Accurate Prediction Results Research Articles

Machine learning algorithm for predicting seizure control after temporal lobe resection using peri-ictal electroencephalography

Brain resection is curative for a subset of patients with drug resistant epilepsy but up to half will fail to achieve sustained seizure freedom in the long term. There is a critical need for accurate prediction tools to identify patients likely to have recurrent postoperative seizures. Results from preclinical models and intracranial EEG in humans suggest that the window of time immediately before and after a seizure (“peri-ictal”) represents a unique brain state with implications for clinical outcome prediction. Using a dataset of 294 patients who underwent temporal lobe resection for seizures, we show that machine learning classifiers can make accurate predictions of postoperative seizure outcome using 5 min of peri-ictal scalp EEG data that is part of universal presurgical evaluation (AUC 0.98, out-of-group testing accuracy > 90%). This is the first approach to seizure outcome prediction that employs a routine non-invasive preoperative study (scalp EEG) with accuracy range likely to translate into a clinical tool. Decision curve analysis (DCA) shows that compared to the prevalent clinical-variable based nomogram, use of the EEG-augmented approach could decrease the rate of unsuccessful brain resections by 20%.

Psychosis Prognosis Predictor: A continuous and uncertainty-aware prediction of treatment outcome in first-episode psychosis.

Machine learning models have shown promising potential in individual-level outcome prediction for patients with psychosis, but also have several limitations. To address some of these limitations, we present a model that predicts multiple outcomes, based on longitudinal patient data, while integrating prediction uncertainty to facilitate more reliable clinical decision-making. We devised a recurrent neural network architecture incorporating long short-term memory (LSTM) units to facilitate outcome prediction by leveraging multimodal baseline variables and clinical data collected at multiple time points. To account for model uncertainty, we employed a novel fuzzy logic approach to integrate the level of uncertainty into individual predictions. We predicted antipsychotic treatment outcomes in 446 first-episode psychosis patients in the OPTiMiSE study, for six different clinical scenarios. The treatment outcome measures assessed at both week 4 and week 10 encompassed symptomatic remission, clinical global remission, and functional remission. Using only baseline predictors to predict different outcomes at week 4, leave-one-site-out validation AUC ranged from 0.62 to 0.66; performance improved when clinical data from week 1 was added (AUC = 0.66-0.71). For outcome at week 10, using only baseline variables, the models achieved AUC = 0.56-0.64; using data from more time points (weeks 1, 4, and 6) improved the performance to AUC = 0.72-0.74. After incorporating prediction uncertainties and stratifying the model decisions based on model confidence, we could achieve accuracies above 0.8 for ~50% of patients in five out of the six clinical scenarios. We constructed prediction models utilizing a recurrent neural network architecture tailored to clinical scenarios derived from a time series dataset. One crucial aspect we incorporated was the consideration of uncertainty in individual predictions, which enhances the reliability of decision-making based on the model's output. We provided evidence showcasing the significance of leveraging time series data for achieving more accurate treatment outcome prediction in the field of psychiatry.

Predictive value of follow-up infarct volume on functional outcomes in middle cerebral artery M2 segment vessel occlusion stroke treated with mechanical thrombectomy.

