RESUMEN

Temporal interference deep-brain magnetic stimulation (TI-DMS) induces rhythmic electric field (EF) in the hippocampus to normalize cognitive function. The rhythmic time series of the hippocampal EF is essential for the assessment of TI-DMS. However, the finite element method (FEM) takes several hours to obtain the time series of EF. In order to reduce the time cost, the temporal convolutional network (TCN) model is adopted to predict the time series of hippocampal EF induced by TI-DMS. It takes coil configuration and loaded current as input and predicts the time series of maximum and mean values of the left and right hippocampal EF. The prediction takes only a few seconds. The model parameter combination of kernel size and layers is selected optimally by cross-validation method. The experimental results for multiple subjects show that the R2 of all the time series predicted by the model exceed 0.98. And the prediction accuracy is even higher as the input parameters approach the training set. These results demonstrate that the adopted model can quickly predict the time series of hippocampal EF induced by TI-DMS with relatively high accuracy, which is beneficial for future clinical applications.

RESUMEN

In many emerging nations, rapid industrialization and urbanization have led to heightened levels of air pollution. This sudden rise in air pollution, which affects global sustainability and human health, has become a significant concern for citizens and governments. While most current methods for predicting air quality rely on shallow models and often yield unsatisfactory results, our study explores a deep architectural model for forecasting air quality. We employ a sophisticated deep learning structure to develop an advanced system for ambient air quality prediction. We utilize three publicly available databases and real-world data to obtain accurate air quality measurements. These four datasets undergo a data cleaning to yield a consolidated, cleaned dataset. Subsequently, the Fused Eurasian Oystercatcher-Pathfinder Algorithm (FEO-PFA)-a dual optimization method combining the Eurasian Oystercatcher Optimizer (EOO) and Pathfinder Algorithm (PFA)-is applied. This method aids in selecting weighted features, optimizing weights, and choosing the most relevant attributes for optimal results. These optimal features are then incorporated into the Multiscale Depth-wise Separable Adaptive Temporal Convolutional Network (MDS-ATCN) for the ambient Air Quality Prediction (AQP) process. The variables within MDS-ATCN are further refined using the proposed FEO-PFA to enhance predictive accuracy. An empirical analysis is performed to compare the efficacy of our proposed model with traditional methods, underscoring the superior effectiveness of our approach. The average cost function is reduced to 5.5%, the MAE to 28%, and the RMSE to 14% by the suggested method, according to the performance research conducted with regard to all datasets.

RESUMEN

In this paper, the experimental results on microphone virtualization in realistic automotive scenarios are presented. A Temporal Convolutional Network (TCN) was designed in order to estimate the acoustic signal at the driver's ear positions based on the knowledge of monitoring microphone signals at different positions-a technique known as virtual microphone. An experimental setup was implemented on a popular B-segment car to acquire the acoustic field within the cabin while running on smooth asphalt at variable speeds. In order to test the potentiality of the TCN, microphone signals were recorded in two different scenarios, either with or without the front passenger. Our experimental results show that, when training is performed in both scenarios, the adopted TCN is able to robustly adapt to different conditions and guarantee a good average performance. Furthermore, an investigation on the parameters of the Neural Network (NN) that guarantee the sufficient accuracy of the estimation of the virtual microphone signals while maintaining a low computational complexity is presented.

RESUMEN

The increased risk of cardiovascular disease in people with spinal cord injuries motivates work to identify exercise options that improve health outcomes without causing risk of musculoskeletal injury. Handcycling is an exercise mode that may be beneficial for wheelchair users, but further work is needed to establish appropriate guidelines and requires assessment of the external loads. The goal of this research was to predict the six-degree-of-freedom external loads during handcycling from data similar to those which can be measured from inertial measurement units (segment accelerations and velocities) using machine learning. Five neural network models and two ensemble models were compared against a statistical model. A temporal convolutional network (TCN) yielded the best predictions. Predictions of forces and moments in-plane with the crank were the most accurate (r = 0.95-0.97). The TCN model could predict external loads during activities of different intensities, making it viable for different exercise protocols. The ability to predict the loads associated with forward propulsion using wearable-type data enables the development of informed exercise guidelines.

