy-encoded using PNG and Deflate encoding for the predictive results, achieving compression. Experimental verification was conducted using a test dataset comprised of cross-sectional images extracted from multiple scan sequences. The results indicate that the proposed algorithm achieves an average compression ratio of 1:3.00926, outperforming other lossless compression algorithms based on linear prediction, such as PNG and TIFF.

Infantile hemangiomas (IH) and venous malformations (VM) are two common vascular diseases that occur during childhood. Both diseases present similar features in grayscale ultrasound images; therefore, using separate grayscale ultrasound images to classify IH and VM can result in misclassification owing to insufficient comprehensible information in the single-modality images. To address this problem, this study proposes an ultrasound multimodal classification model. The model uses grayscale ultrasound images and the corresponding color ultrasound images of IH and VM as input to the feature extraction modules. The feature outputs from the two modules are spliced and fused. Finally, the fused features are input into the fully connected layer to be classified and recognized. The data consists of 1498 grayscale images and 1498 color images. The experimental results showed that the accuracy of the ultrasound multimodal classification model was 96.9%, which was higher than that of seven existing unimodal models.

In the teaching of traditional Chinese medicine, due to the large differences in the theoretical principles of Chinese and Western medicine, it is difficult for students to understand the profoundness and abstraction of Chinese medicine clinical medicine students. This article proposes to build a knowledge map of viscera state based on Neo4j, resolve difficulties, reduce gradient, and use students' subjective initiative. First of all, based on teaching materials, curriculum outlines, teaching plans and lectures to determine the knowledge points of the curriculum content; second, the five internal organs of Chinese medicine are centered, covering the disease, syndrome, prescriptions, and drugs. Relationships, semantic relationships, and instantize; then, use CNN-BILSTM-CRF and K-means to extract knowledge; Finally, the visual display and storage of the relationship between viscera state knowledge points and relationships based on the Neo4j diagram database. The constructed five internal organs knowledge maps, using the connection between key knowledge points, laid a certain foundation for the discovery of the potential relationship of Traditional Chinese Medicine viscera state theory.

In order to achieve accurate segmentation of lung bronchus and obtain the inner diameter of the bronchus, a new method of lung bronchus segmentation design and post-processing optimization based on 3D UX-Net model was proposed. Firstly, 3D UX-Net model was used to train bronchial data. Secondly, a new post-processing process to improve segmentation accuracy is combined with region selection, threshold adjustment and maximum connected domain analysis. Finally, a comparison experiment with other algorithm models is conducted on the open data set. The experimental results showed that the Dice coefficient evaluation index based on the 3D UX-Net model was improved from 0.83 to 0.85, and the error of the published and clinical data sets on the inner diameter of the bronchus was within 0.2mm. It was concluded that the model and post-processing method proposed in this study could segment the bronchus quickly and accurately.

Positron emission tomography(PET) and computed tomography(CT) imaging can obtain the functional metabolic information and anatomical information of the human body, respectively. In response to the problem of how to merge the complementary information of PET and CT, a lung tumor image fusion method based on VGG19 and GAN model is proposed in this paper. The basic idea is to extract the content of the PET image and the style of the CT image separately, and then merge these two to generate the fusion image. The style image is represented using a series of statistical methods that capture features such as color, texture, and style. The content image is represented using a series of deep learning techniques that capture features such as shapes, structures, and semantic information. The experimental results show that the proposed method has better subjective and objective results than the compared algorithms, which can generate higher-quality PET/CT fusion images. This method provides comprehensive diagnostic information, which is helpful for the diagnosis of lung tumors in clinical treatment and reduces misdiagnosis and underdiagnoses.

Optical fiber photoacoustic transducers are a type of medical imaging technology that combines the high resolution of ultrasound with the versatility and precision of fiber optics. This paper presents the development and application of an optical fiber photoacoustic transducer based on candle soot (CS) carbon nanoparticles for medical imaging. The transducer leverages the photoacoustic effect, which involves the conversion of light energy into ultrasound through thermal expansion and contraction. The study focuses on the preparation of the transducer using a CS/PDMS composite material, which is dip-coated onto the end of an optical fiber with a peak-to-peak acoustic pressure of ~4.5 MPa and a -6dB bandwidth of approximately ~23.8 MHz (198%, center frequency: ~12 MHz). The resulting device demonstrated high-resolution imaging capabilities for fish head part. This result demonstrates the potential of optical fiber photoacoustic transducers in medical imaging.

Inner speech is an internal language that expresses thoughts and ideas within the mind, without producing sound. Recording inner speech activities through scalp electroencephalography (EEG) provides insights into the mechanism of language production. Our current study aims to explore the neural mechanisms of inner speech among Chinese native speakers through collecting EEG signals during inner speech tasks, and extract topographical and time-frequency related features for decoding the brain’s language formation. Specifically, we designed a picture-induced speech and non-speech task. EEG signals were collected from 7 subjects across three different tasks: inner speech, pronounced speech, and visual tasks. Through computational analysis of event-related potential and time-frequency domain features, comparisons were made regarding power discrepancies among different tasks and words across six frequency bands. Power difference topographic maps were also generated to compare spatial disparities between tasks and words. Our results indicate that the inner speech in Chinese Mandarin shares a similar articulatory encoding process with pronounced speech. After the first second of inner speech execution, the spatial distribution of gamma power may help distinguish directional words such as "up", "down", "left", and "right". This research provides insights for further exploration of inner speech decoding among Chinese native speakers.

Based on thermal injury to biological tissues, this paper proposes a temperature-adaptive control strategy for heating atherosclerotic plaques in the blood flow field under focused ultrasound (FU). The temperatures in the spatial distribution of the blood vessel walls and their changes over time were obtained by solving a thermal-fluid-solid coupling model using the finite element method. The temperatures in the entire vascular region were automatically tracked using a local search procedure. The maximum temperature of the region is controlled below 42 °C by using a proportional-integral-derivative (PID) controller, which can achieve the stability control on a specified power output of FU transducer. The results showed that, compared to the traditional PID control, the PID control optimized by noise mean method had better tracking control in noisy environments. The model can realize rapid and accurate control according to the target function, and may provide a technique for FU hyperthermia in the treatment of atherosclerosis.

Breast cancer is the highest incidence rate tumor in women, and it is also the main cause of death. Metabolic dysregulation and heterogeneous are considered as important driver in the development and progression of breast cancer, while the heterogeneous of metabolism in breast cancers still remains to be elucidated. Identification of heterogeneous-related metabolites or genes may help to reveal new breast cancer targets. In this study, the breast cancer metabolites data and bulk RNA-seq data were integrated to identify the potential feature metabolites and feature genes associated with estrogen receptor status of breast cancer. The identified feature metabolites such as maltopentaose and feature genes such as SPDEF, TFF1 and TFF3 are associated with metabolic process and immune response of breast cancer. Our study may help to further understand the metabolic mechanism and human adaptive immune response in breast cancer.

This study aimed to investigate the potential targets and molecular mechanisms underlying the therapeutic effects of Farrerol on spinal cord injury (SCI) using various methods. The databases TCMSP, PharmMapper, CTD, and GeneCards were utilized to collect Farrerol and SCI-related targets. Molecular docking, molecular dynamics, and cell experiments were conducted to explore the interaction between Farrerol and its target proteins. GO and KEGG pathway enrichment analyses were performed to identify the biological processes and pathways involved in Farrerol's therapeutic effects. The results showed that Farrerol significantly reduced TNF-α production in LPS-stimulated microglia, improved microglial cell apoptosis, and reduced reactive oxygen species generation. Additionally, Farrerol could decrease P65 nuclear translocation mediated by LPS, thereby suppressing the inflammatory response of microglia. The study concluded that Farrerol may improve pathological changes in spinal cord injury by inhibiting inflammation and improving cell apoptosis.

In this paper, we propose a graph structure clustering-based spatio-temporal graph convolutional network (GSC-GCN) to classify Autism Spectrum Disorder. Firstly, a sliding window with fixed size is utilized to sample the preprocessed restingstate functional magnetic resonance imaging time series. Then functional connectivity matrices for all the subsequences are constructed and a series of dynamical temporal and spatial graphs are built accordingly. Next, to effectively reduce dimensionality and capture key functional connections, three different sizes of coarsening graphs are generated using clustering methods. Subsequently, a spatial-temporal graph convolution network is designed which focuses on the temporal dynamics of graph series as well as the spatial characteristics of functional brain network. In order to verify the effectiveness of our model, we test our GSC-GCN on the Autism Brain Imaging Data Exchange dataset and compare it with the state-of-the-art techniques. Results show that GSC-GCN outperforms other methods, indicating that the comprehensive use of both temporal and spatial features deserves further exploration.