Medium vessel occlusion (MeVO) strokes, particularly affecting the M2 segment of the middle cerebral artery, represent a critical proportion of acute ischemic strokes, posing significant challenges in management and outcome prediction. The efficacy of mechanical thrombectomy (MT) in MeVO stroke may warrant reliable predictors of functional outcomes. This study aimed to investigate the prognostic value of follow-up infarct volume (FIV) for predicting 90-day functional outcomes in MeVO stroke patients undergoing MT. This multicenter, retrospective cohort study analyzed data from the Multicenter Analysis of primary Distal medium vessel occlusions: effect of Mechanical Thrombectomy (MAD-MT) registry, covering patients with acute ischemic stroke due to M2 segment occlusion treated with MT. We examined the relationship between 90-day functional outcomes, measured by the modified Rankin Scale (mRS), and follow-up infarct volume (FIV), assessed through CT or MRI within 12-36 h post-MT. Among 130 participants, specific FIV thresholds were identified with high specificity and sensitivity for predicting outcomes. A FIV ⩽5 ml was highly specific for predicting favorable and excellent outcomes. The optimal cut-off for both prognostications was identified at ⩽15 ml by the Youden Index, with significant reductions in the likelihood of favorable outcomes observed above a 40 ml threshold. Receiver Operator Curve (ROC) analyses confirmed FIV as a superior predictor of functional outcomes compared to traditional recanalization scores, such as final modified thrombolysis in cerebral infarction score (mTICI). Multivariable analysis further highlighted the inverse relationship between FIV and positive functional outcomes. FIV within 36 h post-MT serves as a potent predictor of 90-day functional outcomes in patients with M2 segment MeVO strokes. Establishing FIV thresholds may aid in the prognostication of stroke outcomes, suggesting a role for FIV in guiding post intervention treatment decisions and informing clinical practice. Future research should focus on validating these findings across diverse patient populations and exploring the integration of FIV measurements with other clinical and imaging markers to enhance outcome prediction accuracy.

Neural networks in oncourology

In recent decades, neural networks have been widely applied in many fields of science and medicine. Accurate and early diagnosis of malignancies is a key challenge in oncology. Neural networks can analyse a wide range of medical data and identify relationships between qualitative and quantitative features. This allows for more precise and timely diagnoses. Moreover, they can be used to predict tumour progression, evaluate treatment effectiveness, and optimise treatment plans for each patientIn oncourology, the use of neural networks offers new perspectives for the diagnosis, prognosis, and treatment of various cancer conditions related to the urinary tract and male reproductive system. This review article explores how neural networks are being used in this field and present research into the use of neural networks for diagnosing, predicting the course and treating urological oncological diseases. The advantages and limitations of using neural networks in this field are demonstrated, and possible directions for future research are suggested. The application of neural networks in oncourology opens new horizons for the development of a personalised approach to diagnosing and treating oncological diseases. Artificial intelligence has the potential to become a powerful tool for improving the accuracy of patient outcome predictions and reducing undesirable side effects of therapy. Introducing neural networks into oncourological practice creates new opportunities for enhancing the work of healthcare organisations and improving the quality of care provided to patients. This can lead to better treatment outcomes and improved patient satisfaction.

Impact of crop management practices on maize yield: Insights from farming in tropical regions and predictive modeling using machine learning

Understanding the influence of crop management practices on maize yield is crucial amidst the changing environmental conditions. Smallholder farming in tropical regions has long puzzled decision-makers in terms of maize management. Balancing alternative management practices, environmental factors, and economic outcomes is essential. In this study, we examine the relationships between maize yield and various factors including climate variables, soil quality parameters, cultivars, tillage practices, and fertilizer usage in Chiapas, Mexico. Pearson's correlation coefficient and T-test were employed to determine the statistical significance of the correlations. Our findings reveal strong positive associations between maize yields and factors such as planting, cultivar, vapor pressure, temperature, solar radiations and fertilizer inputs for nitrogen, phosphorus, and potassium in lower elevation farms. Conversely, there is a negative correlation with elevation, slope, and system. Extensive data analysis was conducted to investigate the impact of different crop management practices on yield, utilizing both data visualization and analytics. Additionally, maize yield prediction was performed using six hybrid machine learning (ML) models, employing Grid Search and Random Search optimization techniques. Based on evaluation, Grid Search-based XGBoost exhibited the most accurate prediction results. Furthermore, explainable artificial intelligence (XAI) was employed to assess the individual impact of each input parameter on the ML model output.

Predictive models and biomarkers for survival in stage III breast cancer: a review of clinical applications and future directions

Stage III breast cancer, characterized by locally advanced tumors and potential regional lymph node involvement, presents a formidable challenge to both patients and healthcare professionals. Accurate prediction of survival outcomes is crucial for guiding treatment decisions and optimizing patient care. This publication explores the potential clinical utility of predictive tools, encompassing genetic markers, imaging techniques, and clinical parameters, to improve survival outcome predictions in stage III breast cancer. Multimodal approaches, integrating these tools, hold the promise of delivering more precise and personalized predictions. Despite the inherent challenges, such as data standardization and genetic heterogeneity, the future offers opportunities for refinement, driven by precision medicine, artificial intelligence, and global collaboration. The goal is to empower healthcare providers to make informed treatment decisions, ultimately leading to improved survival outcomes and a brighter horizon for individuals facing this challenging disease.

