This paper addresses the challenges of dynamic environments and multimodal data fusion in multimodal transport path optimization for smart ports by proposing a GL-SSL Model that integrates Graph Neural Networks (GCN), Long Short-Term Memory (LSTM), and Self-Supervised Learning (SSL). The model fully exploits the graph-structured information of port transport networks and their temporal variations, while SSL enhances feature representation, enabling efficient optimization of path planning. Experiments were conducted on multiple public datasets, including AIS data from the Port of Rotterdam, global shipping data, and port net revenue data. Results show that the GL-SSL Model achieved significant improvements in key performance metrics. Specifically, the optimized path length reached 80 km, the transport cost was reduced to 200 cost-units (a composite metric reflecting fuel consumption, equipment wear, and labor cost), and the delay rate was maintained at 0.05 (5%), all of which are substantially better than traditional algorithms and other deep learning models. Furthermore, the model demonstrated stable performance under complex scenarios such as peak traffic, adverse weather, and equipment failures, with rapid convergence of training loss and strong robustness. These findings highlight the model’s adaptability and practical application potential. Overall, this work provides effective technical support for multimodal transport path optimization in smart ports and carries important theoretical significance and broad application prospects.

The Corsi Block Tapping Test (CBT) assesses short-term visuospatial memory, while the Walking Corsi Test (WalCT) introduces greater motor and spatial demands. The impact of visual trajectory presentation on these tests remains debated. To examine the effect of visual presentation of the trajectory on visuospatial short-term memory and topographic memory during the CBT and WalCT. A total of 37 students completed the Corsi task paradigm under two conditions: CBT and WalCT. Both were performed in classical versions (with visual trajectory presentation) and automated versions (without visual trajectory presentation). Each test was conducted in forward and backward modalities. Span and Total Product values were recorded. Sex, age and physical activity level were considered in the analysis. For Span, only the type of test influenced performance, with higher results observed in the CBT (7.29 ± 1.13) compared to the WalCT (6.18 ± 1.55) (p < 0.001). Regarding Total Product, significant effects were found for both the type of test (CBT = 84.83 ± 26.21; WalCT = 58.76 ± 28.99; p = 0.026) and the modality (Forward = 74.85 ± 31.56; Backward = 68.74 ± 29.23; p = 0.026). Age significantly interacted as a covariate in both analyses (p < 0.001). The visual presentation of the trajectory did not improve performance. However, the type and modality of the test directly influenced final performance. Additionally, age emerged as a factor affecting performance in the Corsi paradigms, while physical activity level and sex showed no significant effects.

This study investigates the effectiveness of embedding fiber Bragg grating (FBG) sensors in power transmission towers to assess the remaining service life of the structures following impacts from strong winds and hurricanes. FBG sensors monitor the structural integrity of the tower using online measurement of strain variations at critical structural points. The novelty of this work lies in employing a compact long short-term memory (LSTM) framework to estimate the remaining useful lifetime (RUL) from real-time FBG sensor data under both stable and fluctuating wind conditions. To estimate RUL of the tower, LSTM neural network has been implemented, providing predictive insights for proactive maintenance and risk mitigation. A prototype transmission tower was built and experimentally evaluated in a wind tunnel to assess the effectiveness and performance of the proposed RUL model. To simulate different hurricane categories, the experiment was conducted across wind speeds between 0 and 150 mph. FBG sensors installed at critical locations continuously captured real-time strain data, which was transmitted via a low-power micro FBG interrogator to a computer for input into the RUL prediction model. The proposed three-layer LSTM converges rapidly, reducing training and validation loss by nearly two orders of magnitude within 40 epochs, and achieves robust RUL predictions with an average bias of about 50 s on the test set. To quantify structural health, a mathematical health indicator was formulated based on the observed strain responses. The FBG sensors demonstrated high effectiveness in accurately detecting strain variations and monitoring the tower's dynamic behavior under extreme wind loads. The findings support the implementation of condition-based maintenance strategies, enhance safety assessments, and enable early failure detection. This approach not only improves operational reliability but also facilitates timely intervention and maintenance during critical events. Received: 26 June 2025 | Revised: 12 September 2025 | Accepted: 11 November 2025 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement Data are available from the corresponding author upon reasonable request. Author Contribution Statement Yu Shi: Methodology, software, formal analysis, data curation, writing – original draft, writing – review &amp; editing, visualization, supervision, and project administration. Abolghassem Zabihollah: Conceptualization, methodology, software, validation, investigation, resources, data curation, writing – original draft, writing – review &amp; editing, visualization, supervision, and project administration. Yao-Chi Yu: Methodology, software, formal analysis, data curation, and writing – original draft. Arunima Pathak: Investigation, writing – original draft, and visualization. Oluwaseyi Oyetunji: Investigation.

