Early diagnosis and precise treatment of gastrointestinal (GI) diseases are crucial for reducing mortality and improving quality of life. In this context, the detection and classification of abnormalities in endoscopic images is an important support for specialists during the diagnostic process. In this study, an innovative deep learning approach for the segmentation and classification of pathological regions in the GI system is presented. In the first phase of the study, a novel segmentation network called GISegNet was developed. GISegNet is a deep learning-based architecture tailored for accurate detection of anomalies in the GI system. Experiments conducted on the Kvasir dataset showed that GISegNet achieved excellent results on performance metrics such as Jaccard and Dice coefficients and outperformed other segmentation models with a higher accuracy rate (93.16%). In the second phase, a hybrid deep learning method was proposed for classifying anomalies in the GI system. The features extracted from the transformer-based models were fused and optimized using the Minimum Redundancy Maximum Relevance (mRMR) algorithm. The classification process was performed using Support Vector Machines (SVM). As a result of feature fusion and selection, the second model, which combined features from DeiT and ViT models, achieved the best performance with an accuracy rate of 95.2%. By selecting a subset of 300 features optimized by the mRMR algorithm, the accuracy (95.3%) was maintained while optimizing the computational cost. These results show that the proposed deep learning approaches can serve as reliable tools for the detection and classification of diseases of the GI system.

Conventional manual or feature-based ECG analysis methods are limited by time inefficiencies and human error. This study explores the potential of transforming 1D signals into 2D Gramian Angular Field (GAF) images for improved classification of four ECG categories: Atrial Fibrillation (AFib), Left Ventricular Hypertrophy (LVH), Right Ventricular Hypertrophy (RVH), and Normal ECG.•The study employed GAF transformations to convert 1D ECG signals into 2D representations at three resolutions: 5000 × 5000, 512 × 512, and 256 × 256 pixels.•Segmentation methods were applied to enhance feature localization.•The ConvNext deep learning model, optimized for image classification, was used to evaluate the transformed ECG images, with performance assessed through accuracy, precision, recall, and F1-score metrics.The 512 × 512 resolution achieved the optimal balance between computational efficiency and accuracy. F1-score for AFib, LVH, RVH and Normal ECG were 0.781, 0.71, 0.521 and 0.792 respectively. Segmentation methods improved classification performance, especially in detecting conditions like LVH and RVH. The 5000 × 5000 resolution offered the highest accuracy but was computationally intensive, whereas the 256 × 256 resolution showed reduced accuracy due to loss details.

Evaluating the plant risk of soil pollution by plant physiological indices usually requires a long cycle and has significant uncertainty. In this study, oxidative potential (OP) of the in situ heavy metal contaminated soils was measured by the dithiothreitol method. The oxidative stress response of the model plant ryegrass (Lolium perenne L.) induced by heavy metal contaminated soil was evaluated by the biomarkers, including superoxide dismutase and total antioxidant capacity. The comprehensive biomarker response index has a significant exponential correlation with the OP of soil (r ​= ​0.923, p ​< ​0.01) in ryegrass. Metabolomics analysis also showed a significant relationship of the metabolic effect level index of amino acids and sugars with OP. Random forest was selected from four machine learning models to screen the metabolites most relevant to OP, and Shapley additive explanations analysis was used to explain the contribution and the influence direction of the features on the model. Based on the selected 20 metabolites, the metabolic pathways most related to OP in plants, including alkaloid synthesis and amino acids metabolism, were identified. Compared to the plant physiological indices, OP is a more stable and faster indicator for the plant risk assessment of heavy metals contaminated soil.

Breast cancer is the most common cancer in women globally, imposing a significant burden on global public health due to high death rates. Data from the World Health Organization show an alarming annual incidence of nearly 2.3 million new cases, drawing the attention of patients, healthcare professionals, and governments alike. Through the examination of histopathological pictures, this study aims to revolutionize the early and precise identification of breast cancer by utilizing the capabilities of a deep convolutional neural network (CNN)-based model. The model's performance is improved by including numerous classifiers, including support vector machine (SVM), decision tree, and K-nearest neighbors (KNN), using transfer learning techniques. The studies include evaluating two separate feature vectors, one with and one without principal component analysis (PCA). Extensive comparisons are made to measure the model's performance against current deep learning models, including critical metrics such as false positive rate, true positive rate, accuracy, precision, and recall. The data show that the SVM algorithm with PCA features achieves excellent speed and accuracy, with an amazing accuracy of 99.5%. Furthermore, although being somewhat slower than SVM, the decision tree model has the greatest accuracy of 99.4% without PCA. This study suggests a viable strategy for improving early breast cancer diagnosis, opening the path for more effective healthcare treatments and better patient outcomes.

Oral cavity carcinomas (OCCs) represent roughly 50% of all head and neck cancers. The risk of occult neck metastases for early-stage OCCs ranges from 15% to 35%, hence the need to develop tools that can support the diagnosis of detecting these neck metastases. Machine learning and radiomic features are emerging as effective tools in this field. Thus, the aim of this study is to demonstrate the effectiveness of radiomic features to predict the risk of occult neck metastases in early-stage (T1-T2/N0) OCCs.

Retrospective study.

A single-institution analysis (Maxillo-facial Surgery Unit, University of Naples Federico II).

A retrospective analysis was conducted on 75 patients surgically treated for early-stage OCC. For all patients, data regarding TNM, in particular pN status after the histopathological examination, have been obtained and the analysis of radiomic features from MRI has been extrapolated.

56 patients confirmed N0 status after surgery, while 19 resulted in pN+. The radiomic features, extracted by a machine-learning algorithm, exhibited the ability to preoperatively discriminate occult neck metastases with a sensitivity of 78%, specificity of 83%, an AUC of 86%, accuracy of 80%, and a positive predictive value (PPV) of 63%.

Our results seem to confirm that radiomic features, extracted by machine learning methods, are effective tools in detecting occult neck metastases in early-stage OCCs. The clinical relevance of this study is that radiomics could be used routinely as a preoperative tool to support diagnosis and to help surgeons in the surgical decision-making process, particularly regarding surgical indications for neck lymph node treatment.

Lung cancer is one of the most prevalent diseases affecting people and is a main factor in the rising death rate. Recently, Machine Learning (ML) and Deep Learning (DL) techniques have been utilized to detect and classify various types of cancer, including lung cancer. In this research, a DL model, specifically a Convolutional Neural Network (CNN), is proposed to classify lung cancer stages for two types of lung cancer (LUAD and LUSC) using a gene dataset. Evaluating and validating the performance of the proposed model required addressing some common challenges in gene datasets, such as class imbalance and overfitting, due to the low number of samples and the high number of features. These issues were mitigated by deeply analyzing the gene dataset and lung cancer stages from a medical perspective, along with extensive research and experiments. As a result, the optimized CNN model using F-test feature selection method, achieved high classification accuracies of approximately 93.94% for LUAD and 88.42% for LUSC.

Non-alcoholic steatohepatitis (NASH), a prominent driver of hepatocellular carcinoma (HCC) besides virus and alcohol, induces a series of complex liver structural and immune microenvironment changes, which make the early diagnosis and treatment of NASH-associated HCC (NASH-HCC) more challenging. This study aims to identify signature genes and explore the role of immune cell infiltration in NASH-HCC to improve early detection and prognosis assessment.