A comprehensive comparison of film cooling characteristics of shaped holes on the leading edge of rotating blade

This study numerically investigates the cooling characteristics and flow features of various hole geometries on the rotating blade leading edge. A comprehensive comparison of cooling effectiveness, heat transfer coefficients, net heat flux reduction, and discharge coefficients is conducted among laidback fan-shaped, fan-shaped, laidback, cone holes, and cylindrical holes with two different diameters. Detailed flow and thermal field analysis reveals the impact of geometry on cooling performance. The blade rotational speed is 600 rpm, with a Reynolds number of 12.4 × 104, blowing ratios from 0.5 to 2.0, and fixed injection and spanwise angles at 90° and 30°. Shaped holes share a uniform hole exit-to-inlet area ratio, and cylindrical holes match their inlet or exit areas. The RANS equations were closed using the SST k-ω turbulence model coupled with the Gamma Theta transition model, and an unstructured grid of 13.0 million cells was employed to attain relatively accurate predictive results. Results show that the lateral expansion on the leading edge significantly enhances cooling performance, with fan-shaped holes outperforming others, especially at high blowing ratios. Additionally, cooling effectiveness and discharge coefficient on the leading edge correlate strongly with hole geometric parameters. Shaped holes with equal area ratios show similar discharge coefficients. Cooling effectiveness at high blowing ratios is proportional to the effective exit width, provided that the coolant flows smoothly without severe lift-off.

Limitations of Separating Athletes into High or Low-Risk Groups based on a Cut-Off. A Clinical Commentary.

Athlete injury risk assessment and management is an important, yet challenging task for sport and exercise medicine professionals. A common approach to injury risk screening is to stratify athletes into risk groups based on their performance on a test relative to a cut-off threshold. However, one potential reason for ineffective injury prevention efforts is the over-reliance on identifying these 'at-risk' groups using arbitrary cut-offs for these tests and measures. The purpose of this commentary is to discuss the conceptual and technical issues related to the use of a cut-off in both research and clinical practice. How can we better assess and interpret clinical tests or measures to enable a more effective injury risk assessment in athletes? Cut-offs typically lack strong biologic plausibility to support them; and are typically derived in a data-driven manner and thus not generalizable to other samples. When a cut-off is used in analyses, information is lost, leading to potentially misleading results and less accurate injury risk prediction. Dichotomizing a continuous variable using a cut-off should be avoided. Using continuous variables on its original scale is advantageous because information is not discarded, outcome prediction accuracy is not lost, and personalized medicine can be facilitated. Researchers and clinicians are encouraged to analyze and interpret the results of tests and measures using continuous variables and avoid relying on singular cut-offs to guide decisions. Injury risk can be predicted more accurately when using continuous variables in their natural form. A more accurate risk prediction will facilitate personalized approaches to injury risk mitigation and may lead to a decline in injury rates. 5.

UCN-Centric Prognostic Model for Predicting Overall Survival and Immune Response in Colorectal Cancer.

Colorectal cancer (CRC), a prevalent malignancy, ranks third in global incidence and second in mortality rates. Despite advances in screening methods such as colonoscopy, the accurate diagnosis of CRC remains challenging due to the absence of reliable biomarkers. This study aimed to develop a robust prognostic model for precise CRC outcome prediction. Employing weighted co-expression network analysis (WGCNA) and Cox regression analysis on data from The Cancer Genome Atlas (TCGA), we identified a panel of 12 genes strongly associated with patient survival. This gene panel facilitated accurate CRC outcome predictions, which is also validated via the external validation cohort GSE17536. We conducted further investigations into the key gene, urocortin (UCN), using single-cell transcriptomic data and immune infiltration analysis in CRC patients. Our results revealed a significant correlation between high UCN expression and the reduced prevalence of key immune cells, including B cells, CD4+ cytotoxic T cells, CD8+ T cells, and NKT cells. Functional experiments showed that UCN gene interference in the CRC cell lines significantly decreased cancer cell proliferation, underscoring UCN's role in intestinal immunity modulation. The UCN-centric prognostic model developed enhances prognosis prediction accuracy and offers critical insights for CRC diagnosis and therapeutic interventions.