Multiple implants need to be placed in different regions of the mandible based on the remaining bone mass assessment and implantation procedure of different patients with complete edentulousness. However, little is known about the impact of the implant sites arrangement on framework, distal implant and surrounding bone stress. This paper was to explore the biomechanical behavior changes of the implant sites distribution on the framework in the axial design and compared with the inclined design. Arched mandible models with four different implant placement arrangements were built using 4 axial implants. Another alternative inclined design was generated for comparison with the axial design while maintaining an even distribution of the implant positions. A 300 N oblique load at 45° to the occlusal plane was applied to bilateral posterior-tooth area for finite element analysis. For the axial design, the maximum stress occurred in groups with multiple implants adjacently arranged, whereas the minimum peak stress occurred in groups with each implant evenly spaced. The inclined group exhibited lower stress levels on the distal implant and surrounding bone. For full-arch implant restorations, an inclined designed with 4 implants evenly distributed on the mandible is preferred to obtain optimal stress distribution.

To study the non-EEG sleep assessment method suitable for wearable devices, a complex network-based sleep stage characteristic analysis method for short-term heart rate variability data is proposed. Short-term heartbeat interval time series in different sleep stages are mapped to directed unweighted networks, respectively. Then, several characteristic parameters, including the average out-degree, the entropy of the degree distribution, the average clustering coefficient, and the entropy of clustering coefficient are calculated and analyzed. The results indicate that the various network measures show significant differences in the amplitude levels during wakefulness, light sleep, deep sleep, and rapid eye movement sleep. These differences can reflect the dynamic characteristics of short-term heart rate interval series in different sleep stages, helping to reveal microvariations and patterns during sleep processes. Furthermore, these findings align with existing physiological cognition, offering new supplementary parameters for sleep staging research based on short-term heart rate variability, which has the potential to enhance the effectiveness of sleep staging in wearable devices.

The emergence of advanced gene sequencing technologies has transformed the study of structural and genetic variations within the human genome. This has spurred increased research into the genetic underpinnings of complex diseases. While medical research often utilizes ordinal response variables to reflect stages or severity of diseases, most existing methods focus on binary or continuous traits, neglecting vital ordinal data. To address this gap, this article introduces the Partial Linear Cumulative Logistic (PLCL) method within a mixed model framework, designed to accurately process ordinal traits and explore their genetic correlations with diseases. The PLCL method employs B-spline functions for modeling, uses Sieve maximum likelihood estimation for regression coefficients, and determines P-values via a likelihood ratio test. An aggregated Cauchy association test then combines these P-values for comprehensive analysis. Simulation results affirm PLCL's robustness, efficiency in P-value computation without permutations, strict Type I error control, and superior performance compared to traditional methods.

Purpose: Positron Emission Tomography (PET) is an advanced clinical imaging modality within the field of nuclear medicine. Despite strict adherence to standardized imaging protocols, variations in image indices caused by various factors complicate the comparison of multicenter PET data. This study aims to investigate whether a Gamma model-based standardized uptake value (SUV) harmonization algorithm can improve the classification of imaging radiomics features across multicenter datasets. Methods: This study includes PET/CT and PET/MR image data of 251 patients from Guangdong Provincial People's Hospital without significant lesions in the liver, spleen, and bone regions. A Gamma modelbased harmonization algorithm is developed and evaluated for its impact on radiomic-based tissue classification. The effectiveness of the harmonization algorithm is assessed based on classification accuracy, in comparison with unharmonized and ComBat-harmonized data. Results: The study identifies 86 radiomic features spanning five categories from volumes of interest (VOIs) in three tissues. Our harmonization algorithm significantly improved tissue classification accuracy compared to unharmonized data. By utilizing the PCA-reduced 10-dimensional radiomic features, the algorithm achieved a mean accuracy of 94.5% and 93.0% in the training and validation sets, respectively, representing a substantial improvement over the unharmonized accuracies of 55.8% and 52.0%. Conclusion: These findings suggest that the Gamma model-based harmonization algorithm facilitates the investigation of imaging radiomics in multicenter data.

This study evaluates the performance of various supervised machine learning models in analyzing highly correlated neural signaling data from the Adolescent Brain Cognitive Development (ABCD) Study, with a focus on predicting obsessive-compulsive disorder scales. We simulated a dataset to mimic the correlation structures commonly found in imaging data and evaluated logistic regression, elastic networks, random forests, and XGBoost on their ability to handle multicollinearity and accurately identify predictive features. Our study aims to guide the selection of appropriate machine learning methods for processing neuroimaging data, highlighting models that best capture underlying signals in high feature correlations and prioritize clinically relevant features associated with Obsessive-Compulsive Disorder (OCD).

Cardiovascular illnesses are becoming a major global health concern requires sophisticated, accurate and efficient cardiac computed tomography heart image analysis, to assist diagnosis and therapy planning, particularly accurate left ventricle segmentation. In order to overcome this, this paper created an enhanced U-Net model that greatly enhances left ventricular detail identification by combining a novel multi-scale feature fusion with an attention mechanism. By comparing the performance on the same dataset, the enhanced U-Net shows relative advantages in important metrics such as precision, recall, F1 score, IoU and Dice coefficient. The enhanced U-Net model is more effective in acquiring knowledge and more reliable in reaching a steady state, which is significant for practical clinical applications. Furthermore, the model is optimised by using depthwise separable convolution, which guarantees lightness and increases training speed. Extensive experiments on many cardiac CT datasets validated the model's exceptional segmentation accuracy, computational performance, and flexibility, outperforming state-of-the-art deep learning techniques and offering new opportunities for accurate diagnosis and treatment of cardiovascular diseases.

Deep learning techniques are extensively employed for the classification and diagnosis of Alzheimer's disease (AD). Existing classification methods often use a single feature from functional magnetic resonance imaging (fMRI) and neglect fusion among multiple features. Thus, this paper proposes a framework of GCN combined with MLP model (ie, GCN-MLP) by utilizing fMRI multi-feature analysis, which includes amplitude of low-frequency fluctuations (ALFF), regional homogeneity (ReHo), and degree centrality (DC), for the classification of AD patients. The KNN algorithm is used to construct the adjacency matrix to capture the connectivity relationship between nodes, thus enabling the GCN to propagate and learn features using this relationship, and output the updated feature matrix for further integration with the MLP model to improve the predictive ability of the model. In addition, an interpretable feature learning method is employed to provide interpretable disease diagnosis information that enhances the interpretability of the model. It is found that the suggested approach achieves a classification accuracy of 95.2% and outperforms existing AD classification methods.

MS-DNA(Microsatellite DNA) is widely used in gene mapping, clinical medicine, paternity testing, anthropology, population genetics, and species identification. In this study, we examined the variation and distribution of MS-DNA in the mitochondrial genome (mtDNA) and chloroplast genome (cpDNA) of cotton to develop more valuable MS-DNA molecular markers. Based on bioinformatics techniques, we selected full sequences of mtDNA and cpDNA from eight cotton species and utilized Perl language and open-source code (MISA) to create a tool for identifying and localizing MSDNA sequences for analysis. The results revealed that the lengths of mtDNA among the eight cotton species ranged from 644,311bp to 687,482bp, while their cpDNA lengths were between 160,072bp and 160,317bp, indicating that the length of mtDNA was approximately four times larger than that of cpDNA. The frequency of MS-DNA was in cpDNA higher than mtDNA. Regarding distribution characteristics, MS-DNA in the cotton cytoplasmic genome were mainly located in intergenic regions, with AT content exceeding CG content. Compared to the chloroplast genome, the mitochondrial genome contained more repetitive sequence regions, resulting in higher richness of MS-DNA. In terms of types, the microsatellite categories showed a higher prevalence of P4 type repeats over P3 type, with P5 and P6 repeats exhibiting specificity. Additionally, a total of 31 types (61 instances) of compound MS-DNA were detected across the eight cotton mtDNAs, while 18 types (32 instances) were found in cpDNAs. These loci can serve as candidate sites, providing assistance for the subsequent development of specific molecular markers.