Differential gene and immune cell infiltration are important indicators for predicting the progress of oncology and responsiveness of tumor patients to immunotherapy, usually confirmed through biopsy tests with poor patient compliance. To obtain a highly correlated signature gene set and validate immune cell infiltration status, the GSE164760 and GSE102079 datasets from the Gene Expression Omnibus (GEO) database were analyzed using machine learning algorithms. Feature genes were identified based on differentially expressed genes and key modular genes identified by weighted gene co-expression network analysis (WGCNA). The signature genes were screened using the least absolute shrinkage and selection operator (LASSO), random forest, and support vector machine recursive feature elimination (SVM-RFE) machine learning algorithms. Subsequently, the signature genes were subjected to diagnostic efficacy tests, gene set enrichment analysis, immune cell infiltration assessment and real-time reverse transcription polymerase chain reaction (RT-qPCR) validation.

Six signature genes were identified, including C-C motif chemokine ligand 14 (CCL14), C-type lectin domain family 4 member G (CLEC4G), ficolin-2 (L-ficolin, FCN2), insulin-like growth factor binding protein 3 (IGFBP3), C-X-C motif chemokine ligand 14 (CXCL14), and vasoactive intestinal polypeptide type I receptor (VIPR1). The area under the receiver operating characteristic (ROC) curve for the six signature genes was between 0.927-0.958, and the calibration curves also indicated that they had high prediction accuracy. Six signature genes were positively associated with NASH pathological process pathways including butyric acid metabolism and fatty acid degradation. The infiltration of immune cells such as M2-type macrophages was significantly positively correlated with the signature genes. RT-qPCR revealed a significant decrease in the expression of CLEC4G and IGFBP3 in the NASH-HCC model.

CLEC4G and IGFBP3 hold potential as biomarkers for clinical surveillance, offering new insights for early detection and prognosis evaluation.

Predicting drug-target interactions (DTIs) is essential for advancing drug discovery and personalized medicine. However, accurately capturing the intricate binding relationships between drugs and targets remains a significant challenge, particularly when attempting to fully leverage the vast correlation information inherent in molecular data. This complexity is further exacerbated by the structural differences and sequence length disparities between drug molecules and protein targets, which can hinder effective feature alignment and interaction modeling. To address these challenges, we propose a model named LAM-DTI. First, drug and target features are extracted from the original molecular sequence data using a multilayer convolutional neural network. To address the sequence length discrepancy between drug and target features, we apply a connectionist temporal classification module to generate normalized feature sequences. Building on this, we introduce a learnable association information matrix as a flexible intermediary, which dynamically adjusts to capture accurate DTI association information, thereby enhancing cross-modal mapping within a unified latent space. This progressive mapping strategy enables the model to form an interaction projection between drugs and targets, effectively identifying critical interaction regions and guiding the capture of complex interaction-related features. Extensive experiments on three well-known benchmark data sets demonstrate that LAM-DTI significantly outperforms previous models.

This study develops and validates a machine learning model using peritoneal cytology to predict distant metastasis in uterine carcinosarcoma, aiding clinical decision-making.

This study utilized detailed clinical data and peritoneal cytology findings from uterine carcinosarcoma patients in the SEER database. Eight machine learning algorithms-Logistic Regression, SVM, GBM, Neural Network, RandomForest, KNN, AdaBoost, and LightGBM-were applied to predict distant metastasis. Model performance was assessed using AUC, calibration curves, DCA, confusion matrices, sensitivity, and specificity. The Logistic Regression model was visualized with a nomogram, and its results were analyzed. SHAP values were used to interpret the best-performing machine learning model.

Peritoneal cytology, T stage, age, and tumor size were key factors influencing distant metastasis in uterine carcinosarcoma patients. Peritoneal cytology had significant weight in the prediction models. The logistic regression model demonstrated excellent predictive performance with an AUC of 0.882 in the training set and 0.881 in the internal test set. The model was visualized and interpreted using a nomogram. In comprehensive evaluations, GBM was identified as the best-performing model and was explained using SHAP values. Additionally, calibration and DCA curves indicated that both models have significant potential clinical utility.

This study introduces the first effective tool for predicting distant metastasis in uterine carcinosarcoma patients by integrating peritoneal cytology features into model construction. It aids in early identification of high-risk patients, enhancing follow-up and monitoring during tumor development, and supports the optimization of personalized treatment strategies.

High-grade serous tubo-ovarian cancer (HGSC) is marked by substantial inter- and intra-tumor heterogeneity. The tumor microenvironments (TME) of HGSC show pronounced variability in cellular make-up across metastatic sites, which is linked to poorer patient outcomes. The influence of cellular composition on therapy sensitivity, including chemotherapy and targeted treatments, has not been thoroughly investigated. In this study, we examined the premise that the variations in cellular composition can forecast drug efficacy. Using a high-throughput 3D in vitro tumoroid model, we assessed the drug responses of twenty-three distinct cellular configurations to an assortment of five therapeutic agents, including carboplatin and paclitaxel. By amalgamating our experimental findings with random forest machine learning algorithms, we assessed the influence of TME cellular composition on treatment reactions. Our findings reveal notable disparities in drug responses correlated with tumoroid composition, underscoring the significance of cellular diversity within the TME as a predictor of therapeutic outcomes. However, our work also emphasizes the complex nature of cell composition's influence on drug response. This research establishes a foundation for employing human tumoroids with varied cellular composition as a method to delve into the roles of stromal, immune, and other TME cell types in enhancing cancer cell susceptibility to various treatments. Additionally, these tumoroids can serve as a platform to explore pivotal cellular interactions within the TME that contribute to chemoresistance and cancer recurrence.

The global increase in diabetes prevalence necessitates advanced diagnostic methods. Machine learning has shown promise in disease diagnosis, including diabetes.

We used a dataset collected from the Medical City Hospital laboratory and the Specialized Center for Endocrinology and Diabetes at Al-Kindy Teaching Hospital in Iraq. This dataset includes 1000 physical examination samples from both male and female patients. The samples are categorized into three classes: diabetic (Y), nondiabetic (N), and predicted diabetic (P). The dataset contains twelve attributes and includes outlier data. Outliers in medical studies can result from unusual disease attributes. Therefore, consulting with a specialist physician to identify and handle these outliers using statistical methods is necessary. The main contribution of this study is the proposal of two hybrid models for diabetes diagnosis in two scenarios: (1) Scenario 1 (presence of outlier data): Hybrid Model 1 combines the K-medoids clustering algorithm with a Gaussian naive Bayes (GNB) classifier based on kernel density estimation (KDE) to handle outliers and (2) Scenario 2 (after removing outlier data): Hybrid Model 2 combines the K-means clustering algorithm with a GNB classifier based on KDE with suitable bandwidth. We performed principal component analysis to minimize dimensionality and evaluated the models using fivefold cross-validation.

All experiments were conducted in identical settings. Our proposed hybrid models demonstrated superior performance in two scenarios, handling and rejecting outliers, compared to other machine-learning models in this study, including support vector machines (with radial-based, polynomial, linear, and sigmoid kernel functions), decision trees (J48), and GNB classifiers for diabetes prediction. The average accuracy for Scenario 1 with Hybrid Model 1 was 0.9743, and for Scenario 2 with Hybrid Model 2, it was 0.9867. We also evaluated precision, sensitivity, and F1-score as performance metrics.

This study presents two hybrid models for diabetes diagnosis, demonstrating high accuracy in distinguishing between diabetic and nondiabetic patients and effectively handling outliers. The findings highlight the potential of machine-learning techniques for improving the early diagnosis and treatment of diabetes.

Sepsis, a complex inflammatory condition with high mortality rates, lacks effective treatments. This study explores the therapeutic mechanisms of Calculus Bovis in sepsis using network pharmacology and RNA sequencing.