Asunto(s)

RESUMEN

Secondary structure prediction is a key step in understanding protein function and biological properties and is highly important in the fields of new drug development, disease treatment, bioengineering, etc. Accurately predicting the secondary structure of proteins helps to reveal how proteins are folded and how they function in cells. The application of deep learning models in protein structure prediction is particularly important because of their ability to process complex sequence information and extract meaningful patterns and features, thus significantly improving the accuracy and efficiency of prediction. In this study, a combined model integrating an improved temporal convolutional network (TCN), bidirectional long short-term memory (BiLSTM), and a multi-head attention (MHA) mechanism is proposed to enhance the accuracy of protein prediction in both eight-state and three-state structures. One-hot encoding features and word vector representations of physicochemical properties are incorporated. A significant emphasis is placed on knowledge distillation techniques utilizing the ProtT5 pretrained model, leading to performance improvements. The improved TCN, achieved through multiscale fusion and bidirectional operations, allows for better extraction of amino acid sequence features than traditional TCN models. The model demonstrated excellent prediction performance on multiple datasets. For the TS115, CB513 and PDB (2018-2020) datasets, the prediction accuracy of the eight-state structure of the six datasets in this paper reached 88.2%, 84.9%, and 95.3%, respectively, and the prediction accuracy of the three-state structure reached 91.3%, 90.3%, and 96.8%, respectively. This study not only improves the accuracy of protein secondary structure prediction but also provides an important tool for understanding protein structure and function, which is particularly applicable to resource-constrained contexts and provides a valuable tool for understanding protein structure and function.

Asunto(s)

RESUMEN

Mold contamination poses a significant challenge in the processing and storage of Chinese herbal medicines (CHM), leading to quality degradation and reduced efficacy. To address this issue, we propose a rapid and accurate detection method for molds in CHM, with a specific focus on Atractylodes macrocephala, using electronic nose (e-nose) technology. The proposed method introduces an eccentric temporal convolutional network (ETCN) model, which effectively captures temporal and spatial information from the e-nose data, enabling efficient and precise mold detection in CHM. In our approach, we employ the stochastic resonance (SR) technique to eliminate noise from the raw e-nose data. By comprehensively analyzing data from eight sensors, the SR-enhanced ETCN (SR-ETCN) method achieves an impressive accuracy of 94.3%, outperforming seven other comparative models that use only the response time of 7.0 seconds before the rise phase. The experimental results showcase the ETCN model's accuracy and efficiency, providing a reliable solution for mold detection in Chinese herbal medicine. This study contributes significantly to expediting the assessment of herbal medicine quality, thereby helping to ensure the safety and efficacy of traditional medicinal practices.

Asunto(s)

RESUMEN

Accurate penetration rate prediction enhances rock-breaking efficiency and reduces disc cutter damage in tunnel boring machine (TBM) construction. However, this process faces significant challenges such as the high uncertainty of ground conditions and the complexity of maintaining optimal TBM operation in long and large tunnels. To address these challenges, we propose TCN-SENet++, a novel hybrid multistep real-time penetration rate prediction model that combines a temporal convolutional network (TCN) and a squeeze-and-excitation (SENet) block for aided tunneling. This study aims to demonstrate the application of TCN-SENet++, as well as other models such as RNN, LSTM, GRU, and TCN, for TBM penetration rate prediction. The model was developed using actual datasets collected from the Yin-Song diversion project. We employ a 30-s time step to predict the future time steps of the penetration rate (1st, 3rd, 5th, 7th, and 9th). The features that influence the penetration rate, such as the cutterhead torque, thrust, and cutterhead power, were considered. A comparative analysis using the mean absolute error and mean squared error revealed that the TCN-SENet++ model outperformed the other models, including RNN, LSTM, GRU, TCN, and TCN-SENet+. In comparison, TCN-SENet++ achieved average MSE reductions of 18%, 6%, 3%, 1%, and 2%, respectively. The TCN-SENet++ model demonstrated fewer errors in the new project, validating its effectiveness and suitability for real-time penetration rate prediction in TBM construction.