USP22, as a deubiquitinating enzyme, is highly expressed in various types of cancer and promotes the progression of the cell cycle and cancer development by interacting with cell cycle regulatory factors. This study, for the first time, successfully expressed a highly active USP22 complex in vitro using a method based on the PQLink series of vectors. It was found that USP22 must form a complex with the auxiliary proteins ATXN7, ATXN7L3, and ENY2 to possess deubiquitinating enzyme activity. Through optimizing gene sequences, the construction of protein expression vectors, expression-test, overexpression and purification, we obtained a highly active USP22 complex and verified its deubiquitinating enzyme activity using Ub-AMC hydrolysis assay. In addition, the experimental results showed that the enzyme activity of the USP22 complex would be lost if any of the auxiliary proteins were missing. This study not only provides important information for understanding the mechanism of USP22 in cancer but also lays the foundation for the development of small molecule inhibitors targeting USP22, which is expected to become a new strategy for cancer treatment.

In genomics analysis, with the development of sequencing technology and the rapid growth of data volume, how to effectively identify and analyze important information in genomic data from massive data has become an important challenge. Identify key patterns and features in genomic data through advanced community testing methods. First, genomic data often contains a lot of noise and redundant information that needs to be processed through steps such as normalization, filtering, and dimensionality reduction. Normalization makes data at different scales comparable, filtering out data points with low quality or missing values, and dimensionality reduction reduces data dimensions through principal component analysis (PCA) and retains the main information. The gene co-expression network establishes the connection relationship between genes by calculating the expression similarity matrix between genes, and applies the Louvain algorithm to maximize the modularity of the network, aggregate nodes layer by layer, form a community structure, and identify the complex relationship and potential structure between genes. Finally, we utilize the cancer dataset to evaluate the proposed model. From our extensive experimental results, we can observe that Louvain's algorithm indicates outperformances and accuracy in the recognition and analysis of genomic data.

Improving the regeneration efficiency of poplar trees is an important role in their genetic transformation. In this study, a regeneration system of 741 poplar tree, a hybrid descendant of Populus alba, was developed through cluster seedling differentiation. By utilizing different hormone concentrations and their combinations to induce callus tissue, cluster seedling formation, and cluster seedling stem extraction, the regeneration system of 741 poplar planting plants was optimized. The optimized concentration and combination of induced callus culture hormones is 1.0 mg/L 6-BA+0.2 mg/L NAA, that of cluster seedlings differentiation is 0.5 mg/L 6-BA+0.1 mg/L NAA, and that of stem extraction is 0.3 mg/L 6-BA+0.05 mg/L NAA. In addition, compared with normal photoperiod, it was found that continuous dark cultivation of tender leaves for 7 days and of petioles for 5 days significantly increased the induction rate of callus tissue. In summary, we established an efficient regeneration system for 741 poplar via cluster seedling differentiation involved in hormone level and combination optimization, photoperiod regulation, and sucrose concentration screening, which provides assurance for genetic transformation of poplar.

Lung cancer is a prevalent and life-threatening malignant tumor that significantly impacts human health. For clinical diagnosis of this cancer, positron emission tomography (PET) and computed tomography (CT) are frequently employed imaging methods. Using PET/CT images to accurate pathological classify lung cancer is crucial for effective clinical treatment. However, obtaining complete PET/CT data is often challenging due to clinical limitations. Therefore, developing a classification approach that addresses the problem of missing modalities has substantial clinical importance. This article introduces a novel neural network based on knowledge distillation to tackle the issue of missing modality, specifically designed for scenarios where only CT data is available. Our method effectively transfers knowledge from PET/CT data to CT data, capturing both intra-class and inter-class features. These features enable student models to learn robust feature representations from teacher models, using only CT data for inference. Our experiments resulted in an accuracy of 0.847. The findings suggest that our proposed method performs best when the inference data comprises only the CT modality.

This paper proposes a stable, deployable, and user-friendly management system that integrates data visualization and RFID-based Internet of Things (IoT) technologies to optimize the workflow of the Central Sterile Supply Department (CSSD) and achieve more scientific information management. The implementation of the CSSD management system optimizes the workflow of sterile item supply, improves the level of hospital intelligent management and medical safety services.

Accurate assessment of rehabilitation movements is crucial for improving the rehabilitation outcome of patients. However, traditional assessment methods have problems such as incomplete data and interfering information, and the study proposes a rehabilitation movement assessment technique based on 3D coordinate data to improve the accuracy and efficiency of the assessment. The study used Azure Kinect to acquire 3D coordinate data of the human body and utilized a dynamic time regularization algorithm to assess rehabilitation movements. The results show that the improved dynamic time regularization algorithm outperforms the traditional algorithm in terms of action segmentation accuracy, feature extraction accuracy, and action assessment accuracy. The average similarity of the improved dynamic temporal regularization algorithm reaches 0.9958, which is significantly higher than other algorithms. The technique proposed in the study can effectively improve the accuracy of rehabilitation action assessment, which has important application value in the field of rehabilitation medicine.

Schizophrenia (SZ) is a complex and heterogeneous disorder. Characterizing functional magnetic resonance imaging (fMRI) imaging patterns specific to SZ is challenging and can benefit greatly from advanced artificial intelligence techniques. Prior research has shown that fMRI data can be summarized in functional network connectivity matrices (FNCs) using group independent component analysis (ICA), significantly reducing computational burden for tasks such as classification which also maintaining relevant information. Recent advances in the development of state-of-the-art latent diffusion models (LDMs) have shown promise for various generative tasks, such as image generation and data augmentation. This study leverages a lightweight vision transformer (ViT) encoder-decoder architecture instead of the traditional U-net as an effective denoising function in LDMs for FNC generation. This approach addresses the challenge of small datasets that often lead to lower accuracy such as in the use of neuroimaging data to diagnose schizophrenia.

Volatile components in Galla chinensis have the potential to exhibit anti-breast cancer (BC) properties, but as of now, no studies are reporting the specific mechanisms of action. In this study, volatile components and their relative contents in G. chinensis were identified using HS-SPME-GC-MS technology, and the potential mechanisms of the anti-BC properties of these volatile components in G. chinensis were explored using bioinformatics methods. Through analysis, 28 compounds were identified, from which 10 candidate compounds were selected. A search and prediction yielded 160 drug-related targets and 18,061 disease-related targets. Using bioinformatics methods, 2 hub genes (MAOA and MAOB) closely related to BC were identified. Through GO and KEGG analysis, enrichment was obtained for 264 biological processes, 124 molecular functions, 61 cellular components, and 104 important pathways closely related to BC. The study results indicate that volatile components in G. chinensis may target MAOA and MAOB through compounds such as Naphthalene, decahydro-4a-methyl-1-methylene-7-(1-methylethenyl)-, [4aR-(4aà,7à,8aá)]-, and Benzoic acid, 2-ethylhexyl ester, inhibiting signal transduction in drug metabolism pathways such as Drug metabolism- cytochrome P450 and Tyrosine metabolism. The expression validation of hub genes further demonstrates significant differential expression of MAOA and MAOB genes between normal individuals and cancer patients. This study preliminarily explored the potential mechanism of the volatile components of G. chinensis in anti-BC. The findings not only provide new insights and methods for the treatment of BC but also offer new revelations and possibilities for the entire field of cancer research.

Cancer is a global health challenge, and the utilization of multi-omics data offers a comprehensive approach to understanding and predicting human disease. Despite the abundance of gene expression, DNA methylation, and protein-protein interaction data, the integration of these datasets for accurate cancer classification prediction remains challenging. This paper introduces MoSwformer, a novel multi-omics framework for multi-omics analysis that combines Transformer based on sliding window embedding (SET), multi-omics attention mechanisms and omics integrated representation learning (MAL). By leveraging SET for feature extraction and MAL for multi-omics data fusion, MoSwformer improves cancer subtype classification. The framework serves as a powerful tool for multi-omics data analysis, highlighting its potential in advancing cancer research and precision medicine. Additionally, the findings from the ablation study provide convincing evidence for the indispensable role of the three integral modules within the MoSwformer framework. MoSwformer is freely available at https://github.com/ShiLab-GitHub/MoSwformer.

Currently, the treatment methods for defects in oral soft and hard tissues primarily focus on non-regenerative restoration, bone transplantation and oral medicine treatments. However, these methods have certain drawbacks and risks, such as rejection reactions and the need for secondary removal. As a result, researchers have turned their attention to stem cells as a new approach. Stem cells possess immunomodulatory and regenerative abilities, and they can differentiate into various cell types, providing a foundation for tissue regeneration. During the repair process, scaffold materials are often utilized to load stem cells, enabling three-dimensional cell growth and tissue contact. The multilineage differentiation potential of stem cells, combined with the three-dimensional environment provided by scaffolds, establishes the groundwork for the application of stem cell-loaded scaffolds in oral and maxillofacial tissue regeneration. While there is a wide range of stem cell-loaded scaffold materials available, there is currently a lack of comprehensive research summaries in the field of dental applications. Therefore, this article aims to provide a review of commonly used stem cell-loaded scaffolds in oral tissue engineering, with the goal of offering reliable references for clinical selection.