Sepsis data from the China National GeneBank Database were analyzed for differentially expressed genes (FC ≥ 2, FDR < 0.05). Active components of Calculus Bovis were identified via the HERB and BATMAN-TCM databases, with target interactions assessed through protein-protein interaction (PPI) networks. GO and KEGG analyses identified pathway enrichments (P ≤ .01). Survival analysis using the GSE65682 database evaluated prognosis-related genes (P < .05). Four machine learning models (XGBoost, SVM, Decision Tree, KNN) were constructed to assess diagnostic potential, with AUC values evaluating accuracy. Immunofluorescence and single-cell RNA sequencing localized key genes, while molecular docking and molecular dynamics simulations (MD) assessed binding affinities and stability of Calculus Bovis compounds with target proteins.

We identified 593 targets for Calculus Bovis and 4329 sepsis-related genes, with 149 overlapping. Key genes ADAM17, CASP1, CD81, and MGMT were linked to improved prognosis (P < .05) and involved in inflammatory responses and pyroptosis (P ≤ .01). The XGBoost model achieved high diagnostic accuracy (AUC: training = 1.000, test = 0.964). Molecular docking showed strong binding (energy < -6.0 kcal/mol), and MD indicated stable interactions, particularly with ADAM17 and CD81.

This study highlights the potential of Calculus Bovis in sepsis treatment, identifying key genes as therapeutic targets.

Pterygium is a prevalent ocular disease characterized by abnormal conjunctival tissue proliferation, significantly impacting patients' quality of life. However, the underlying molecular mechanisms driving pterygium pathogenesis remain inadequately understood. This study aimed to investigate gene expression changes following pterygium excision and their association with immune cell infiltration. Clinical samples of pterygium and adjacent relaxed conjunctival tissue were collected for transcriptomic analysis using RNA sequencing combined with bioinformatics approaches. Machine learning algorithms, including LASSO, SVM-RFE, and Random Forest, were employed to identify potential diagnostic biomarkers. GO, KEGG, GSEA, and GSVA were utilized for enrichment analysis. Single-sample GSEA was employed to analyze immune infiltration. The GSE2513 and GSE51995 datasets from the GEO database, along with clinical samples, were selected for validation analysis. Differentially expressed genes (DEGs) were identified from the PRJNA1147595 and GSE2513 datasets, revealing 2437 DEGs and 172 differentially regulated genes (DRGs), respectively. There were 52 co-DEGs shared by both datasets, and four candidate biomarkers (FN1, SPRR1B, SERPINB13, EGR2) with potential diagnostic value were identified through machine learning algorithms. Single-sample GSEA demonstrated increased Th2 cell infiltration and decreased CD8 + T cell presence in pterygium tissues, suggesting a crucial role of the immune microenvironment in pterygium pathogenesis. Analysis of the GSE51995 dataset and qPCR results revealed significantly higher expression levels of FN1 and SPRR1B in pterygium tissues compared to conjunctival tissues, but SERPINB13 and EGR2 expression levels were not statistically significant. Furthermore, we identified four candidate drugs targeting the two feature genes FN1 and SPRR1B. This study provides valuable insights into the molecular characteristics and immune microenvironment of pterygium. The identification of potential biomarkers FN1 and SPRR1B highlights their significance in pterygium pathogenesis and lays a foundation for further exploration aimed at integrating these findings into clinical practice.

The integration of Deep Learning with sensor technologies has significantly advanced the field of intelligent sensing and decision making by enhancing perceptual capabilities and delivering sophisticated data analysis and processing functionalities. This review provides a comprehensive overview of the synergy between Deep Learning and sensors, with a particular focus on the applications of triboelectric nanogenerator (TENG)-based self-powered sensors combined with artificial intelligence (AI) algorithms. First, the evolution of Deep Learning is reviewed, highlighting the advantages, limitations, and application domains of several classical models. Next, the innovative applications of intelligent sensors in autonomous driving, wearable devices, and the Industrial Internet of Things (IIoT) are discussed, emphasizing the critical role of neural networks in enhancing sensor precision and intelligent processing capabilities. The review then delves into TENG-based self-powered sensors, introducing their self-powered mechanisms based on contact electrification and electrostatic induction, material selection strategies, novel structural designs, and efficient energy conversion methods. The integration of TENG-based self-powered sensors with Deep Learning algorithms is showcased through their groundbreaking applications in motion recognition, smart healthcare, smart homes, and human-machine interaction. Finally, future research directions are outlined, including multimodal data fusion, edge computing integration, and brain-inspired neuromorphic computing, to expand the application of self-powered sensors in robotics, space exploration, and other high-tech fields. This review offers theoretical and technical insights into the collaborative innovation of Deep Learning and self-powered sensor technologies, paving the way for the development of next-generation intelligent systems.

Intracranial aneurysm (IA) is a critical cerebrovascular condition, and nicotine exposure is a known risk factor. This study delves into the toxicological mechanisms of nicotine in IA, aiming to identify key biomarkers and therapeutic targets.

Gene Set Variation Analysis (GSVA), Weighted Gene Co-Expression Network Analysis (WGCNA), and enrichment analyses were conducted on differentially expressed genes (DEGs) from the GSE122897 dataset. Additionally, nicotine-related targets were identified using CTD, SwissTargetPrediction, and Super-PRED databases. Integrative machine learning approaches, such as Random Forest (RF) and Support Vector Machine (SVM), were employed to pinpoint key toxicity targets. Molecular docking and immune cell infiltration analyses were also performed.

DEGs in IA showed significant alterations in metabolic, secretory, signaling, and homeostatic pathways. Several immune and metabolic response pathways were notably disrupted. WGCNA identified 1127 DEGs with 37 overlapping toxic targets between IA and nicotine. ssGSEA revealed substantial upregulation in immune response and inflammation-related processes. Integrative analyses highlighted TGFB1, MCL1, and CDKN1A as core toxicity targets, confirmed via molecular docking studies. Immune cell infiltration analysis indicated significant correlations between these core targets and various immune cell populations.

This study uncovers significant disruptions in metabolic and immune pathways in IA under nicotine influence, identifying TGFB1, MCL1, and CDKN1A as critical biomarkers. These findings offer a deeper understanding of IA's molecular mechanisms and potential therapeutic targets for nicotine-related toxicity.

Heart failure (HF) is a multifaceted clinical condition, and our comprehension of its genetic pathogenesis continues to be significantly limited. Consequently, identifying specific genes for HF at the transcriptomic level may enhance early detection and allow for more targeted therapies for these individuals.

HF datasets were acquired from the Gene Expression Omnibus (GEO) database (GSE57338), and through the application of bioinformatics and machine-learning algorithms. We identified four candidate genes (FCN3, MNS1, SMOC2, and FREM1) that may serve as potential diagnostics for HF. Furthermore, we validated the diagnostic value of these genes on additional GEO datasets (GSE21610 and GSE76701). In addition, we assessed the different subtypes of heart failure through unsupervised clustering, and investigations were conducted on the differences in the immunological microenvironment, improved functions, and pathways among these subtypes. Finally, a comprehensive analysis of the expression profile, prognostic value, and genetic and epigenetic alterations of four potential diagnostic candidate genes was performed based on The Cancer Genome Atlas pan-cancer database.