Schizophrenia is a complex and chronic psychiatric disorder that affects millions worldwide, significantly impairing cognitive and emotional functioning. Early detection is crucial for improving patient outcomes, yet traditional diagnostic approaches remain subjective and often delayed. Recent advances in Artificial Intelligence (AI) and Machine Learning (ML) have paved the way for automated and more accurate schizophrenia detection. This review explores various ML techniques, including traditional algorithms such as Support Vector Machines (SVM) and Random Forests, as well as deep learning models like Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks. The study delves into different data modalities, including neuroimaging (MRI, fMRI), Electroencephalography (EEG), and clinical assessments, highlighting their role in early schizophrenia identification. Performance evaluation metrics such as accuracy, sensitivity, specificity, and the Area Under the Curve (AUC) are analyzed across multiple studies. Challenges such as data heterogeneity, small sample sizes, and ethical concerns are discussed alongside potential solutions, including explainable AI and multimodal integration. Future directions emphasize real-time monitoring, AI-powered mobile applications, and the development of robust, generalizable models. This review underscores the transformative potential of ML in mental health diagnostics and its role in revolutionizing schizophrenia detection and intervention strategies. Major Findings: This review highlight that Machine Learning (ML) algorithms, particularly support vector machines, Random Forests, CNNs, and LSTMs, have demonstrated promising accuracy in early schizophrenia detection. Neuroimaging, EEG, and clinical assessments serve as key data sources, with multimodal approaches enhancing diagnostic precision. Performance metrics indicate strong potential, though variability exists due to data quality and sample size. Challenges such as data heterogeneity and ethical concerns persist, but advances in explainable AI and integration of diverse data types offer solutions. The future lies in real-time, mobile, and generalizable AI models that support early, accessible, and objective mental health interventions.

BackgroundNeural stem cells (NSCs), crucial for brain function and repair, are disrupted by high-fructose diet (HFrD) in proliferation and survival, linking to neurogenesis impairment and neuropsychiatric risks. Mechanistic insights remain undefined. MethodsComprehensive behavioral assessments were conducted on HFrD mice, including the tail suspension test (TST) and sucrose preference test (SPT) for depressive-like behaviors, elevated plus maze (EPM) and open field test (OFT) for anxiety-like behaviors, as well as novel object recognition (NOR) and Morris water maze (MWM) for cognition. Hippocampal NSCs and newborn neurons were quantified by immunofluorescence, and fructose-treated NE-4C cells underwent RNA sequencing (RNA-seq) analysis coupled with measurements of proliferation, apoptosis and ferroptosis markers.ResultsHFrD mice showed depressive-like behaviors without anxiety-like behaviors, and exhibited impaired short-term memory in NOR but did not show impaired spatial memory in MWM. Decreased number of hippocampal NSCs and newborn neurons were observed, suggesting impaired neurogenesis. In vitro, fructose-treated NE-4c exhibited altered gene expression profiles, with PCA showing distinct clustering between treated and control groups. Further analysis (GO, KEGG, GSEA) indicated enrichment in energy metabolism pathways, including mitochondrial ATP synthesis (e.g., downregulated ATP5E, ATP5H). Consistently, intracellular ATP levels were elevated, indicating metabolic dysregulation. Further experiments demonstrated that high fructose promoted NSC proliferation via p53/Wnt pathways (upregulated CyclinA2, CDK1) while concurrently inducing apoptosis (BAX, P53 upregulation) and ferroptosis (reduced GPX4, elevated ROS, and lipid peroxidation). ConclusionThis study elucidates the mechanistic link between HFrD-induced metabolic disruption and NSC dysfunction, providing novel insights into the pathogenesis of fructose-associated neuropsychiatric disorders.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12986-025-01043-7.