Abnormal expression of glycogen synthase 2 (GYS2) has been linked to tumorigenesis. However, there is currently a scarcity of reports on the relationship between GYS2 expression and its potential as prognostic factors (PFs) and its link to immune cell infiltration (ICI) in hepatocellular carcinoma (HCC). In this study, bio-informatics methods were applied to determine the potential correlation between GYS2 expression and ICI in tumor tissues, as well as its prognostic value in HCC. The findings displayed that GYS2 expression in cancerous tissues was reduced compared to normal tissue. Additionally, it was found that low expression of GYS2 was associated with decreased disease specific survival (DSS), progression-free survival (PFS), and overall survival (OS) rates in HCC patients. Moreover, both Cox univariate and multivariate analyses indicated that GYS2 expression and pM-stage as independent PFs of OS in HCC patients. Co-expression network revealed that GYS2 exhibited crucial functions in metabolic processes, fatty acid breakdown and steroid hormone biosynthesis. Finally, negative correlations of GYS2 expression with tumor ICI of DC cells, exhausted T cells, M1 macrophages, TAMs, B cells, CD8+ T cells and CD4+ T cells, as well as immune checkpoint molecules (CTLA4, PDCD1, LGALS9) and immunomodulatory factor TGFβ were observed. These findings propose GYS2 as a promising standalone prognostic indicator for HCC patient assessment and risk stratification, and suggest its involvement in ICI, which could influence the response to immunotherapy in HCC.

The complete mRNA sequence of a specific gene in tobacco (Nicotiana tabacum), known as 2-hydroxyisoflavanone dehydratase, was amplified using rapid amplification of cDNA ends (RACE) methods. The mRNA produced by the tobacco 2-hydroxyisoflavanone dehydratasedase gene was determined to have a length of 1,278bp, and it contains an open reading frame of 966 bp that codes for a protein consisting of 321 amino acids. After conducting sequence analysis, it was found that the tobacco 2-hydroxyisoflavanone dehydratasedase exhibits significant similarity to comparable genes in other plant species, including nicotiana tomentosiformis (99%), capsicum annuum (78%), potato (75%), lycopersicon pennellii (73%) and lycopersicon esculentum (72%). Upon further analysis of the tobacco genomic sequences database, it was discovered that this specific gene does not contain any introns and is composed of a single exon. Furthermore, our results suggest a closer genetic relationship between the 2-hydroxyisoflavanone dehydratasedase gene in tobacco and its equivalent in nicotiana tomentosiformis. Furthermore, tissue expression profile analysis demonstrated high expression levels of this gene in both flower and leaf tissues, while moderate expression was noted in both the root and stem tissues. This experiment serves as an important foundation for future investigations into the functions and characteristics of this specific gene in tobacco.

An assessment of a patient's health status significantly relies on the analysis of blood components, particularly through the accurate enumeration of red blood cells, white blood cells, and platelets. Traditional algorithms for target detection typically encounter challenges such as inadequate accuracy, as well as frequent false positives and false negatives in the detection of blood cells. In this study, a multi-scale fusion blood cell detection model AGF-YOLOv8 is proposed, which firstly introduces the CBAM attention mechanism between the ConvModule and the Batch Normalization layer to improve the attention to key features; Secondly, it combines the structure of the SPPF and the FPN to realize multi-scale feature fusion, and adopts the Ghost module to reduce model parameters and computational complexity; Finally, to address sample imbalance, the Focal Loss is used in the Darknet Bottleneck to optimize the recognition of difficult-to-divide samples. Experimental findings indicate that, relative to the original YOLOv8, the detection accuracy of the AGF-YOLOv8 model has increased by four percentage points, which can provide auxiliary diagnosis for healthcare professionals in the future.

Tongue squamous cell carcinoma (TSCC) is an oral cancer affecting the epithelial cells of the tongue. It is a serious and potentially lethal disease that requires prompt diagnosis and treatment. Multiple genes associated with immunity have been discovered as crucial in the growth and progression of cancer. In this research, we analyzed the gene expression profiles of patients with TSCC from TCGA database. A prognostic model for TSCC was established using five immune-related genes: TRAP1, LAT, NPY, IL20, and LBP. The prognostic model demonstrated its effectiveness, reliability, and stability through Kaplan-Meier, ROC curve, and both univariate and multivariate analysis. This immune-based prognostic model offers a biomarker for monitoring long-term treatment outcomes in TSCC but also suggests a potential immunotherapy approach for this disease.

Genome-wide association studies (GWAS) have demonstrated their effectiveness in discovering genetic variants associated with diseases such as Schizophrenia. However, translating these findings into therapeutic discovery and drug prediction remains a significant challenge. The NETTAG approach employs a Graph Convolutional Network (GCN) model named NOCD-G for identifying Alzheimer’s Disease (AD)-associated genes and repurposing drugs. Yet, whether there are better models for such cases, or whether this model continues to perform well in other scenarios, are questions that remain unanswered. Against this backdrop, we introduce CTAG, an optimized workflow designed to identify genes with disease-causing risks and map potential drugs for Schizophrenia. CTAG has made the following improvements over the NETTAG framework: 1. Expanding the Model Pool and Comparing Models 2. Enlarging the Dataset 3. New Findings in Drug Repurposing.

Platelets are essential for the processes of hemostasis and thrombosis formation within the human body. For diseases that are predominantly characterized by immune thrombocytopenia, the administration of platelet suspensions is considered a superior treatment method. As platelets are exclusively derived from voluntary blood donation, a critical shortage of platelets has emerged as a global concern. The mass production of platelets in vitro via laboratory techniques is crucial to meet the present and future demands for transfusions. It has been verified by previous research that platelets can be generated in vitro by reprogramming adherent cells. The difficulty in obtaining and culturing a large quantity of these cells in vitro presents a significant barrier to both the progression of research and the scaling of experimental applications. Consequently, the objective of this study is to investigate the potential of reprogramming B cells, which are also derived from lymphoid stem cells, to produce megakaryocytes and subsequently platelets. B cells, being suspension cells, offer several advantages over adherent cells, including ease of acquisition, low cost, and rapid proliferation. Therefore, in the context of this experiment, overexpression of megakaryocyte differentiation-related transcription factors—GATA1, FLI1, AML1, and NFE2—was induced in primary mouse B cells. Flow cytometric analysis demonstrated the enhanced expression of megakaryocyte-specific surface markers CD41 and CD42. Additionally, microscopic examination revealed significant changes in cell morphology, characterized by the appearance of larger cells with features resembling lobulated nuclei and proplatelets. The findings from these experiments preliminarily validate the feasibility of reprogramming suspension cells to megakaryocyte-like cells, thereby offering a novel strategy for the selection of initial cell populations suitable for large-scale in vitro platelet production via cellular reprogramming.

Due to their high programmability and versatility, DNA circuits have been widely used for molecular information processing at the nanoscale. The biocompatibility of DNA circuits allows them to integrate with other biological macromolecules, such as proteins and RNA, enabling biological signal regulation, drug delivery, and even disease diagnosis. The great potential of DNA circuits in molecular information processing has attracted widespread attention and intensive study. In this paper, we construct a three-level DNA cascading circuit based on combinatorial strand displacement to synergistically control the transmission of DNA signals. This strategy for DNA circuits provides a feasible idea for constructing more complex biological signal regulation systems, which is helpful for the development of biological detection, disease diagnosis, and drug delivery.

Single Nucleotide Polymorphism (SNP) is a common form of genetic variation, and genomic prediction is an emerging technique that utilizes SNP information to predict phenotypes in animals and plants. It is gradually being applied in animal and plant breeding as well as human disease risk assessment. Traditional statistical learning methods can only focus on linear interactions between the genome and phenotypes. Machine learning methods and deep learning methods have become popular due to their ability to recognize non-linear interactions between SNPs. However, existing deep learning methods often only learn short-distance interactions between SNPs and overlook long-distance interactions. Therefore, we propose a genomic prediction method called DCNNCSA (DualCNN Channel Spatial Attention) based on a channel spatial attention mechanism. DCNNCSA first uses a dual-branch convolutional neural network to extract features in the channel and spatial dimensions. Then, it utilizes a channel spatial attention mechanism to identify long-distance interactions between SNPs, thereby improving the accuracy of genomic prediction. Experimental evaluations are conducted on datasets including wheat 2000 and wheat 599 datasets. The results demonstrate that DCNNCSA outperforms rrBLUP, LASSO, XGBoost, Random Forest, DeepGS, and DLGWAS methods in terms of prediction accuracy.