A total of 295 differential genes were identified in the HF dataset, and intersected with the blue module gene with the highest correlation to HF identified by weighted correlation network analysis (r = 0.72, p = 1.3 × 10-43), resulting in a total of 114 key HF genes. Furthermore, based on random forest, least absolute shrinkage and selection operator, and support vector machine algorithms, we finally identified four hub genes (FCN3, FREM1, MNS1, and SMOC2) that had good potential for diagnosis in HF (area under the curve > 0.7). Meanwhile, three subgroups for patients with HF were identified (C1, C2, and C3). Compared with the C1 and C2 groups, we eventually identified C3 as an immune subtype. Moreover, the pan-cancer study revealed that these four genes are closely associated with tumor development.

Our research identified four unique genes (FCN3, FREM1, MNS1, and SMOC2), enhancing our comprehension of the causes of HF. This provides new diagnostic insights and potentially establishes a tailored approach for individualized HF treatment.

To establish a simple and noninvasive screening test for sleep apnea (SA) that imposes less burden on potential patients. The specific objective of this study was to verify the effectiveness of past and future single-lead electrocardiogram (ECG) data from SA occurrence sites in improving the estimation accuracy of SA and sleep apnea syndrome (SAS) using machine learning.

The Apnea-ECG dataset comprising 70 ECG recordings was used to construct various machine-learning models. The time window size was adjusted based on the accuracy of SA detection, and the performance of SA detection and SAS diagnosis (apnea‒hypopnea index ≥ 5 was considered SAS) was compared.

Using ECG data from a few minutes before and after the occurrence of SAs improved the estimation accuracy of SA and SAS in all machine learning models. The optimal range of the time window and achieved accuracy for SAS varied by model; however, the sensitivity ranged from 95.7 to 100%, and the specificity ranged from 91.7 to 100%.

ECG data from a few minutes before and after SA occurrence were effective in SA detection and SAS diagnosis, confirming that SA is a continuous phenomenon and that SA affects heart function over a few minutes before and after SA occurrence. Screening tests for SAS, using data obtained from single-lead ECGs with appropriate past and future time windows, should be performed with clinical-level accuracy.

Esophageal cancer is a highly aggressive gastrointestinal malignancy with a poor prognosis, making accurate prognostic assessment essential for patient care. The performance of the esophageal cancer prognosis model based on conventional radiomics is limited, as it mainly characterizes the spatial features such as texture of the tumor area, and cannot fully describe the complexity of esophageal cancer tumors. Therefore, we incorporate the frequency domain features into radiomics to improve the prognostic ability of esophageal cancer. Three hundred fifteen esophageal cancer patients participated in the death risk prediction experiment, with 80% being the training set and 20% being the testing set. We use fivefold cross validation for training, and fuse the 5 trained models through voting to obtain the final prognostic model for testing. The CatBoost achieved the best performance compared to machine learning methods such as random forests and decision tree. The experimental results showed that the combination of frequency domain and radiomics features achieved the highest performance in death predicting esophageal cancer (accuracy: 0.7423, precision: 0.7470, recall: 0.7375, specification: 0.8030, AUC: 0.8487), which was significantly better than the performance of frequency domain or radiomics features alone. The results of Kaplan-Meier survival analysis validated the performance of our method in death predicting esophageal cancer. The proposed method provides technical support for accurate prognosis of esophageal cancer.

A method for accurately classifying donkey breeds has been developed by integrating single nucleotide polymorphism (SNPs) data with machine learning algorithms. The approach includes preprocessing donkey genomic sequencing data, addressing data imbalance with the Synthetic Minority Over-sampling Technique (SMOTE), and utilizing an improved Leave-One-Out Cross-Validation (LOOCV) for dataset partitioning. Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Random Forest (RF) models were constructed and evaluated. The results demonstrated that different chromosomes significantly influence classifier performance. For instance, chromosome Chr2 showed the highest classification accuracy with KNN, while chromosome Chr19 performed best with SVM and RF models. After enhancing data quality and addressing imbalances, classification performance improved substantially, with accuracy, precision, recall, and F1 score showing increases of up to 15% in certain models, particularly on key chromosomes. This method offers an effective solution for donkey breed classification and provides technical support for the conservation and development of donkey genetic resources.

Microsatellite instability (MSI), caused by defects in mismatch repair (MMR) genes, serves as a critical molecular biomarker with therapeutic implications for endometrial cancer (EC). This study aims to assess the diagnostic performance of radiomics as a non-invasive approach for predicting MSI status in EC.

A systematic search across PubMed, Scopus, Embase, Web of Science, Cochrane library and Clinical Trials was conducted. Quality assessment was performed using QUADAS-2 and METRICS. Pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and area under the curve (AUC) were computed using a bivariate model. Separate meta-analyses for radiomics and combined models were conducted. Subgroup analysis and sensitivity analysis were conducted to find potential sources of heterogeneity. Likelihood ratio scattergram was used to evaluate the clinical applicability.

A total of 9 studies (1650 patients) were included in the systematic review, with seven studies contributing to the meta-analysis of radiomics model and five for combined model. The pooled diagnostic performance of the radiomics model was as follows: sensitivity, 0.66; specificity, 0.89; PLR, 5.48; NLR, 0.43; DOR, 18.56; and AUC, 0.87. For combined model, the pooled sensitivity, specificity, PLR, NLR, DOR, and AUC were 0.58, 0.94, 7.37, 0.50, 16.43, and 0.85, respectively. Subgroup analysis of radiomics models revealed that studies employing non-linear classifiers achieved superior performance compared to those utilizing linear classifiers.

Radiomics showed promise as non-invasive tool for MSI prediction in EC, with potential clinical utility in guiding personalized treatments. However, further studies are required to validate these findings.

The coffee-ring effect, involving spontaneous solute separation, has demonstrated promising potential in the context of patient serum analysis. In this study, an approach leveraging the coffee-ring-based analyte redistribution was developed for spectral analysis of surface-enhanced Raman scattering (SERS). By performing radical SERS scanning through the coffee-ring area and sampling across the coffee ring, complicated chemical information was spatially gathered for further spectra analysis. The corresponding application in classification of serum samples from breast cancer patients was also proposed. A simulated serum environment was constructed by mixing phenylalanine, hypoxanthine, and bovine serum albumin (BSA), yielding the coffee-ring patterns along with gold nanoparticles. Distinct divergence in the distributions between hypoxanthine and phenylalanine within the rings were characterized, which is attributed to the inherent electrostatic properties of the noble metal colloid and the interactions among different solvents. Subsequently, this method was applied to serum samples from patients diagnosed with the four breast cancer subtypes. By preparing serum with SERS substrates and forming the coffee-ring patterns, radial SERS scanning was conducted across the rings. The acquired spectra were spatially segmented and processed by employing a multilayer perceptron for learning and prediction. The classification results demonstrated a predictive accuracy of 85.7% in distinguishing among the four breast cancer subtypes, highlighting the feasibility and effectiveness of the coffee-ring assisted radial SERS analysis.

Background: In recent years, the incidence of Crohn's disease (CD) has shown a significant global increase, with numerous studies demonstrating its correlation with various cancers. This study aims to identify novel biomarkers for diagnosing CD and explore their potential applications in pan-cancer analysis. Methods: Gene expression profiles were retrieved from the Gene Expression Omnibus (GEO) database, and differentially expressed genes (DEGs) were identified using the "limma" R package. Key biomarkers were selected through an integrative machine learning pipeline combining LASSO regression, neural network modeling, and Support Vector Machine-Recursive Feature Elimination (SVM-RFE). Six hub genes were identified and further validated using the independent dataset GSE169568. To assess the broader relevance of these biomarkers, a standardized pan-cancer dataset from the UCSC database was analyzed to evaluate their associations with 33 cancer types. Results: Among the identified biomarkers, S100 calcium binding protein P (S100P) and S100 calcium binding protein A8 (S100A8) emerged as key candidates for CD diagnosis, with strong validation in the independent dataset. Notably, S100P displayed significant associations with immune cell infiltration and patient survival outcomes in both liver and lung cancers. These findings suggest that chronic inflammation and immune imbalances in CD may not only contribute to disease progression but also elevate cancer risk. As an inflammation-associated biomarker, S100P holds particular promise for both CD diagnosis and potential cancer risk stratification, especially in liver and lung cancers. Conclusion: Our study highlights S100P and S100A8 as potential diagnostic biomarkers for CD. Moreover, the pan-cancer analysis underscores the broader clinical relevance of S100P, offering new insights into its role in immune modulation and cancer prognosis. These findings provide a valuable foundation for future research into the shared molecular pathways linking chronic inflammatory diseases and cancer development.