RESUMEN

N4-acetylcysteine (ac4C) is a chemical modification in mRNAs that alters the structure and function of mRNA by adding an acetyl group to the N4 position of cytosine. Researchers have shown that ac4C is closely associated with the occurrence and development of various cancers. Therefore, accurate prediction of ac4C modification sites on human mRNA is crucial for revealing its role in diseases and developing new diagnostic and therapeutic strategies. However, existing deep learning models still have limitations in prediction accuracy and generalization ability, which restrict their effectiveness in handling complex biological sequence data. This paper introduces a deep learning-based model, STM-ac4C, for predicting ac4C modification sites on human mRNA. The model combines the advantages of selective kernel convolution, temporal convolutional networks, and multi-head self-attention mechanisms to effectively extract and integrate multi-level features of RNA sequences, thereby achieving high-precision prediction of ac4C sites. On the independent test dataset, STM-ac4C showed improvements of 1.81%, 3.5%, and 0.37% in accuracy, Matthews correlation coefficient, and area under the curve, respectively, compared to the existing state-of-the-art technologies. Moreover, its performance on additional balanced and imbalanced datasets also confirmed the model's robustness and generalization ability. Various experimental results indicate that STM-ac4C outperforms existing methods in predictive performance. In summary, STM-ac4C excels in predicting ac4C modification sites on human mRNA, providing a powerful new tool for a deeper understanding of the biological significance of mRNA modifications and cancer treatment. Additionally, the model reveals key sequence features that influence the prediction of ac4C sites through sequence region impact analysis, offering new perspectives for future research. The source code and experimental data are available at https://github.com/ymy12341/STM-ac4C.

RESUMEN

Accurate power load forecasting is crucial for the sustainable operation of smart grids. However, the complexity and uncertainty of load, along with the large-scale and high-dimensional energy information, present challenges in handling intricate dynamic features and long-term dependencies. This paper proposes a computational approach to address these challenges in short-term power load forecasting and energy information management, with the goal of accurately predicting future load demand. The study introduces a hybrid method that combines multiple deep learning models, the Gated Recurrent Unit (GRU) is employed to capture long-term dependencies in time series data, while the Temporal Convolutional Network (TCN) efficiently learns patterns and features in load data. Additionally, the attention mechanism is incorporated to automatically focus on the input components most relevant to the load prediction task, further enhancing model performance. According to the experimental evaluation conducted on four public datasets, including GEFCom2014, the proposed algorithm outperforms the baseline models on various metrics such as prediction accuracy, efficiency, and stability. Notably, on the GEFCom2014 dataset, FLOP is reduced by over 48.8%, inference time is shortened by more than 46.7%, and MAPE is improved by 39%. The proposed method significantly enhances the reliability, stability, and cost-effectiveness of smart grids, which facilitates risk assessment optimization and operational planning under the context of information management for smart grid systems.

RESUMEN

When resource demand increases and decreases rapidly, container clusters in the cloud environment need to respond to the number of containers in a timely manner to ensure service quality. Resource load prediction is a prominent challenge issue with the widespread adoption of cloud computing. A novel cloud computing load prediction method has been proposed, the Double-channel residual Self-attention Temporal convolutional Network with Weight adaptive updating (DSTNW), in order to make the response of the container cluster more rapid and accurate. A Double-channel Temporal Convolution Network model (DTN) has been developed to capture long-term sequence dependencies and enhance feature extraction capabilities when the model handles long load sequences. Double-channel dilated causal convolution has been adopted to replace the single-channel dilated causal convolution in the DTN. A residual temporal self-attention mechanism (SM) has been proposed to improve the performance of the network and focus on features with significant contributions from the DTN. DTN and SM jointly constitute a dual-channel residual self-attention temporal convolutional network (DSTN). In addition, by evaluating the accuracy aspects of single and stacked DSTNs, an adaptive weight strategy has been proposed to assign corresponding weights for the single and stacked DSTNs, respectively. The experimental results highlight that the developed method has outstanding prediction performance for cloud computing in comparison with some state-of-the-art methods. The proposed method achieved an average improvement of 24.16% and 30.48% on the Container dataset and Google dataset, respectively.

RESUMEN

PURPOSE: Most studies on surgical activity recognition utilizing artificial intelligence (AI) have focused mainly on recognizing one type of activity from small and mono-centric surgical video datasets. It remains speculative whether those models would generalize to other centers. METHODS: In this work, we introduce a large multi-centric multi-activity dataset consisting of 140 surgical videos (MultiBypass140) of laparoscopic Roux-en-Y gastric bypass (LRYGB) surgeries performed at two medical centers, i.e., the University Hospital of Strasbourg, France (StrasBypass70) and Inselspital, Bern University Hospital, Switzerland (BernBypass70). The dataset has been fully annotated with phases and steps by two board-certified surgeons. Furthermore, we assess the generalizability and benchmark different deep learning models for the task of phase and step recognition in 7 experimental studies: (1) Training and evaluation on BernBypass70; (2) Training and evaluation on StrasBypass70; (3) Training and evaluation on the joint MultiBypass140 dataset; (4) Training on BernBypass70, evaluation on StrasBypass70; (5) Training on StrasBypass70, evaluation on BernBypass70; Training on MultiBypass140, (6) evaluation on BernBypass70 and (7) evaluation on StrasBypass70. RESULTS: The model's performance is markedly influenced by the training data. The worst results were obtained in experiments (4) and (5) confirming the limited generalization capabilities of models trained on mono-centric data. The use of multi-centric training data, experiments (6) and (7), improves the generalization capabilities of the models, bringing them beyond the level of independent mono-centric training and validation (experiments (1) and (2)). CONCLUSION: MultiBypass140 shows considerable variation in surgical technique and workflow of LRYGB procedures between centers. Therefore, generalization experiments demonstrate a remarkable difference in model performance. These results highlight the importance of multi-centric datasets for AI model generalization to account for variance in surgical technique and workflows. The dataset and code are publicly available at https://github.com/CAMMA-public/MultiBypass140.