In the rapidly evolving field of Ribonucleic Acid (RNA) secondary structure prediction, the integration of domain-specific knowledge with advanced machine learning techniques offers a promising pathway to enhance prediction accuracy, particularly in the context of limited quantitative high-quality data. This study introduces an innovative approach termed 'Structural Stability-Aware Deep Learning.' Initially, I developed an Energy Neural Network (EnergyNN) utilizing Convolutional Neural Networks (CNNs) and Graph Neural Networks (GCNs) to estimate the RNA's free energy from its sequence and secondary structure. An energy loss, calculated from both the real and predicted structures, was incorporated as an additional information into existing machine learning-based RNA secondary structure frameworks without altering their architectures. By leveraging the correlation between structural stability and free energy, this method integrates structural stability considerations into the machine learning model, significantly improving the accuracy, generalizability, and interpretability of RNA secondary structure predictions. I tested this integration on two different models, E2EFold and UFold. Rigorous testing against established datasets shows that my structural stability-aware model consistently outperforms the original E2EFold and UFold models in terms of prediction F1 score, precision, and recall. Moreover, this approach exhibits potential for broad applicability across various RNA types and machine learning methods, marking a significant advancement in the generalization of machine learning models for RNA research. This study not only progresses the field of RNA secondary structure prediction but also paves the way for future research that integrates domain knowledge into machine learning frameworks.

Lampreys play an important role in the ecological environment, although they may have some negative effects in some aspects, but overall, their existence is important in maintaining ecological balance and promoting biodiversity. This study first constructed the ecosystem and food web is constructed. Secondly, an improved Volterra predation model is established to adapt to multi-level predation relationships. Finally, through data fitting, a function model of the sex ratio of lampreys changing with the amount of food was constructed, and three sex ratios were divided (similar to males, more females than males, fewer females than males) to analyze the impact on the larger ecosystem when the population of lampreys could change its sex ratio.

Protein structure prediction is a crucial research field in life sciences and medicine, as well as a significant application area of artificial intelligence in scientific research. This study explores the structural characteristics and functional implications of mutation forms in the key protein causing cystic fibrosis—cystic fibrosis transmembrane conductance regulator (CFTR)—by employing the advanced protein structure prediction tool AlphaFold. The analysis particularly focuses on how mutated CFTR manifests in cystic fibrosis and predicts its three-dimensional structure using the AlphaFold model, providing a new perspective for understanding its pathological mechanisms. This paper highlights the feasibility of AlphaFold in predicting unique protein structures in the absence of homologous structural information and utilizes this technology to assess the effects of Trikafta therapy (comprising Elexacaftor, Tezacaftor, and Ivacaftor) on the structure and function of mutated CFTR.

By comparing the structures of normal, mutated, and Trikafta-treated CFTR models, the study reveals the impact of mutations on the protein's function and stability and how Trikafta effectively improves the folding and channel function of mutated CFTR. This research not only demonstrates the critical application of the AlphaFold model in modern drug design and disease understanding but also provides a scientific basis for the treatment strategies of cystic fibrosis, advancing the development of personalized medicine and precision treatment for such diseases.

The aim of this study was to screen and identify extracellular polysaccharide producing bacteria from waste tea residues. There are a variety of bacteria in waste tea residues, from which to find, isolate, screen the ability to produce extracellular polysaccharides a strain of TB16 (Klebsiella pnenmoniae) bacteria with the ability to produce a good biological activity of extracellular polysaccharides and to analyse its genetic background, from the genomic perspective of the analysis of the genome is indispensable, the genome of the means to help us to find out that the genetic information, the function of the genes, genome structure and other aspects of information mining, help us to establish a system of biological evolution and phylogenetic studies. By sequencing the whole genome of this bacterium, TB16 belongs to the genus Klebsiella pneumoniae, which is a group of bacteria widely found in nature with a certain degree of pathogenicity and antibiotic resistance. The bacterium adapts to the host microenvironment in a variety of ways, and may acquire resistance genes through phage-level gene transfer. It was found that this bacterium may also play an important role in biogeochemical cycles.

Overall, whole genome sequencing and analysis can provide basic data support for further research on the diversity, environmental impact, material degradation and infection prevention and control of this bacterium, which will help to deepen the understanding and control of this bacterium.

Simulating the microenvironment in vivo and constructing a neural network closer to it is of great significance to the research of the central nervous system (CNS) injury repair. Based on the above requirements, a microfluidic device for three-dimensional (3D) cell culture has been designed and verified by double-layer lithography in this paper. The device consisted of two cell chambers connected by 100 parallel microchannels which were 8 μm wide and 8 μm deep. The results of numerical simulation showed that the microfluid flow in the device was stable and uniform. 3D culture of neural stem cells (NSCs) was achieved in the device, and identified by fluorescent staining that the differentiated neuronal cell bodies were physically isolated from neurites. The device provides a dynamic, stable, and uniform microenvironment for cell culture in vitro. Together, the microfluidic device provides an important technical platform for further research on cell-cell interaction, especially axonal injury regeneration and repair.

Anaerobic ammonia oxidation (anammox) is an efficient biological technology for nitrogen removal. Hydrazine synthase synthesizes the intermediate hydrazine as one of the key enzymes in the metabolism of anammox bacteria. The genomic DNA of anammox bacteria was extracted as a template. Sequence homology and phylogenetic tree analysis showed that the clone gene was closely related to the cloned hydrazine synthase A-subunit (hzsA) from an uncultured anaerobic ammonium-oxidizing bacterium. The full-length cDNA of the hzsA gene was 1331 bp and contained a 762 bp open reading frames (ORFs) encoding 253 amino acids protein. The deduced protein molecular weight was 28.00 kD and its theoretical isoelectric point was 7.62. Its instability coefficient was 31.91, which meant it was a stable protein, and its average hydrophilicity was -0.563, which meant it was a hydrophilic protein. And it was located in the cytoplasm without transmembrane domain and signal peptide. The secondary structure of hydrazine synthase A-subunit protein was composed of four types, the most of which was random coil, accounting for 52.96%, followed by extended strand (32.02%), beta turn (11.46%) and alpha helix (3.56%). The three-dimensional structure of the protein has been successfully modelled. There were no conserved domains in the protein.

To fill the existing knowledge gap, we analyzed the expression patterns of long non-coding RNAs (lncRNAs) in peripheral blood mononuclear cells (PBMCs) from individuals with mild and severe EV71 infections, as well as from healthy controls. Our study identified 8,541 lncRNAs that were significantly differentially expressed. We employed a comprehensive bioinformatics approach, including serial testing clustering (STC) algorithm, gene ontology analysis, pathway analysis, and lncRNA-mRNA co-expression network analysis, to explore the functional roles and regulatory networks of these lncRNAs. These analyses revealed key pathways and potential roles of lncRNAs in immune and inflammatory responses during EV71 infection. Furthermore, we identified and validated 10 crucial lncRNAs with significant expression changes among patients with mild and severe HFMD and healthy controls. Our findings provide novel insights into the involvement of lncRNAs in EV71-induced HFMD, which are essential for early diagnosis and treatment. Additionally, they offer valuable clues for understanding the mechanisms behind EV71-induced immune and inflammatory responses and the pathogenesis of inflammatory imbalance in severe EV71 infection.

Inframe insertion/deletion (indel) mutations play a crucial role in human genetic variation and are closely associated with various diseases. Currently, most methods for predicting pathogenic inframe indel mutations are based on machine learning, utilizing manually engineered features for prediction, which may overlook certain mutation data. Moreover, existing deep learning methods mainly focus on protein sequence features, neglecting other aspects. This paper proposes a method for predicting pathogenic inframe indel mutations, termed DPPred-indel, based on a combination of a biological language model and feature fusion. DPPred-indel leverages transfer learning to address the limited data availability and integrates features from both protein and DNA sequence levels for prediction. Subsequent experiments validate the effectiveness of transfer learning and feature fusion. DPPred-indel demonstrates comparable performance to state-of-the-art methods on two independent test sets constructed in this study.

Prediction of protein-ligand binding affinity holds significant guidance value in fields such as drug development and protein molecular design. This study proposes a novel deep learning model, ResSepDTAF, for predicting protein-ligand binding affinity. The model integrates depthwise separable convolutions and residual modules, employing supervised learning for regression prediction. We utilize the raw feature vectors of proteins and small molecules, processed through convolutional and fully connected layers to predict affinity values. Experimental results show that ResSepDTAF achieves an RMSE of 1.209, significantly lower than Pafnucy (1.418), DeepDTAF (1.355), and DeepAtom (1.318), indicating higher prediction accuracy. Additionally, the correlation coefficient R for ResSepDTAF is improved by 17.9%, 15.8%, and 13.2% over Pafnucy, DeepDTAF, and DeepAtom, respectively. Other metrics such as MAE, SD, and CI also outperform those of other models. Overall, ResSepDTAF demonstrates excellent performance in predicting protein-ligand binding affinity and shows high potential for practical applications.