Metabolic-bariatric surgery (MBS) is the last effective way to lose weight whom around half of the patients are women of reproductive age. It is recommended an interval of 12 months between surgery and pregnancy to optimize weight loss and nutritional status. Predicting pregnancy up to 12 months after MBS is important for evaluating reproductive health services in bariatric centers; therefore, this study aimed to present a prediction model for pregnancy at the first year following MBS using machine learning (ML) algorithms.

In a nested case-control study of 473 women with a history of pregnancy after MBS during 2009-2023, predisposing factors in pregnancy within 12 months after MBS were identified and subsequently, several ML models, including the classification algorithms and decision trees, as well as regression analyses, were applied to predict pregnancy up to 12 months after MBS.

The highest area under the curve (AUC) was 0.920 ± 0.014 (95%CI 0.906, 0.927) for the C5.0 decision tree with sensitivity and specificity of 0.762 ± 0.044 (95%CI 0.739, 0.801) and 0.916 ± 0.028 (95%CI 0.883, 0.922), respectively. This model considered thirteen important factors to predict pregnancy at the first 12 months following MB, including menstrual irregularity, marital status, a history of abnormal fetal development, age, infertility type, parity, gravidity, fertility treatment, presurgery body mass index (BMI), infertility, infertility duration, polycystic ovary syndrome (PCOS), and type 2 diabetes (T2DM).

Developing the ML models, which predict pregnancy within 12 months after MBS, can help bariatric surgeons and obstetricians to prevent and manage suboptimal surgical response and adverse pregnancy outcomes.

This study aimed to establish and validate the efficacy of a nomogram model, synthesized through the integration of multi-parametric magnetic resonance radiomics and clinical risk factors, for forecasting perineural invasion in rectal cancer. We retrospectively collected data from 108 patients with pathologically confirmed rectal adenocarcinoma who underwent preoperative multiparametric MRI at the First Affiliated Hospital of Bengbu Medical College between April 2019 and August 2023. This dataset was subsequently divided into training and validation sets following a ratio of 7:3. Both univariate and multivariate logistic regression analyses were implemented to identify independent clinical risk factors associated with perineural invasion (PNI) in rectal cancer. We manually delineated the region of interest (ROI) layer-by-layer on T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) sequences and extracted the image features. Five machine learning algorithms were used to construct radiomics model with the features selected by least absolute shrinkage and selection operator (LASSO) method. The optimal radiomics model was then selected and combined with clinical features to formulate a nomogram model. The model performance was evaluated using receiver operating characteristic (ROC) curve analysis, and its clinical value was assessed via decision curve analysis (DCA). Our final selection comprised 10 optimal radiological features and the SVM model showcased superior predictive efficiency and robustness among the five classifiers. The area under the curve (AUC) values of the nomogram model were 0.945 (0.899, 0.991) and 0.846 (0.703, 0.99) for the training and validation sets, respectively. The nomogram model developed in this study exhibited excellent predictive performance in foretelling PNI of rectal cancer, thereby offering valuable guidance for clinical decision-making. The nomogram could predict the perineural invasion status of rectal cancer in early stage.

This study aims to predict and diagnose pediatric septic shock through the screening of immune infiltration-related biomarkers. Three gene expression datasets were accessible from the Gene Expression Omnibus repository. The differentially expressed genes were identified using the R 4.3.2 ( https://www.r-project.org/ ), followed by gene set enrichment analysis. Thereafter, the genes were identified utilizing machine-learning algorithms. The receiver operating characteristic curve was employed to assess the discrimination and effectiveness of the hub genes. The inflammatory and immune status of pediatric septic shock was evaluated through cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT). The correlation between diagnostic markers and infiltrating immune cells was further examined. Overall, we detected 12 differentially expressed genes. CD177, MCEMP1, MMP8, and OLAH were examined as diagnostic indicators for pediatric septic shock, revealing statistically significant differences (P < 0.01) and diagnostic efficacy in the validation cohort. The immune cell infiltration analysis suggests that various immune cells may contribute to the onset of pediatric septic shock. Furthermore, all diagnostic characteristics may exhibit varying degrees of correlation with immune cells. This study identifies four potential biomarkers-CD177, MCEMP1, MMP8, and OLAH-that provide diagnostic value and novel insights into immune dysregulation in pediatric septic shock. Through the integration of bioinformatics and machine learning methodologies, we offer a novel perspective on the immune mechanisms involved in pediatric septic shock, potentially facilitating more targeted and personalized therapies for individual patients.

This study aims to investigate the influencing factors of pregnancy outcomes during in vitro fertilization and embryo transfer (IVF-ET) procedures in clinical practice. Several prediction models were constructed to predict pregnancy outcomes and models with higher accuracy were identified for potential implementation in clinical settings.

The clinical data and pregnancy outcomes of 2625 women who underwent fresh cycles of IVF-ET between 2016 and 2022 at the Reproductive Center of the Affiliated Hospital of Inner Mongolia Medical University were enrolled to establish a comprehensive dataset. The observed features were preprocessed and analyzed. A predictive model for pregnancy outcomes of IVF-ET treatment was constructed based on the processed data. The dataset was divided into a training set and a test set in an 8:2 ratio. Predictive models for clinical pregnancy and clinical live births were developed. The ROC curve was plotted, and the AUC was calculated and the prediction model with the highest accuracy rate was selected from multiple models. The key features and main aspects of IVF-ET treatment outcome prediction were further analyzed.

The clinical pregnancy outcome was categorized into pregnancy and live birth. The XGBoost model exhibited the highest AUC for predicting pregnancy, achieving a validated AUC of 0.999 (95% CI: 0.999-1.000). For predicting live births, the LightGBM model exhibited the highest AUC of 0.913 (95% CI: 0.895-0.930).

The XGBoost model predicted the possibility of pregnancy with an accuracy of up to 0.999. While the LightGBM model predicted the possibility of live birth with an accuracy of up to 0.913.

The number of survivors of cancer is growing, and they often experience negative long-term behavioral outcomes due to cancer treatments. There is a need for better computational methods to handle and predict these outcomes so that physicians and health care providers can implement preventive treatments.

This study aimed to create a new feature selection algorithm to improve the performance of machine learning classifiers to predict negative long-term behavioral outcomes in survivors of cancer.

We devised a hybrid deep learning-based feature selection approach to support early detection of negative long-term behavioral outcomes in survivors of cancer. Within a data-driven, clinical domain-guided framework to select the best set of features among cancer treatments, chronic health conditions, and socioenvironmental factors, we developed a 2-stage feature selection algorithm, that is, a multimetric, majority-voting filter and a deep dropout neural network, to dynamically and automatically select the best set of features for each behavioral outcome. We also conducted an experimental case study on existing study data with 102 survivors of acute lymphoblastic leukemia (aged 15-39 years at evaluation and >5 years postcancer diagnosis) who were treated in a public hospital in Hong Kong. Finally, we designed and implemented radial charts to illustrate the significance of the selected features on each behavioral outcome to support clinical professionals' future treatment and diagnoses.