All-optical artificial synaptic devices offer promising potential for neuromorphic computing, yet their development is hindered by limited spectral tunability and poor plasticity linearity. Here, a broadband all-optical synaptic memtransistor based on organic charge transfer cocrystals (DTT-TCNQ) is reported, which enables fully light-driven and reversible modulation of synaptic weights across a wide wavelength range (395-808nm). The device exhibits bidirectional excitatory and inhibitory photoresponses, and achieves highly linear long-term potentiation and depression (LTP/LTD) characteristics with ultralow nonlinearity (αp = 0.00191, αd = 0.00305) and asymmetry ratio (AR = 0.00114), attributed to a synergistic strategy combining frequency modulation and photoelectric coupling. When integrated into a convolutional neural long short-term memory network (CNN-LSTM)hybrid network, the device enables rapid convergence (98.77% accuracy in 6 training epochs) and robust recognition performance under spatiotemporal noise, outperforming conventional light-write/electric-erasing schemes. This work bridges material-level innovation and system-level functionality, offering a scalable approach toward energy-efficient, noise-resilient neuromorphic vision systems.

Long short-term memory network with exponential gating mechanism: a deep learning approach for cardiovascular stroke risk stratification

This work investigated machine-learning algorithms for remote-control and autonomous operation of the Very-Small, Long-Life, Modular (VSLLIM) microreactor. This walk-away safe reactor can continuously generate 1.0–10 MW of thermal power for 92 and 5.6 full power years, respectively, is cooled by natural circulation of in-vessel liquid sodium, does not require on-site storage of either fresh or spent nuclear fuel, and offers redundant means of control and passive decay heat removal. The two ML algorithms investigated are Supervised Learning with Long Short-Term Memory networks (SL-LSTM) and Soft-Actor Critic with Feedforward Neural Networks (SAC-FNN). They are trained to manage the movement of the control rods in the reactor core during various transients including startup, shutdown, and to change the reactor steady state power up to 10 MW. The trained algorithms are incorporated into a Programmable Logic Controller (PLC) coupled to a digital twin dynamic model of the VSLLIM microreactor. Although the SL-LSTM algorithms demonstrate high prediction accuracy of up to 99.95%, they demonstrate inferior performance when incorporated into the PLC. Conversely, the PLC with SAC-FNN algorithm accurately adjusts the control rods positions during the reactor startup transients to within ±1.6% of target values.

In this study, we explore the integration of a cost-effective triboelectric nanogenerator (TENG) with an large silicon PIN detector (diameter: 12 mm) for intelligent wireless recognition applications. Wireless communication eliminates the need for physical connections, enabling greater flexibility and scalability in deployment. It allows for seamless integration of AI systems into a wide range of environments without the constraints of wiring, reducing installation complexity and enhancing mobility. Additionally, we demonstrate the TENG’s functionality as an autonomous communication unit. The TENG is employed to convert various environmentally triggered signals into digital formats and to autonomously power optoelectronic devices, thus eliminating the need for an external power supply. By integrating optoelectronic components within the self-powered sensing system, the TENG can identify specific trigger information and reduce extraneous noise, thereby improving the accuracy of information transmission. Moreover wireless technology facilitates real-time data transmission and processing. This setup not only enhances the overall efficiency and adaptability of the system but also supports continuous operation in diverse and dynamic settings. This paper introduces a novel convolutional neural network-long short-term memory (CNN-LSTM) fusion neural network model. Utilizing the sensing system in combination with the CNN-LSTM neural network enables the collection and identification of variations in the flicker frequency and luminosity of optoelectronic devices. This capability allows for the recognition of environmental trigger signals generated by the TENG. The classification and recognition results of human body trigger signals indicate a recognition accuracy of 92.94%.