Study on the bending behavior of CFRP-confined square concrete-filled steel tubes reinforced with internal latticed steel angles

The bending mechanical properties of FRP-confined square concrete-filled steel tubes (CFST) reinforced with internal latticed steel angles, including the failure modes, characteristic curves and characteristic mechanical parameters of tested specimens, were investigated through experiments in this paper. The bending mechanisms were also analyzed using an established finite element model, including the analysis of the bending moment borne by each component, the distributions of contact pressure between different components at the mid-span section, and the distributions of the neutral axis under the bending capacity. The key influential parameters, including the tensile strength and thickness of the FRP, were also examined. The results indicate that the FRP has almost no influence on bending stiffness, however, it enhances the yield bending moment and bending capacity to a certain extent. It was also found that the enhancement decreases as the width of steel tube increases, and the influence of FRP is more profound before reaching the peak load. The calculation formulas for the bending capacity were developed using the superposition method and numerical fitting method. Analysis of their accuracy shows that the formulation using the superposition method yields relatively accurate and safe prediction results, with the prediction errors of less than 20 %.

Wave spectrum fitting with multiple parameters based on optimization algorithms and its application

In the realm of ocean engineering and environmental research, the accuracy of the wave spectrum is paramount. Traditional models, constrained by the number of parameters, often struggle to accurately capture the intricacies of real-world wave spectra, especially when confronting the complexities of multi-peak spectral shapes shaped by wind-sea and swell systems. This research develops a novel multi-parameter wave spectrum model that substantially amplifies the capacity to describe the diversity of wave spectra by introducing more shape parameters. The proposed model includes more shape parameters than its predecessors. These parameters can be directly optimized and obtained through the Heterogeneous Comprehensive Learning Particle Swarm Optimizer (HCLPSO). This sophisticated algorithm leverages its robust global search capabilities and swift convergence to fine-tune the model’s parameters, ensuring a high degree of precision in fitting the measured wave spectra. Furthermore, the proposed model also shows great potential in wave spectrum prediction, even when there are fewer training samples, it can still yield relatively accurate predictive results. This research not only improves the applicability and accuracy of the wave spectrum model but also offers a new tool for ocean wave research and real-world applications, thereby contributing important research and practical value to the field.

Implementation of Ensemble Learning Algorithm - Stacking Regressor on PM2.5 Prediction Model

This study aims to develop a PM2.5 concentration prediction model using ensemble learning techniques, focusing on the application of stacking regressor. The developed model is compared with several other basic models, namely LSTM, Random Forest, XGBoost, and GBM, to evaluate its performance in terms of prediction accuracy. The results show that the stacking regressor model provides more accurate prediction results than these basic models. The LSTM model has a Mean Squared Error (MSE) value of 80.42 and a coefficient of determination (R²) of -0.29, which shows unsatisfactory performance despite being able to capture temporal patterns. Random Forest gave better results with an MSE of 10.90 and R² of 0.83, thanks to its ability to handle heterogeneous data. The XGBoost and GBM models showed similar performance, with MSE of 9.01 and 8.81 respectively and R² of 0.86. However, the stacking regressor model with the RidgeCV meta-learner achieved the best results with an MSE of 8.07 and R² of 0.87, indicating that this technique successfully combines the advantages of various base models to improve prediction accuracy. In conclusion, the stacking regressor model proved effective in improving PM2.5 prediction accuracy, making it a potential tool for air quality monitoring and supporting public health policy making. This research makes an important contribution to the development of ensemble learning-based air quality prediction methods, and the results can serve as a reference for future research

State of Health Prediction in Electric Vehicle Batteries Using a Deep Learning Model