In this pilot study, we demonstrated that our approach outperforms the traditional statistical and computation methods, including linear and nonlinear feature selectors, for the addressed top-priority behavioral outcomes. Our approach holistically has higher F1, precision, and recall scores compared to existing feature selection methods. The models in this study select several significant clinical and socioenvironmental variables as risk factors associated with the development of behavioral problems in young survivors of acute lymphoblastic leukemia.

Our novel feature selection algorithm has the potential to improve machine learning classifiers' capability to predict adverse long-term behavioral outcomes in survivors of cancer.

Objective: The primary objective of this study is to enhance the detection and staging of pressure injuries using machine learning capabilities for precise image analysis. This study explores the application of the You Only Look Once version 8 (YOLOv8) deep learning model for pressure injury staging. Approach: We prepared a high-quality, publicly available dataset to evaluate different variants of YOLOv8 (YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x) and five optimizers (Adam, AdamW, NAdam, RAdam, and stochastic gradient descent) to determine the most effective configuration. We followed a simulation-based research approach, which is an extension of the Consolidated Standards of Reporting Trials (CONSORT) and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for dataset preparation and algorithm evaluation. Results: YOLOv8s, with the AdamW optimizer and hyperparameter tuning, achieved the best performance metrics, including a mean average precision at intersection over union ≥0.5 of 84.16% and a recall of 82.31%, surpassing previous YOLO-based models in accuracy. The ensemble model incorporating all YOLOv8 variants showed strong performance when applied to unseen images. Innovation: Notably, the YOLOv8s model significantly improved detection for challenging stages such as Stage 2 and achieved accuracy rates of 0.90 for deep tissue injury, 0.91 for Unstageable, and 0.74, 0.76, 0.70, and 0.77 for Stages 1, 2, 3, and 4, respectively. Conclusion: These results demonstrate the effectiveness of YOLOv8s and ensemble models in improving the accuracy and robustness of pressure injury staging, offering a reliable tool for clinical decision-making.

Cell death has long been a focal point in life sciences research, and recently, scientists have discovered a novel form of cell death induced by copper, termed cuproptosis. This paper aimed to identify genes associated with cuproptosis in systemic lupus erythematosus (SLE) through machine learning, combined with single-cell RNA sequencing (scRNA-seq), to screen and validate related genes. The analytical results were then experimentally verified. Two published microarray gene expression datasets (GSE65391 and GSE61635) from SLE and control peripheral blood samples were downloaded from the GEO database. The GSE65391 dataset was used as the training group, while the GSE61635 dataset served as the validation group. Differentially expressed genes from GSE65391 identified 12 differential genes. Nine diagnostic genes, considered potential biomarkers, were selected using the least absolute shrinkage and selection operator and support vector machine recursive feature elimination analysis. The receiver operating characteristic (ROC) curves for both the training and validation groups were used to calculate the area under the curve to assess discriminatory properties. CIBERSORT was used to assess the relationship between these diagnostic genes and a reference set of infiltrating immune cells. scRNA-seq data (GSE162577) from SLE patients were also obtained from the GEO database and analyzed. Experimental validation of the most important SLE biomarkers was performed. Twelve significantly different cuproptosis-related genes were identified in the GSE65391 training set. Immune cell analysis revealed 12 immune cell types and identified nine signature genes, including PDHB, glutaminase (GLS), DLAT, LIAS, MTF1, DLST, DLD, LIPT1, and FDX1. In the GSE61635 validation set, seven genes were weakly expressed, and two genes were strongly expressed in the treatment group. According to the ROC curves, PDHB and GLS demonstrated significant diagnostic value. Additionally, correlation analysis was conducted on the nine characteristic genes in relation to immune infiltration. The distribution of key genes in immune cells was determined using scRNA-seq data. Finally, the mRNA expression of the nine diagnostic genes was validated using qPCR.

Prostate cancer (PCa) recurrence affects between 20% and 40% of patients, being a significant challenge for predicting clinical outcomes and increasing survival rates. Although serum PSA levels, Gleason score, and tumor staging are sensitive for detecting recurrence, they present low specificity. This study compared the performance of three supervised machine learning models, Naive Bayes (NB), Support Vector Machine (SVM), and Artificial Neural Network (ANN) for classifying PCa recurrence events using a dataset of 489 patients from The Cancer Genome Atlas (TCGA). Besides comparing the models performance, we searched for analyzing whether the incorporation of specific genes expression in the predictor set would enhance the prediction of PCa recurrence, then suggesting these genes as potential biomarkers of patient prognosis. The models showed accuracy above 60% and sensitivity above 65% in all combinations. ANN models were more consistent in their performance across different predictor sets. Notably, SVM models showed strong results in precision and specificity, particularly considering the inclusion of genes selected by feature selection (NETO2, AR, HPN, and KLK3), without compromising sensitivity. However, the relatively high standard deviations observed in some metrics indicate variability across simulations, suggesting a gap for additional studies via different datasets. These findings suggest that genes are potential biomarkers for predicting PCa recurrence in the dataset, representing a promising approach for early prognosis even before the main treatment.

Severe Community-Acquired Pneumonia (SCAP) is a serious global health issue with high incidence and mortality rates. In recent years, the role of biomarkers such as Connective Tissue Growth Factor (CTGF) and Milk Fat Globule-Epidermal Growth Factor 8 (MFG-E8) in disease diagnosis and prognosis has increasingly gained attention. However, their specific functions in SCAP have still remained unclear. By conducting a prospective analysis, this study has explored the relationship between these two proteins and the diagnosis and mortality of SCAP patients. Additionally, founded on comparing the applications of machine learning and nomograms as predictive models in forecasting the 28-day mortality risk of SCAP patients, this paper has discussed their performance in different medical scenarios to provide more accurate treatment options and improve prognosis.

198 patients diagnosed with SCAP, 80 patients with CAP and 80 healthy individuals were encompassed in the study. Demographic characteristics, clinical features and biomarkers were extracted. The ELISA method was employed to measure the levels of MFG-E8 and CTGF in the three groups. The 28-day mortality of SCAP patients was tracked. Eleven models, including XGBoost and CatBoost, were used as prediction models and compared with a nomogram. And 14 scoring methods, like F1 Score and AUC Score, were used to evaluate the prediction models.

Compared to healthy controls, SCAP patients had higher serum levels of CTGF and MFG-E8, suggesting that these biomarkers are associated with poor prognosis. Compared to CAP patients, SCAP patients had lower levels of MFG-E8 and higher levels of CTGF. In the deceased group of SCAP patients, their CTGF levels were higher and MFG-E8 levels were lower. Using the CatBoost model for prediction, it performed the best, with key predictive features including Oxygenation Index, cTnT, MFG-E8, Dyspnea, CTGF and PaCO2.

This study has highlighted the critical role of clinical and biochemical markers such as CTGF and MFG-E8 in assessing the severity and prognosis of SCAP. The CatBoost model has shown the significant potential in predicting mortality risk by virtue of its unique algorithmic advantages and efficiency.

The etiology of interstitial cystitis/painful bladder syndrome (IC/BPS) remains elusive, presenting significant challenges in both diagnosis and treatment. To address these challenges, we employed a comprehensive approach aimed at identifying diagnostic biomarkers that could facilitate the assessment of immune status in individuals with IC/BPS.