A hybrid power load forecasting model based on evolutionary strategy and long short term memory

Meteorological drought severity forecasting utilizing blended modelling.

Tracking is now popular in real world. Precise tracking of objects in real-time videos is a challenging task. With billions of fans, football is a rapidly expanding sport that has proven essential to many nations and their citizens in particular. None of the numerous great target tracking algorithms have surfaced in recent years primarily deep learning and correlation filtering that can track players in soccer game videos with high accuracy. In this paper, the proposed system is used You Only Look Once version 8-nano (YOLOv8n) for Multi-Object Detection (MOD) to get higher detection accuracy results. Moreover, this system is based on the hybrid method for tracking. The hybrid method is combined with stacked Long Short Term Memory (LSTM) and Fairness of Detection and Re-identification in Multipe Object Tracking (FairMOT). The experimental analysis shows that the proposed system is efficiently and better accuacy because the best detection results with YOLOv8n is 93% for precision, 91% for recall and 92% for mAP(50) with own dataset. After using the proposed system, the average of the Multi Object Tracking Accuracy (MOTA) is 80 % at IoU-Threshold 0.5, the average of the Multi Object Tracking Precision (MOTP) is 89% at IoU-Threshold 0.8 and the average of the final mAP is 96% at IoU- Threshold 0.5 by using hybrid method for tracking.

This study explores the application of Long Short-Term Memory (LSTM) neural networks for predicting short-term price movements of the TON-IRT trading pair in the cryptocurrency market. Given the high volatility and complexity of cryptocurrency prices, traditional models like Linear Regression and ARIMA often fail to capture the underlying non-linear and temporal dependencies. To address this, we implemented an LSTM model, a type of recurrent neural network specifically designed for sequential data. The model was trained on historical hourly data, utilizing various technical indicators and lagged features to improve prediction accuracy. Our results demonstrated that the LSTM model significantly outperformed traditional methods, achieving a Mean Absolute Error (MAE) of 0.0274, a Root Mean Squared Error (RMSE) of 0.0321, and an R-squared (R²) value of 0.8743, which indicated that the model captured over 87% of the variance in the actual price data. Visual analysis of predicted versus actual prices revealed a strong alignment, though some lag in predictions during high-volatility periods was observed. The model also showed a tendency to underestimate price peaks, highlighting areas for further refinement. This study contributes to the field of blockchain trading analytics by demonstrating the effectiveness of LSTM models in addressing the unique challenges of cryptocurrency price prediction. Practical implications for traders and investors include the ability to enhance trading strategies, optimize entry and exit points, and improve risk management. Future research could integrate additional external factors, such as market sentiment and news events, or explore advanced architectures like Transformer models. By doing so, the predictive capabilities of LSTM models in volatile markets like cryptocurrency could be further refined, leading to more robust and accurate forecasting tools for financial decision-making.

Cardiotocography (CTG) is a widely used technique for fetal monitoring. This study presents CTGFusionNet, a novel multimodal adaptive framework designed for prenatal analysis. The framework integrates attention-based adaptive Bi-Directional Convolutional Neural Networks (Bi-CNN) with Long Short-Term Memory (LSTM) networks to improve the accuracy of fetal distress prediction. The methodology begins with an initial data preprocessing phase, followed by signal segmentation and enhancement. Thereafter, the FHR and UC signals are transformed into two-dimensional representations using embedding layers and subsequently integrated through concatenation. The spatial features of the synchronized signals are extracted using the proposed adaptive Bi-CNN. Multi-head attention is then applied to emphasize the most relevant information, and the temporal features are captured using an LSTM network. In the final stage, the most relevant features from the perinatal clinical data are identified using the Relief, Lasso, and Information Gain algorithms and then integrated with the processed signals. Furthermore, classification results are obtained using a fully connected layer and sigmoid function. The results demonstrate that CTGFusionNet leads to significant improvements in performance measures, namely accuracy, sensitivity, and specificity, with values of 97.85%, 97.07%, and 98.65%, respectively. This suggests that CTGFusionNet-a multimodal approach that combines FHR, UC, and clinical data, provides a more reliable and precise method for the early detection and prediction of fetal distress. The proposed approach has the potential to significantly improve prenatal care outcomes by enabling accurate interventions.