Accurately estimating the state of health (SOH) of lithium-ion batteries plays a significant role in the safe operation of electric vehicles. Deep learning (DL)-based approaches for estimating state of health (SOH) have consistently been the focus of study in recent years. In the current era of electric mobility, the utilization of lithium-ion batteries (LIBs) has evolved into a necessity for energy storage. Ensuring the safe operation of EVs requires a precise assessment of the state-of-health (SOH) of LIBs. To estimate battery SOH accurately, this paper employs a deep learning (DL) algorithm to enhance the estimation accuracy of SOH to obtain accurate SOH measurements. This research introduces the Diffusion Convolutional Recurrent Neural Network (DCRNN) with a Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithm (DCRNN + SVM-RFE) for enhancing classification and feature selection performance. The data gathered from the dataset were pre-processed using the min–max normalization method. The Center for Advanced Life Cycle Engineering (CALCE) dataset from the University of Maryland was employed to train and evaluate the model. The SVM-RFE algorithm was used for feature selection of pre-processed data. DCRNN algorithm was used for the classification process to enhance prediction precision. The DCRNN + SVM-RFE model’s performance was calculated using Mean Absolute Percentage Error (MAPE), Mean Squared Error (MAE), Mean Squared Error (MSE), and Root MSE (RMSE) metric values. The proposed model generates accurate results for SOH prediction; all RMSEs are within 0.02%, MAEs are within 0.015%, MSEs were within 0.032%, and MAPEs are within 0.41%. The mean values of RMSE, MSE, MAE, and MAPE were 0.014, 0.026, 0.011, and 0.32, respectively. Experiments confirmed that the DCRNN + SVM-RFE model has the highest accuracy among those that predict SOH.

Plaque enhancement of middle cerebral artery and pre-stroke diet are associated with prognosis of subacute ischemic stroke: A prospective high-resolution MR vessel wall imaging study

ObjectivesTo explore the value of middle cerebral artery (MCA) plaque characteristics in predicting the outcomes of subacute ischemic stroke and the incremental value of the previous diet on predictive performance. MethodsOne hundred and thirty-seven subacute ischemic stroke patients attributed to MCA plaques were included and analyzed in this prospective study. The National Institute of Health Stroke Scale (NIHSS) score, Mediterranean Diet Adherence Screener (MEDAS) score, and other clinical data were assessed. The plaque area, degree of stenosis, plaque burden, enhancement ratio, remodeling type, and intraplaque hemorrhage were measured using high-resolution MR vessel wall imaging (HR-VWI). Multivariable logistic regression analysis and receiver operating characteristic curve analysis were performed to assess the predictive performance of clinical and plaque characteristics for subacute ischemic stroke outcomes at 3 months. ResultsPatients with poor outcomes exhibited high NIHSS scores, and low MEDAS scores (P<0.001). Plaque burden, enhancement ratio, and degree of stenosis were significantly higher in patients with poor outcomes (P<0.001). Multivariate analyses further indicated that NIHSS score (P=0.001), MEDAS score (P=0.013), and enhancement ratio (P=0.011) were independent predictors of subacute ischemic stroke outcomes. The three models’ area under the curve (AUC) values were 0.811, 0.844, and 0.794. Combining these three factors resulted in an AUC of 0.908 (P<0.001). ConclusionsThe combination of NIHSS score, MEDAS score, and enhancement ratio showed significant superiority in the prognostic evaluation of subacute ischemic stroke. Clinical data combined with plaque characteristics improves the accuracy of 3-month outcome prediction on subacute ischemic stroke.

Exosome- Machine Learning Integration in Biomedicine: Advancing Diagnosis and Biomarker Discovery.

Exosomes, small extracellular vesicles (sEVs) secreted by various cell types, play crucial roles in intercellular communication and are increasingly recognized as valuable biomarkers for disease diagnosis and therapeutic targets. Meanwhile, machine learning (ML) techniques have revolutionized biomedical research by enabling the analysis of complex datasets and highly accurate prediction of disease outcomes. Exosomes, with their diverse cargo of proteins, nucleic acids, and lipids, offer a rich source of molecular information reflecting the physiological state of cells. Integrating exosome analysis with ML algorithms, including supervised and unsupervised learning techniques, allows for identifying disease-specific biomarkers and predicting disease outcomes based on exosome profiles. Integrating exosome biology with ML presents a promising avenue for advancing biomedical research and clinical practice. This review explores the intersection of exosome biology and ML in biomedicine, highlighting the importance of integrating these disciplines to advance our understanding of disease mechanisms and biomarker discovery.