Transcriptome data from IC/BPS patients were sourced from the Gene Expression Omnibus (GEO) database. We identified differentially expressed genes (DEGs) crucial for gene set enrichment analysis. Key genes within the module were revealed using weighted gene co-expression network analysis (WGCNA). Hub genes in IC/BPS patients were identified through the application of three distinct machine-learning algorithms. Additionally, the inflammatory status and immune landscape of IC/BPS patients were evaluated using the ssGSEA algorithm. The expression and biological functions of key genes in IC/BPS were further validated through in vitro experiments.

A total of 87 DEGs were identified, comprising 43 up-regulated and 44 down-regulated genes. The integration of predictions from the three machine-learning algorithms highlighted three pivotal genes: PLAC8 (AUC: 0.887), S100A8 (AUC: 0.818), and PPBP (AUC: 0.871). Analysis of IC/BPS tissue samples confirmed elevated PLAC8 expression and the presence of immune cell markers in the validation cohorts. Moreover, PLAC8 overexpression was found to promote the proliferation of urothelial cells without affecting their migratory ability by inhibiting the Akt/mTOR/PI3K signaling pathway.

Our study identifies potential diagnostic candidate genes and reveals the complex immune landscape associated with IC/BPS. Among them, PLAC8 is a promising diagnostic biomarker that modulates the immune response in patients with IC/BPS, which provides new insights into the future diagnosis of IC/BPS.

The clinical manifestation of aortic dissection (AD) is complex and varied, making early diagnosis crucial for patient survival. This study aimed to identify immune-related markers to establish a nomogram model for AD diagnosis. Three datasets from GEO-GSE52093, GSE147026 and GSE153434-were combined and used for identification of immune-related causative genes using weighted gene co-expression network analysis, and 136 immune-related genes were obtained. Then, 15 pivotal genes were screened by the protein-protein interaction network. Through machine learning including the Least Absolute Shrinkage and Selection Operator algorithm, random forest algorithm, and multivariate logistic regression, four key feature genes were obtained-CXCL1, ITGA5, PTX3, and TIMP1-and the diagnostic scores based on these four genes were proved to be effective in distinguishing between AD patients and healthy donors. External dataset (GSE98770 and GSE190635) validation revealed this nomogram displayed strong predictive significance. Further analysis revealed that these genes are related with neutrophils, resting NK cells, resting mast cells, activated mast cells, activated dendritic cells, central memory CD4 T cells, γδ T cells, natural killer T cells, and myeloid-derived suppressor cells in AD. Finally, these four genes were validated to be upregulated in AD patients' tissue and serum samples compared with controls. These results suggest that this nomogram model, using machine learning identified four immune-related genes CXCL1, ITGA5, PTX3, and TIMP1, displays superior diagnostic ability in distinguishing AD and healthy individuals, and immune cells commonly associated with these hub genes may be therapeutic targets for AD.

Bladder, kidney, and prostate cancers are prevalent urinary cancers, and developing efficient detection methods is of significance for the early diagnosis of them. However, noninvasive and sensitive detection of urinary cancers still challenges traditional techniques. In this study, we developed a SERS-based method to analyze serum samples from patients with urinary cancers. Rapid, label-free, and highly sensitive detection of human sera is achieved by cleaning and aggregating silver nanoparticles. Furthermore, a long short-term memory deep learning algorithm is used to distinguish serum spectra, and the performance of the model is evaluated by comparing the accuracy, sensitivity, specificity, and receiver operating characteristic curves. Taking advantage of SERS and machine learning in sensitivity and data processing, the three urinary cancers are clearly classified. This is the first attempt to exploit the SERS-machine learning strategy to discriminate multiple urinary cancers with clinical serum samples, and our results showed the potential application of this method in the early diagnosis and screening of cancers.

This study aims to utilize proteomics, bioinformatics, and machine learning algorithms to identify diagnostic biomarkers in the serum of patients with acute and chronic brucellosis.

Proteomic analysis was conducted on serum samples from patients with acute and chronic brucellosis, as well as from healthy controls. Differential expression analysis was performed to identify proteins with altered expression, while Weighted Gene Co-expression Network Analysis (WGCNA) was applied to detect co-expression modules associated with clinical features of brucellosis. Machine learning algorithms were subsequently used to identify the optimal combination of diagnostic biomarkers. Finally, ELISA was employed to validate the identified proteins.

A total of 1,494 differentially expressed proteins were identified, revealing two co-expression modules significantly associated with the clinical characteristics of brucellosis. The Gaussian Mixture Model (GMM) algorithm identified six proteins that were concurrently present in both the differentially expressed and co-expression modules, demonstrating promising diagnostic potential. After ELISA validation, five proteins were ultimately selected.

These five proteins are implicated in the innate immune processes of brucellosis, potentially associated with its pathogenic mechanisms and chronicity. Furthermore, we highlighted their potential as diagnostic biomarkers for brucellosis. This study further enhances our understanding of brucellosis at the protein level, paving the way for future research endeavors.

Steroid-induced osteonecrosis of the femoral head (SIONFH) is a universal hip articular disease and is very hard to perceive at an early stage. The understanding of the pathogenesis of SIONFH is still limited, and the identification of efficient diagnostic biomarkers is insufficient. This research aims to recognize and validate the latent exosome-related molecular signature in SIONFH diagnosis by employing bioinformatics to investigate exosome-related mechanisms in SIONFH.

The GSE123568 and GSE74089 datasets were employed to conduct differentially expressed genes (DEGs) analysis, and the GSE123568 dataset was subjected to perform weighted genes co-expression network analysis (WGCNA). The exosome-related genes (ERGs) were retrieved from the GeneCards database. We identified differentially expressed exosome-related genes (DEERGs) between healthy controls (HC) and SIONFH patients, and a consensus clustering analysis was then implemented to group the SIONFH patients. The CIBERSORT was implemented to calculate the immune cell infiltration. Gene Set Variation Analysis (GSVA), Gene Ontology (GO), and Kyoto Encyclopedia of Genes and Genomes (KEGG) were conducted to investigate latent enriched pathways. In addition, machine-learning algorithms were applied to refine the DEERGs. Ultimately, we verified the diagnostic significance and expression of the hub genes using the SIONFH datasets and performing quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis.

This study identified twenty DEERGs from the peripheral serum and hip articular cartilage samples of SIONFH patients and HC. Two SIONFH subtypes related to ERGs were identified, and distinctions in pathways and immune cell infiltration patterns were compared. SIONFH's high-risk subpopulation exhibited enriched immune-related pathways and high immune cell infiltration, such as M0 macrophages, resting mast cells, and neutrophils. Three machine-learning algorithms then determined LCP1, PNP, UBE2V1, and ZFP36 as four exosome-related hub genes (ERHGs). Compared to HC samples, these ERHGs showed excellent diagnostic efficiency (overall AUC for ERHGs is in the range of 0.923 to 0.970 in GSE123568) in SIONFH samples. LCP1, PNP, UBE2V1, and ZFP36 expressions were validated in the GSE123568 and GSE74089 datasets and finally detected in peripheral serum samples with accordant expression by RT-qPCR.

Twenty potential exosome-related genes involved in SIONFH were identified through bioinformatics analysis. LCP1, PNP, UBE2V1, and ZFP36 might become candidate biomarkers and therapeutic targets because they have an intimate relationship with exosomes. These findings shed light on the exosome-related acquaintance of SIONFH and might contribute to the diagnosis and prognosis of SIONFH.