RESUMEN

Electroencephalography (EEG) plays a crucial role in epilepsy analysis, and epileptic seizure prediction has significant value for clinical treatment of epilepsy. Currently, prediction methods using Convolutional Neural Network (CNN) primarily focus on local features of EEG, making it challenging to simultaneously capture the spatial and temporal features from multi-channel EEGs to identify the preictal state effectively. In order to extract inherent spatial relationships among multi-channel EEGs while obtaining their temporal correlations, this study proposed an end-to-end model for the prediction of epileptic seizures by incorporating Graph Attention Network (GAT) and Temporal Convolutional Network (TCN). Low-pass filtered EEG signals were fed into the GAT module for EEG spatial feature extraction, and followed by TCN to capture temporal features, allowing the end-to-end model to acquire the spatiotemporal correlations of multi-channel EEGs. The system was evaluated on the publicly available CHB-MIT database, yielding segment-based accuracy of 98.71%, specificity of 98.35%, sensitivity of 99.07%, and F1-score of 98.71%, respectively. Event-based sensitivity of 97.03% and False Positive Rate (FPR) of 0.03/h was also achieved. Experimental results demonstrated this system can achieve superior performance for seizure prediction by leveraging the fusion of EEG spatiotemporal features without the need of feature engineering.

Asunto(s)

RESUMEN

Research evidence shows that physical rehabilitation exercises prescribed by medical experts can assist in restoring physical function, improving life quality, and promoting independence for physically disabled individuals. In response to the absence of immediate expert feedback on performed actions, developing a Human Action Evaluation (HAE) system emerges as a valuable automated solution, addressing the need for accurate assessment of exercises and guidance during physical rehabilitation. Previous HAE systems developed for the rehabilitation exercises have focused on developing models that utilize skeleton data as input to compute a quality score for each action performed by the patient. However, existing studies have focused on improving scoring performance while often overlooking computational efficiency. In this research, we propose LightPRA (Light Physical Rehabilitation Assessment) system, an innovative architectural solution based on a Temporal Convolutional Network (TCN), which harnesses the capabilities of dilated causal Convolutional Neural Networks (CNNs). This approach efficiently captures complex temporal features and characteristics of the skeleton data with lower computational complexity, making it suitable for real-time feedback provided on resource-constrained devices such as Internet of Things (IoT) devices and Edge computing frameworks. Through empirical analysis performed on the University of Idaho-Physical Rehabilitation Movement Data (UI-PRMD) and KInematic assessment of MOvement for remote monitoring of physical REhabilitation (KIMORE) datasets, our proposed LightPRA model demonstrates superior performance over several state-of-the-art approaches such as Spatial-Temporal Graph Convolutional Network (STGCN) and Long Short-Term Memory (LSTM)-based models in scoring human activity performance, while exhibiting lower computational cost and complexity.

Asunto(s)

RESUMEN

Brain Computer Interface (BCI) is a highly promising human-computer interaction method that can utilize brain signals to control external devices. BCI based on functional near-infrared spectroscopy (fNIRS) is considered a relatively new and promising paradigm. fNIRS is a technique of measuring functional changes in cerebral hemodynamics. It detects changes in the hemodynamic activity of the cerebral cortex by measuring oxyhemoglobin and deoxyhemoglobin (HbR) concentrations and inversely predicts the neural activity of the brain. At the present time, Deep learning (DL) methods have not been widely used in fNIRS decoding, and there are fewer studies considering both spatial and temporal dimensions for fNIRS classification. To solve these problems, we proposed an end-to-end hybrid neural network for feature extraction of fNIRS. The method utilizes a spatial-temporal convolutional layer for automatic extraction of temporally valid information and uses a spatial attention mechanism to extract spatially localized information. A temporal convolutional network (TCN) is used to further utilize the temporal information of fNIRS before the fully connected layer. We validated our approach on a publicly available dataset including 29 subjects, including left-hand and right-hand motor imagery (MI), mental arithmetic (MA), and a baseline task. The results show that the method has few training parameters and high accuracy, providing a meaningful reference for BCI development.