Deep learning approach for classifying grazing behavior in yearling horses using triaxial accelerometer data: A pilot study.

Abstract Remaining useful life (RUL) prediction of fuel cells is a topic of significant research attention. This paper presents a convolutional neural network (CNN)-attention hybrid model for predicting the RUL of fuel cells. We used road data to verify the predictive capabilities of the model and compared it with the long short-term memory (LSTM) model, CNN model, and attention model. The results showed that the CNN-attention hybrid model can effectively combine the local feature extraction capability of CNNs and global feature capture capability of attention to fully extract high-accuracy voltage attenuation information from the input data. At a training ratio of 70%, the mean absolute percentage error was 17% lower than that of the LSTM model, 25% lower than that of the CNN model, and 24% lower than that of the attention model. In terms of the stability of the model output results, the maximum deviation of four independent predictions did not exceed 0.5%, indicating that the model outputs have high stability. We further analyzed the long-term prediction performance of the model, and the results showed that the hybrid model can achieve effective long-term RUL prediction of fuel cells.

Various industries, such as healthcare and surveillance, depend heavily on the ability to recognize human activity. The “human activity recognition (HAR) using smartphones data set” can be found in the UCI online repository and includes accelerometer and gyroscope readings recorded during a variety of human activities. The accelerometer and gyroscope signals are also subjected to a band-pass filter to eliminate unwanted frequencies and background noise. This method effectively decreases the dimensionality of the feature space while improving the model's accuracy and efficiency. “Convolutional neural networks (CNNs)” and “long shortterm memory (LSTM)” networks are combined to create pyramidal dilated convolutional memory network (PDCMN), which is the final proposal. Results from experiments demonstrate the effectiveness and reliability of the suggested method, demonstrating its potential for precise and effective HAR in actuality schemes.

Emotion recognition (ER) is crucial for understanding human behaviours, social interactions, and psychological well-being. Electroencephalography (EEG) has emerged as a promising tool for capturing the neural correlates of emotions. This work is a systematic review of articles in ER using EEG signals. A total of 120 articles from 1041 articles were selected based on PRISMA guidelines using defined inclusion and exclusion criteria, published between 2018 and 2024. This article aims to provide an in-depth understanding of the current landscape of ER from EEG signals utilizing deep learning (DL). This review offers valuable guidance for researchers and practitioners seeking more refined and reliable emotion classification systems. To explore the effectiveness of DL models in EEG-based ER, several potential DL models, such as convolutional neural network, long short-term memory (LSTM), gated recurrent unit (GRU), hybrid bidirectional LSTM (BiLSTM), bidirectional GRU, and advanced DL models such as convolutional recurrent neural network and EEG-Conformer models are applied to two popular datasets, SEED and GAMEEMO, respectively, to depict the full process of ER. Additionally, the performance of DL models is also compared with the performance of basic machine learning (ML) models such as SVM, k-nearest neighbors, logistic regression, and boosting algorithms such as AdaBoost, XGBoost and LightGBM. Through extensive experiments and performance evaluations, the performance of different models when applied to the datasets mentioned above is compared. The accuracy, precision, recall, and F1-scores are analysed to determine the most effective model for EEG-based ER. The findings of this study demonstrate that the performance of hybrid DL models is more efficacious than that of ML models. The best-performing model (BiLSTM) classified the emotions, with an accuracy of 90.54% when applied to the GAMEEMO dataset. This research contributes to the growing body of literature on ER and provides insights into the feasibility of using EEG signals to understand emotional states, and presents a structured roadmap for future exploration. The findings can aid in the development of more accurate and reliable ER systems, which can have wide-ranging applications in psychology, social sciences, and human-computer interactions.

Are Vision-xLSTM-embedded U-Nets better at segmenting medical images?

Semantic-based audiovisual integration enhances active storage in visual working memory.