Acute Myeloid Leukemia (AML) with KMT2A rearrangements (KMT2Ar), found on chromosome 11q23, is often called KMT2A-rearranged AML (KMT2Ar-AML). This variant is highly aggressive, characterized by rapid disease progression and poor outcomes. Growing knowledge of epigenetic changes, especially lactylation, has opened new avenues for investigation and management of this subtype. Lactylation plays a significant role in cancer, inflammation, and tissue regeneration, but the underlying mechanisms are not yet fully understood. This research examined the influence of lactylation on gene expression within KMT2Ar-AML, initially identifying twelve notable lactylation-dependent differentially expressed genes (DEGs). Using advanced machine learning techniques, six key lactylation-associated genes (PFN1, S100A6, CBR1, LDHB, LGALS1, PRDX1) were identified as essential for prognostic evaluation and linked to relevant disease pathways. The study also suggested PI3K inhibitors and Pevonedistat as possible therapeutic options to modulate immune cell infiltration. Our findings confirm the critical role of lactylation in KMT2Ar-AML and identify six key genes that may serve as biomarkers for diagnosis and treatment. In addition to highlighting the need for further validation in clinical settings, these findings contribute to our understanding of KMT2Ar-AML's molecular mechanisms.

Intramuscular fat (IMF) is an important indicator for evaluating meat quality. Transcriptome sequencing (RNA-seq) is widely used for the study of IMF deposition. Machine learning (ML) is a new big data fitting method that can effectively fit complex data, accurately identify samples and genes, and it plays an important role in omics research. Therefore, this study aimed to analyze RNA-seq data by ML method to identify differentially expressed genes (DEGs) affecting IMF deposition in pigs. In this study, a total of 74 RNA-seq data from muscle tissue samples were used. A total of 155 DEGs were identified using a limma package between the two groups. 100 and 11 significant genes were identified by support vector machine recursive feature elimination (SVM-RFE) and random forest (RF) models, respectively. A total of six intersecting genes were in both models. KEGG pathway enrichment analysis of the intersecting genes revealed that these genes were enriched in pathways associated with lipid deposition. These pathways include α-linolenic acid metabolism, linoleic acid metabolism, ether lipid metabolism, arachidonic acid metabolism, and glycerophospholipid metabolism. Four key genes affecting intramuscular fat deposition, PLA2G6, MPV17, NUDT2, and ND4L, were identified based on significant pathways. The results of this study are important for the elucidation of the molecular regulatory mechanism of intramuscular fat deposition and the effective improvement of IMF content in pigs.

Polycystic ovary syndrome (PCOS) is a common endocrine disorder affecting women of reproductive age. It is characterized by symptoms such as hyperandrogenemia, oligo or anovulation and polycystic ovarian, significantly impacting quality of life. However, the practical implementation of machine learning (ML) in PCOS diagnosis is hindered by the limitations related to data size and algorithmic models. To address this research gap, we have increased the sample size in our study and aim to utilize two ML algorithms to analyze and validate diagnostic biomarkers, as well as explore immune cell infiltration patterns in PCOS.

We performed RNA-seq analysis on granulosa cell, including 13 samples from normal controls and 25 samples from women with PCOS. The data from our study were combined with publicly available databases. Batch effects were corrected using the 'sva' package in R software. Differential expression analysis was performed to identify genes that exhibited significant differences between the two groups. These differentially expressed genes (DEGs) were further analyzed for Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Hub genes were selected by intersecting the results of both methods after using LASSO and SVM-RFE for central gene selection for DEGs. Receiver Operating Characteristic (ROC) curves were employed to verify the accuracy of models by SVM and XGBoost. CIBERSORT analysis was performed to determine the relative abundances of immune cell populations. GSEA was analyzed to illustrate the expression patterns of genes within highly enriched functional pathways. RT-qPCR was used to validate the reliability of hub genes.

824 DEGs were found between the normal control and PCOS groups, including 376 upregulated and 448 downregulated genes. These DEGs were associated with endocytosis, salmonella infection and focal adhesion based on the KEGG enrichment analysis. Through overlapping LASSO and SVM-RFE algorithms, we identified four hub genes (CNTN2, CASR, CACNB3, MFAP2) that are significantly associated with the PCOS group. The diagnostic efficacy validation set using SVM and XGBoost yielded AUC values of 0.795 and 0.875, respectively, indicating their potential as diagnostic biomarkers. Consistent with the data analysis, the upregulation of CNTN2, CASR, CACNB3, and MFAP2 in PCOS was confirmed by RT-qPCR analysis on human granulosa cells. Furthermore, according to CIBERSORT analysis, a significant reduction in CD4 memory resting T cells was revealed in the PCOS group compared to the normal control group (P < 0.05).

This study identified CNTN2, CASR, CACNB3, and MFAP2 as potential diagnostic biomarkers for PCOS, which provides strong evidence for existing research on hub genes. Furthermore, the analysis of immune cell infiltration revealed the significant involvement of CD4 memory resting T cells in the onset and progression of PCOS. These findings shed light on potential mechanisms underlying PCOS pathogenesis and provide valuable insights for future research and therapeutic interventions.

Diagnosis and treatment of endometrial diseases are crucial for women's health. Over the past decade, ultrasound has emerged as a non-invasive, safe, and cost-effective imaging tool, significantly contributing to endometrial disease diagnosis and generating extensive datasets. The introduction of artificial intelligence has enabled the application of machine learning and deep learning to extract valuable information from these datasets, enhancing ultrasound diagnostic capabilities. This paper reviews the progress of artificial intelligence in ultrasound image analysis for endometrial diseases, focusing on applications in diagnosis, decision support, and prognosis analysis. We also summarize current research challenges and propose potential solutions and future directions to advance ultrasound artificial intelligence technology in endometrial disease diagnosis, ultimately improving women's health through digital tools.

Anxi Tieguanyin belongs to the oolong tea category and is one of the top ten most famous teas in China. In this study, hyperspectral imaging (HSI) technology was combined with chemometric methods to achieve the rapid determination of free amino acid and tea polyphenol contents in Tieguanyin tea. Here, the spectral data of Tieguanyin tea samples of four quality grades were obtained via visible near-infrared hyperspectroscopy in the range of 400-1000 nm, and the free amino acid and tea polyphenol contents of the samples were detected. First derivative (1D), normalization (Nor), and Savitzky-Golay (SG) smoothing were utilized to preprocess the original spectrum. The characteristic wavelengths were extracted via principal component analysis (PCA), competitive adaptive reweighted sampling (CARS), and the successive projection algorithm (SPA). The contents of free amino acid and tea polyphenol in Tieguanyin tea were predicted by the back propagation (BP) neural network, partial least squares regression (PLSR), random forest (RF), and support vector machine (SVM). The results revealed that the free amino acid content of the clear-flavoured Tieguanyin was greater than that of the strong-flavoured type, that the tea polyphenol content of the strong-flavoured Tieguanyin was greater than that of the clear-flavoured type, and that the content of the first-grade product was greater than that of the second-grade product. The 1D preprocessing improved the resolution and sensitivity of the spectra. When using CARS, the number of wavelengths for free amino acids and tea polyphenols was reduced to 50 and 70, respectively. The combination of 1D and CARS is conducive to improving the accuracy of late modelling. The 1D-CARS-RF model had the highest accuracy in predicting the free amino acid (RP2 = 0.940, RMSEP = 0.032, and RPD = 4.446) and tea polyphenol contents (RP2 = 0.938, RMSEP = 0.334, and RPD = 4.474). The use of hyperspectral imaging combined with multiple algorithms can be used to achieve the fast and non-destructive prediction of free amino acid and tea polyphenol contents in Tieguanyin tea.