Prediction of sepsis mortality in ICU patients using machine learning methods

ProblemSepsis, a life-threatening condition, accounts for the deaths of millions of people worldwide. Accurate prediction of sepsis outcomes is crucial for effective treatment and management. Previous studies have utilized machine learning for prognosis, but have limitations in feature sets and model interpretability.AimThis study aims to develop a machine learning model that enhances prediction accuracy for sepsis outcomes using a reduced set of features, thereby addressing the limitations of previous studies and enhancing model interpretability.MethodsThis study analyzes intensive care patient outcomes using the MIMIC-IV database, focusing on adult sepsis cases. Employing the latest data extraction tools, such as Google BigQuery, and following stringent selection criteria, we selected 38 features in this study. This selection is also informed by a comprehensive literature review and clinical expertise. Data preprocessing included handling missing values, regrouping categorical variables, and using the Synthetic Minority Over-sampling Technique (SMOTE) to balance the data. We evaluated several machine learning models: Decision Trees, Gradient Boosting, XGBoost, LightGBM, Multilayer Perceptrons (MLP), Support Vector Machines (SVM), and Random Forest. The Sequential Halving and Classification (SHAC) algorithm was used for hyperparameter tuning, and both train-test split and cross-validation methodologies were employed for performance and computational efficiency.ResultsThe Random Forest model was the most effective, achieving an area under the receiver operating characteristic curve (AUROC) of 0.94 with a confidence interval of ±0.01. This significantly outperformed other models and set a new benchmark in the literature. The model also provided detailed insights into the importance of various clinical features, with the Sequential Organ Failure Assessment (SOFA) score and average urine output being highly predictive. SHAP (Shapley Additive Explanations) analysis further enhanced the model’s interpretability, offering a clearer understanding of feature impacts.ConclusionThis study demonstrates significant improvements in predicting sepsis outcomes using a Random Forest model, supported by advanced machine learning techniques and thorough data preprocessing. Our approach provided detailed insights into the key clinical features impacting sepsis mortality, making the model both highly accurate and interpretable. By enhancing the model’s practical utility in clinical settings, we offer a valuable tool for healthcare professionals to make data-driven decisions, ultimately aiming to minimize sepsis-induced fatalities.

Fungal membrane determinants affecting sensitivity to antifungal cyclic lipopeptides from Bacillus spp.

Bacillus spp. produce numerous antimicrobial metabolites. Among these metabolites, cyclic lipopeptides (CLP) including fengycins, iturins, and surfactins are known to have varying antifungal activity against phytopathogenic fungi. The differential activities of CLP have been attributed to diverse mechanisms of action on fungal membranes. However, the precise biochemical determinants driving their antifungal modes of action have not been conclusively identified. In this study, three plant pathogenic fungi of varying lipopeptide sensitivities, Alternaria solani, Cladosporium cucumerinum, and Fusarium sambucinum, were studied to determine how their cell membrane lipid compositions may confer sensitivity and/or tolerance to fengycin, iturin, and surfactin. Results indicated that sensitivity to all three lipopeptides correlated with lower ergosterol content and elevated phospholipid fatty acid unsaturation. Fungal sensitivity to surfactin was also notably different than fengycin and iturin, as surfactin was influenced more by lower phosphatidylethanolamine amounts, higher levels of phosphatidylinositol, and less by phospholipid fatty acyl chain length. Results from this study provide insight into the fungal membrane composition of A. solani, F. sambucinum, and C. cucumerinum and the specific membrane characteristics influencing the antifungal effectiveness of fengycin, iturin, and surfactin. Understanding of these determinants should enable more accurate prediction of sensitivity-tolerance outcomes for other fungal species exposed to these important CLP.