The ability of compounds to pass through the blood-brain barrier is an important factor in drug development related to the central nervous system. Therefore, predicting the blood-brain barrier permeability of compounds at high throughput and providing appropriate candidate compounds are crucial for the development of related drugs. Although traditional experimental methods can also predict the blood-brain barrier permeability of compounds, they are costly and have long time cycle. To assist in related research, this article proposes a neural network model using deep learning algorithm to complete the task of predicting blood-brain barrier permeability of compounds, and names it DL_BBBP. In DL_BBBP, various compounds are characterized using molecular graphs and MACCS molecular fingerprints. Specifically, we conducted feature complementarity processing on MACCS, removed information about molecular substructures to prevent duplication and omission. By extracting features from the MACCS molecular fingerprints and molecular graphs of the compounds, we predict the blood-brain barrier permeability of the compounds and compare the results with some current deep learning models and machine learning methods. The validation results verify that the model’s performance is better than state of the art models, and the prediction of the blood-brain barrier permeability of the compounds is accurate and effective. Therefore, it is believed that this model has great potential in the field of drug development.

Significant emphasis has been placed on advancing therapeutic interventions and medicines to treat Alzheimer's disease, the leading cause of dementia. BACE1 inhibitors have shown considerable promise in reducing amyloid-β levels in the brain and potentially halting the progression of Alzheimer disease. However, human clinical trials are fraught with risk and exorbitant cost. In addressing these challenges, this investigation has developed a deep learning model for the prediction of interactions between BACE1 inhibitors and ligand. The model represents compounds as molecular graphs and SMILES strings, which are then processed through Graph Neural Network and Convolutional Neural Network channels, respectively. This approach allows comprehensive prediction of IC50 values and classification of compound activity with the BACE1 inhibitor. The model can be used as a convenient tool for the development of BACE1 inhibitors and also for virtual screening of molecules to identify potential inhibitors.

Background: The structure of circular ribonucleic acids (circRNAs) is a distinct closed loop. As tumor inhibitors or oncogenes, circRNAs play the important roles in the development of tumors. Previous studies indicate that circRNAs can influence the occurrence and development of various diseases. Many studies have found that circRNAs may play a crucial role in the carcinogenic effects of different types of cancer. Methods: Here, we applied a bioinformatics analysis to explore the potential regulation of circRNAs in prostate cancer (PCa). The differentially expressed (DE) RNAs (DE-circRNAs and DE-mRNAs) were identified based on GEO and TCGA data analyses. And the tumor immune microenvironment of PCa was also analysed. Results: 201 DE-mRNAs and 607 DE-circRNAs were identified, and the regulation network was constructed. Some genes regulated by multiple circRNAs, such as DPYS and KIF4A are associated with survival and prognosis. Conclusion: This study may elucidate the potential mechanisms underlying the pathogenesis of PCa and may help serve new biomarkers or targets for PCa.

Long-term neocoronitis remains a complex and not fully understood medical condition, manifesting with persistent symptoms such as fatigue, cognitive problems and joint discomfort long after the acute phase of neocoronitis has passed. Salvia miltiorrhiza is an important element of Chinese medicine that has shown promise in alleviating these persistent symptoms. However, the biochemical pathways by which it exerts its therapeutic effects have not been clearly defined. In this study, we investigated the possible molecular pathways by which danshen alleviates prolonged coronavirus symptoms using computational biology techniques. We analyzed the protein targets activated by metabolites formed after salvia ingestion and identified potential therapeutic targets for prolonged coronavirus COVID. By constructing a protein-protein interaction network, performing gene ontology, and enrichment analysis of cross genes by Kyoto Gene and Genome Encyclopedia, we identified key targets and biological pathways. This study reveals the underlying biochemical mechanisms by which Salvia divinorum exerts its therapeutic effects on long coronaviruses and provides a basis for the development of more targeted therapeutic strategies against this complex disease.

With the improvement of people's quality of life and the modern sub-healthy lifestyle, the proportion of diseases including diabetes, obesity, hypertension, hyperlipidemia, and coronary heart disease caused by them has significantly increased in the population. GIP and GLP-1 and are both important targets for the treatment of TIID and obesity and have attracted increasing attention in both basic and clinical research. This article mainly discusses the mechanisms of GLP-1 and GIP, two intestinal hormones, in TIID, obesity, and the immune system, as well as their roles in metabolism. It also provides an overview of the clinical effects of related drugs, such as semaglutide and dual receptor agonists. This article has guiding significance for future basic research and clinical medication and underscores the importance of comprehensive research in GLP pharmacology to optimize GLP-targeted drug development and clinical outcomes.

Sophisticated prosthetic devices use classification algorithms to recognize electromyography (EMG) signal variations, accurately identifying user-intended motion gestures. Achieving high classification accuracy is crucial for the real-time control of prosthetic limbs. In this paper, we introduce an extended version of common spatial pattern (CSP) method by incorporating time delay embedding to extract spatial and temporal information into the classifier on myoelectric control performance. Nineteen able-bodied subjects in the experiment, and the experiment data from eight classes are used to classify based on the new spatiotemporal domain space proposed. The performance of the proposed approach surpasses that of traditional methods relying on time-domain features and autoregressive model and achieves a 10% increase in average classification accuracy (p < 0.05).

Lung cancer is the highest mortality as well as prevalence of tumor-related diseases in the world, and there is no very effective way for early screening of lung cancer, so finding an accurate biomarker is necessary for early diagnosis of lung cancer. In this research, the MPC1 gene expression level in a variety of tumors was analyzed by bio-informatics analysis, and prognostic analysis was performed, focusing on lung cancer patients to check the diagnostic accuracy of ROC for MPC1, followed by KEGG and GO analysis to find out the biological processes that MPC1 may be involved in, and then immune infiltration analysis to further confirm some of the possible MPC1 functions, and finally, MPC1 related proteins were searched by SRTING protein interaction analysis. The analysis results revealed that MPC1 expression was downregulated in a variety of cancers, and in lung cancer, MPC1 expression was significantly down-regulated, and the significance reached p < 0.005, with a high diagnostic accuracy of AUC: 0.828 for ROC, and the expression level of MPC1 was significantly down-regulated in patients who were smokers, and was correlated with the age of cigarette smoking, age, and the degree of malignancy of the tumors. By immune infiltration analysis, it was found that in lung cancer, the expression of MPC1 was positively correlated with a variety of immune cells, among which, the strongest correlation was with Follicular helper T cell (TFH), R=0.291, followed by enosinophils, R=0.281, these results indicated that the low expression of MPC1 was probably related to the immune escape of the tumor Related. Finally, the nine proteins with the highest correlation with MPC1 were found to be related to inflammatory factors. These results suggest that MPC1 may be a new direction for tumor immunity, and restoring the body's tumor immunity by promoting the expression of MPC1 will make MPC1 a new marker for lung cancer and a new target for lung cancer treatment.

Objective: To assess combined validation of tuberculosis (TB) clinical etiological diagnostic items in phenotypical, immunological, and molecular biological technologies. Method: A comprehensive study is conducted on the 'Chinese core journal database', covering phenotypical, immunological, and molecular biological clinical etiological diagnostic technologies of TB. Retrieved papers are screened, extracted, and systematically evaluated for quality. Meta-analysis is performed on qualified papers. The combined effect values such as combined sensitivity (SEN), combined specificity (SPE), combined positive likelihood ratio (+LR), combined negative likelihood ratio (-LR), combined diagnostic odds ratio (DOR), combined summary receiver operating characteristic area under the curve (SROC AUC) are calculated. Result: A total of 271 relevant papers are retrieved, and 49 qualified papers are included, including 54 TB phenotypical diagnostic technological papers, 39 TB immunological diagnostic technological papers, and 25 TB molecular biological diagnostic technological papers. Combined SEN, SPE, +LR, -LR, DOR, SROC AUC are 0.47 (0.41-0.53), 0.99 (0.98- 0.99), 40.40 (21.80-75.00), 0.53 (0.48-0.60), 76.00 (40.00-144.00), 0.85 (0.82-0.88) in phenotypical diagnostic technology; 0.72 (0.65-0.78), 0.89 (0.85-0.93), 6.70 (4.60-9.80), 0.31 (0.25-0.39), 22.00 (13.00-36.00), 0.88 (0.85-0.91) in immunological diagnostic technology; 0.78 (0.69-0.85), 0.97 (0.95-0.98), 26.10 (15.10-44.80), 0.23 (0.16-0.32), 115.00 (62.00-210.00), 0.96 (0.94-0.97) in molecular biological diagnostic technology. Conclusion: TB phenotypical diagnostic technology has a high combined SPE; TB molecular biological diagnostic technology has a high combined SEN and validation. The combined validation of TB immunological diagnostic technology is between the abovementioned technologies. Appropriate clinical etiological diagnostic technology could be selected based on TB diagnostic laboratories' different levels and functions.

This study employed network pharmacological analysis and molecular docking methods to identify and confirm the likely molecular targets and mechanisms through which pterostilbene may exert its effects on schizophrenia. Swiss Target Prediction and Gene Cards databases were used to retrieve targets related to astragalus and schizophrenia, and Venny was used to construct an intersection map of astragalus targets and schizophrenia targets. Protein-protein interactions of common genes were obtained from the STRING database to construct a PPI network map for Pterostilbene -Schizophrenia. Utilizing the Metascape platform, an enrichment analysis of GO and KEGG pathways was conducted to delineate the potential targets associated with both Pterostilbene and schizophrenia. Subsequently, the selected targets were subjected to molecular docking and graphical representation through the Auto Dock software. Pterostilbene has 40 relevant targets that can act on schizophrenia, and the core targets involve BCL2, ESR1, EGFR, MTOR, etc. In the GO analysis, the MF enrichment results showed that the treatment of schizophrenia by pterostilbene is closely related to enzyme binding, and the KEGG pathway enrichment results showed that the TNF signaling pathway has a significant correlation with the antischizophrenic effect of pterostilbene. Analysis of the KEGG pathway enrichment data revealed significant correlations between the TNF and thyroid hormone signaling pathways and the efficacy of pterostilbene in countering schizophrenia. Molecular docking analysis indicated stable interactions between pterostilbene and its respective target. This study suggests that pterostilbene exerts antipsychotic effects through multiple targets and pathways, providing a target drug for the treatment of schizophrenia.

Objective: This study explores the correlations among tongue appearance, tongue coating microbiota, and clinical symptoms across different stages of chronic gastritis. Methods: We collected 133 samples from patients categorized into inflammation (53), atrophy (31), and intestinal metaplasia (49) groups. Clinical data were analyzed using baseline, Mental, and MaAsLin statistical methods to investigate the relationships between tongue characteristics, microbiota, and symptoms. Results: (1) Tongue and Symptom Correlation: Progression in gastritis was associated with a decrease in pale red tongues and an increase in dark-colored, red, dark red, purple, large, and greasy tongues. Although dominant bacterial genera were consistent across groups, unique genera varied significantly, with the most diversity in the intestinal metaplasia group. (2) Microbiota Variability: Statistical analysis showed no significant variations in Hp infection rates among the groups. However, specific microbiota were correlated with particular symptoms; for instance, greasy and slippery coatings were linked with positive correlations to Cyanobacteria and Chloroflexi. (3) Symptom Correlations: Stomach pain and bloating were positively correlated with Firmicutes and negatively with Bacteroidetes. Oral symptoms like bitterness were more common in the atrophy and metaplasia groups, significantly differing from the inflammation group. Conclusion: Changes in tongue appearance and microbiota across different gastritis stages mirror clinical symptoms, suggesting a diagnostic link. These findings could aid in developing precise diagnostic tools and early warning systems for chronic gastritis based on microbiome data.

Research on the core parameter detection technology of PET/MR based on YY/T 1719—2023 "General requirements of positron emission tomography and magnetic resonance equipment", which is the first industry standard related to PET/MR quality control in China. Taking uPMR 790 PET/MR as an example, using models such as the ACR MRI phantom and the NEMA IEC body phantom, performance tests and data analysis were conducted on PET/MR parameters such as signalto- noise ratio, uniformity, spatial resolution, and registration accuracy and the testing methods were explored and studied. The results obtained are all within the standard requirements and the normal range provided by the manufacturer, the signal-to-noise ratio of the MR system is 779.23, with uniformity of 91.04%, relative error of slice thickness of 4.02%, maximum geometric distortion of 0.40 mm, and visible spokes of 34 in low contrast; the tomography sensitivity of the PET system is 17.8 cps/kBq, with scattering fraction of 44%, peak NECR132.7 kcps, and maximum registration accuracy deviation of 1.4 mm. The testing methods used in YY/T 1719—2023 are mostly commonly used and mature technologies both domestically and internationally, which are executable. However, some of the methods used for certain parameters have some lag compared to international standards and have a certain level of complexity in practical operation.

This study aims to develop an efficient and reliable chronic disease monitoring platform using RT-Thread real-time operating system. The system software integrates the three modules of EEG data annotation, large language model and intelligent family doctor to achieve real-time collection, processing and remote transmission of physiological data of patients with chronic diseases, as well as continuous monitoring of sleep data, supporting medical professionals to carry out effective health management and decision-making. The hardware of the system includes five parts: embedded development board module, WiFi RF transceiver module, blood oxygen saturation monitoring module, EEG acquisition module and Bluetooth communication module. Through the integration of advanced technology, the system realizes the real-time and reliability of chronic disease management, improves the efficiency of patient health management, reduces the work burden of medical staff, and has a wide application prospect. The test results show that the SalientSleepNet model, semi-supervised learning training method and sensor selected in this work are feasible.

With the advancement of computer technology and artificial intelligence, the significance of clinical decision support systems in the medical field has escalated. These systems, leveraging deep mining of multimodal data and feature fusion, furnish clinicians with precise diagnostic insights, significantly augmenting diagnostic efficiency and accuracy. This study, centered on multimodal inference, has crafted an intelligent diagnostic decision model, validated its efficacy through experimentation. It delves into the critical facets of technology adoption and application, drawing from rational behavior theory, planned behavior theory, and the technology acceptance model. Additionally, a meticulous analysis of multimodal fusion methods geared towards features, decisions, and processes has been conducted, laying a robust foundation for future investigations. The experimental segment has subjected algorithmic performance to comprehensive testing and assessment, revealing the pronounced advantage of multimodal fusion methods in enhancing diagnostic decision accuracy. This research proffers novel avenues and methodologies to propel the advancement of intelligent healthcare.

Centrifugal microfluidic chips have excellent automation potential. Various detection technologies have been integrated into centrifugal microfluidic chips. To realize multi-step reaction processes, centrifugal microfluidic chips require the integration of several valves. Valve systems using only capillary valves may lead to excessively high rotational speeds. This is undesirable for manufacturing portable, low-power centrifugal microfluidic chip mating devices. Through valve mating and chamber layout, a valve system with low rotational speed characteristics was designed to overcome the dependence of centrifugal microfluidic chips on high rotational speed. The system includes 2 hydrophobic valves, 2 siphon valves and a Coriolis force structure. This study simulated the parametrically designed valve and Coriolis force structure by COMSOL and obtained the liquid control effect as expected. Using fluid simulation, the development cycle of the centrifugal microfluidic chip was shortened, saving time and expense. Through experiments, this study proved that our centrifugal microfluidic chip can accomplish 4-step reaction and 7-step pipetting control of 5 reagents. It has consistency and stability for the distribution of products.

Drug combination therapy is essential in cancer treatment, significantly enhancing the survival rates of cancer patients. In previous studies, researchers have investigated various methods to predict drug synergy, such as utilizing drug chemical information and protein-protein interaction network (PPI) to obtain features of both drugs and cancer cell lines. However, there has been comparatively little research on the information regarding drug molecular structures and the associations between cell target proteins and cancer cell line genes. Therefore, this paper sets out to investigate this aspect. We propose a model that utilizes a three-layer Graph Isomorphism Network (GIN) to capture the structural information inherent in the SMILES representations of drug molecules. Additionally, we employ a graph attention networks (GAT) to learn features of PPI network. Finally, we integrate drug features, cancer cell line target protein features, and gene features, and utilize XGBoost model to predict the synergistic effects between drugs. We evaluate our model on the DrugComDB and Oncology-Screen datasets, and experimental results demonstrate its superior performance. On the DrugComDB dataset, our model achieves F1, Recall, and ACC scores of 0.738, 0.718, and 0.768, respectively. Similarly, on the Oncology- Screen dataset, our model demonstrates F1, Recall, and ACC scores of 0.810, 0.825, and 0.790, respectively. Our model outperforms several established models across all metrics. We believe it can effectively identify combinations of anticancer drugs.