Problems and Countermeasures in the Water Injection Development Process of Strong Water Sensitive Sand Conglomerate Reservoir

Due to the unique physicochemical property of strong water sensitive glutenite reservoirs, a series of problems may occur during water injection development, seriously affecting production efficiency and economic benefits. This study analyzed the problems and their causes in the water injection development process, constructed a water injection optimization model based on machine learning algorithms, and rigorously verified and evaluated the performance of the model. The results show that the algorithm proposed in this article has achieved significant advantages in prediction accuracy, reducing prediction error by 28.26%, and maintaining a high recall. In addition, as the sample size increases, the algorithm in this article performs more stably. When facing complex reservoir environments and a large amount of data, this algorithm can provide more accurate and reliable prediction results, providing strong support for the optimization of reservoir water injection development process. The research results can provide an effective decision support tool for the field of petroleum engineering, which helps to improve the efficiency and economy of oil extraction, and promote the sustainable growth of the petroleum industry.

An intelligent hybrid deep learning model for rolling bearing remaining useful life prediction

ABSTRACT Remaining Useful Life (RUL) prediction of rolling bearings is one of the intricate and important issues for equipment intelligent maintenance and health management. Various machine learning models and methods have been applied to rolling bearing RUL prediction. However, a single model cannot effectively extract state information and obtain accurate prediction results, and its generalisation is not stable under the condition of small sample data. Therefore this paper proposes an intelligent hybrid deep learning model for achieving accurate RUL prediction of rolling bearings. Firstly, the one-dimensional vibration signal is transformed into the corresponding two-dimensional time-frequency diagram via Continuous Wavelet Transform (CWT). Secondly, the diagram is input into a Multilayer Perceptron (MLP) consisting of a basic three-layer feed-forward network to obtain a one-dimensional feature vector. And lastly, the obtained feature vector is input into an integrated model based on Deep Autoregressive and Transformer to produce the probability distribution and obtain the prediction results of rolling bearing RUL. Extensive experiments on two rolling bearing datasets show that the proposed model outperforms six other comparative models in extracting bearing fault features and predicting bearing RUL, which demonstrates that the proposed model can effectively extract bearing fault features and accurately predict bearing remaining useful life.

Landslide susceptibility prediction based on landform predisposing indexes − An example from the Beiluo River Basin

The Beiluo River basin, which flows through the central part of the Loess Plateau, has experienced intense soil erosion and significant geomorphic change, which has provided favorable conditions for the occurrence of a large number of landslides. Landform indexes, which can express geomorphologic development state and internal rules, can transfer the development process information of surface morphology into the evaluation of landslide susceptibility, and help get more accurate landslide susceptibility prediction results. Taking the Beiluo River Basin as an example, a landslide susceptibility prediction model based on landform index is proposed by comparing the importance of landform index. In order to improve the accuracy of LSP, 10 kinds of general predictors indexes, 5 kinds of landform predisposing indexes and 1821 landslide points were compiled, and the geographic information system of Beiluo River Basin was constructed. Through the correlation test and CF model, the environmental indexes were evaluated to obtain the sensitive index results, and the combination of different environmental predictors indexes were classified according to the sensitive index results. Based on the combined classification results, the Max Entropy (MaxEnt) model was used to evaluate the Landslide susceptibility prediction (LSP), while the calculated results were evaluated and compared using the receiver operating characteristic (ROC) curve and landslide density. The results show that the vertical erosion factor, elevation, rainfall, horizontal erosion factor, slope angle and NDVI play a key role in controlling the spatial distribution of landslides in the study area. At the same time, the accuracy of landslide susceptibility is compared by AUC value. According to the calculation results, the Group5 (AUC = 0.803) with reasonable terrain index performs better in the training and test stages, and the relative accuracy is improved by 6.22 % compared with the non-introduction of terrain index and the omission rate difference is the best (omission rate difference = 0.0005), indicating that the introduction of landform index can effectively improve the landslide susceptibility prediction. The distribution of different sensitive areas was observed. The high sensitive areas and very high sensitive areas are mainly distributed in the southern Luochuan loess tableland and the northern Wuqi loess hilly area. The research results provide a scientific basis for landslide susceptibility prediction with rational introduction of landform indexes and regional infrastructure construction.