The application of Internet of Things (IoT) technology promotes refined management in various fields. This paper proposes an IoT-based medical waste management and tracking system to implement closed-loop digital management of medical waste throughout its entire life cycle, from generation, transportation, storage to disposal. The system visualizes key stages of medical waste collection, storage, and transportation across multiple hospital campuses, enabling real-time data transmission and traceability. This unified and standardized digital management process improves the overall operational efficiency and homogeneity of medical waste management across multiple campuses, facilitating resource optimization and enhancing service quality and management level. Ultimately, the system aims to better serve patients.

In Chinese acupuncture, moxibustion of the abdomen can energize the meridians, dissipate cold and relieve pain, promote water retention and reduce swelling, and warm the kidney and yang. The traditional abdominal acupoint identification, which mainly relies on the experience of doctors, is highly subjective and uncertain. To crack this problem, this paper proposes an abdominal meridian detection and abdominal acupoint location method based on geometric topological relations. First, we use MediaPipe to extract 33 human skeletal key points and YOLOv8 to detect the left and right nipples and navel, which are used as the reference points for detecting meridians and acupoints. Then the geometric topological relationships of the meridians and acupoints are established through the human body reference points, so as to determine the major meridians in the abdomen, such as the Conception Vessel Meridian, the Spleen Meridian, the Stomach Meridian, and the Sidney Meridian. Finally, under the guidance of the meridians, each acupoint in the abdomen was determined by the bone degree and cun method. The experimental results have showed that the method was able to accurately identify and localize the major 35 abdominal acupoints with an average error of less than 3 mm.

In order to further improve the accuracy and measurement efficiency of the plethysmography method, this paper proposes a genetic algorithm-based optimized back-propagation (GA-BP) neural network method for the correction of the pressure inside the box of the plethysmography box. In this paper, we first analyze how the temperature affects the accuracy and measurement efficiency of plethysmography testing box results, and give the expression of the relationship between the local temperature inside the plethysmography testing box and the pressure inside the box, through the correction of the measurement pressure, the temperature influence of the human body into the plethysmography testing box can be reduced, so as to improve the accuracy of the measurement results. By setting the appropriate temperature detection points, establishing the BP neural network pressure prediction model based on the optimization of genetic algorithm, using the model to predict the pressure changes of different human bodies entering the box, and using the prediction to correct the effect of human body temperature on the pressure inside the box in real time during the measurement process. Finally, the results of simulation and experimental data verified its superiority over the traditional method in terms of accuracy and efficiency performance, which has an important application value in clinical lung function testing.

Nasopharyngeal carcinoma (NPC) is a common type of malignant tumor in the head and neck, for which treatment and prognostic evaluation are critically important for patient management. Traditional prognostic methods often rely on limited clinical indicators and fail to fully reveal the complex pathophysiological status of NPC patients. This study introduces a network analysis-based feature selection method for predicting the overall survival (OS) prognosis of NPC. Utilizing clinical data from 792 newly diagnosed NPC patients, a co-occurrence matrix was constructed. By calculating centrality scores, key features closely related to OS prognosis were identified. Through an iterative feature selection process, starting with the structures having the highest centrality scores, the number of features was progressively increased to assess prognostic performance on the training set, culminating in the identification of the top 14 features as criteria for stratifying high-risk populations. On an independent test set, patients were divided into high- and low-risk groups based on these 14 features, demonstrating a significant difference in prognosis between the two groups (p<0.05). These results not only validate the accuracy and effectiveness of the selected features but also affirm the powerful potential of network analysis techniques in constructing medical prognostic models. This study provides a new tool for the prognostic evaluation of NPC, aiding doctors in accurately assessing patient outcomes and offering scientific support for personalized treatment plans.

Last few years, numerous related reaserches have demonstrated that microRNAs (miRNAs) influenced the evolution and advancement of intricate human illnesses, so observing associations of miRNA-disease can contribute to exploring and treating these diseases. However, based on biological experiments, it is frequently observed that the experimentation process tends to be costly and time-intensive, often performed on a small-scale. Therefore, the development of multiple algorithms plays an important role in predicting the potential association of mirnas with disease. In the research, we presented a framework of computation that combined autoencoder with extreme gradient boosting to predict unknown miRNAdisease association (AEXGB). Firstly, the similarity between miRNA and various diseases is synthesized, and representative disease similarity and miRNA similarity can be constructed, respectively. In addition, disease similarity and miRNA similarity can be combined, and the original features of mirNA-disease pairs can be constructed. These features were send as in put to autoencoder (AE) for extracting hidden biological patterns. Unverified associations dependent on IRNA-disease for extracted deep features can be inferred by extreme gradient enhancement (XGBoost).

To enable methods to be evaluated, case studies and cross-validation experiments can be used, and the effectiveness of AEXGB in observing potential mirNA-disease associations can be observed.

Objective To explore the effect of positive psychological intervention based on PERMA model on the subjective wellbeing of nursing undergraduates. Methods Convenience sampling method was used to divide 72 sophomore nursing students into an intervention group of 36 people and a control group of 36 people. The control group participated in the routine daily activities of college students, and the intervention group received eight consecutive positive psychological workshops (interviews and exercises) based on the PERMA model. The subjective well-being index scale (IWB) was used to evaluate the effect before the intervention, at the end of the intervention, and one month after the end of the intervention. Results At the end of the intervention and one month after the end of the intervention, the total score of subjective wellbeing, emotional index, and life satisfaction scores of the intervention group were higher than those of the control group, and the difference was statistically significant (P<0.05). The results of repeated measures analysis of variance showed that the main effect of the intervention and the time effect of the total subjective well-being score and emotional index, and life satisfaction were statistically significant (P<0.05), and the interaction effect of the total subjective well-being score and emotional index was statistically significant (P<0.05). Scientific significance (P<0.05) Conclusion The positive psychological intervention based on the PERMA model can improve the subjective well-being of nursing undergraduates.

Prostate cancer is not easy to be detected, and the current detection methods have the shortcomings of low detection limit and insufficient sensitivity, so it is urgent to prepare an efficient and sensitive sensor. In this paper, PCN(Hf)-222-24 was used as a luminescent probe and combined with immunoaffinity and boric acid nucleophilicity to prepare a sensitive and simple prostate antigen sensor. PCN(Hf)-222 has a large surface area and excellent ECL emission and is used as a boric acid nucleophile receptor for specific detection of prostate antigen. The results show that the prepared electrode has high stability, excellent detection ability, satisfactory selectivity and repeatability, and the detection limit is 0.21 pg/mL, which has better performance than other types of sensors.

Real-time home healthcare, enabled by future sixth-generation (6G) communication technology, offers a promising application for advanced clinical gait analysis. Utilization of dynamic sensors in clinical gait analysis presents a definitive approach for quantifying variances in gait patterns between patient cohorts and healthy populations. However, variations in demographic and anthropometric parameters, including gender, age, height, and Body Mass Index, among these groups, may result in inaccurate clinical evaluations. This study introduces an innovative clustering methodology that incorporates custom-weighted distances based on demographic and anthropometric parameters. This approach is designed to establish a dynamic, normative gait framework pertinent to healthy individuals. It facilitates the classification of individuals into distinct categories grounded on their demographic and anthropometric parameters, enabling the provision of a customized normative gait model appropriate for each category.

In the context of the ongoing progress of modern technology, research into Traditional Chinese Medicine (TCM) is being deepened. Advances in modern pharmacology and molecular biology are progressively uncovering the mechanisms of action, efficacy principles, and predictive effects of the components of TCM. Faced with the complexity of TCM components and the intricacies of their mechanisms of action, the traditional compound-target relationship model has limitations in its predictive capabilities. At present, constructing complex heterogeneous graph networks and applying machine learning or deep learning for prediction have become a trend. This paper introduces a novel prediction method based on the efficacy-herb-target-pathway network, with the innovation of incorporating the Metapath2vec. This algorithm trains the model on a heterogeneous graph using manually defined metapaths, capturing the complex relationships within the network more effectively than the traditional node2vec algorithm. In addition, we have developed a custom prediction module based on the transformer architecture, which significantly enhances the accuracy of the predictions. Our method has demonstrated outstanding performance in terms of AUC_ROC, AUC_PR, and F1 evaluation metrics, as evidenced by testing on the collected dataset. This approach not only enhances the accuracy of predictions but also offers a new perspective and tool for predicting TCM targets, thereby adding more practical value to the development of traditional Chinese medicine. MT-HTI is freely available at https://github.comShiLab-GitHub/MT-HTI.