Asunto(s)

RESUMEN

A single network model exhibits limitations in the life prediction of rotating machinery for the various fault types and uncertain fault occurrence. Therefore, a network prediction model combining multi-domain feature fusion (MDFF) and distributed TCN-Attention-BiGRU (DITCN-ABiGRU) is proposed to enable a more accurate life prediction of rotating machinery. Firstly, the features of vibration signals collected from multiple sensors are extracted in the time, frequency, and time-frequency domains. Subsequently, dimensionality reduction optimization is conducted on these multi-domain features to eliminate useless information features. The temporal convolutional network (TCN) model is constructed to capture the critical information reflecting the fault characteristics of rotating machinery through the attention mechanism, and the dependencies of the whole training process are captured by the BiGRU network. Finally, precise prediction of the lifespan of rotating machinery is achieved by constructing a health indicator curve (HI). The proposed methods are verified through the life prediction of rolling bearings from the IEEE PHM Challenge 2012 dataset and ball screw pairs from a designed experiment. The experimental results show that the proposed MDFF and DITCN-ABiGRU model achieves a better score and lower error than the convolutional neural network (CNN) and GRU models.

RESUMEN

BACKGROUND: In clinical medicine, fetal heart rate (FHR) monitoring using cardiotocography (CTG) is one of the most commonly used methods for assessing fetal acidosis. However, as the visual interpretation of CTG depends on the subjective judgment of the clinician, this has led to high inter-observer and intra-observer variability, making it necessary to introduce automated diagnostic techniques. METHODS: In this study, we propose a computer-aided diagnostic algorithm (Hybrid-FHR) for fetal acidosis to assist physicians in making objective decisions and taking timely interventions. Hybrid-FHR uses multi-modal features, including one-dimensional FHR signals and three types of expert features designed based on prior knowledge (morphological time domain, frequency domain, and nonlinear). To extract the spatiotemporal feature representation of one-dimensional FHR signals, we designed a multi-scale squeeze and excitation temporal convolutional network (SE-TCN) backbone model based on dilated causal convolution, which can effectively capture the long-term dependence of FHR signals by expanding the receptive field of each layer's convolution kernel while maintaining a relatively small parameter size. In addition, we proposed a cross-modal feature fusion (CMFF) method that uses multi-head attention mechanisms to explore the relationships between different modalities, obtaining more informative feature representations and improving diagnostic accuracy. RESULTS: Our ablation experiments show that the Hybrid-FHR outperforms traditional previous methods, with average accuracy, specificity, sensitivity, precision, and F1 score of 96.8, 97.5, 96, 97.5, and 96.7%, respectively. CONCLUSIONS: Our algorithm enables automated CTG analysis, assisting healthcare professionals in the early identification of fetal acidosis and the prompt implementation of interventions.

Asunto(s)

RESUMEN

Automatic seizure detection plays a key role in assisting clinicians for rapid diagnosis and treatment of epilepsy. In view of the parallelism of temporal convolutional network (TCN) and the capability of bidirectional long short-term memory (BiLSTM) in mining the long-range dependency of multi-channel time-series, we propose an automatic seizure detection method with a novel end-to-end TCN-BiLSTM model in this work. First, raw EEG is filtered with a 0.5-45 Hz band-pass filter, and the filtered data are input into the proposed TCN-BiLSTM network for feature extraction and classification. Post-processing process including moving average filtering, thresholding and collar technique is then employed to further improve the detection performance. The method was evaluated on two EEG database. On the CHB-MIT scalp EEG database, our method achieved a segment-based sensitivity of 94.31%, specificity of 97.13%, and accuracy of 97.09%. Meanwhile, an event-based sensitivity of 96.48% and an average false detection rate (FDR) of 0.38/h were obtained. On the SH-SDU database we collected, the segment-based sensitivity of 94.99%, specificity of 93.25%, and accuracy of 93.27% were achieved. In addition, an event-based sensitivity of 99.35% and a false detection rate of 0.54/h were yielded. The total detection time consumed for 1[Formula: see text]h EEG data was 5.65[Formula: see text]s. These results demonstrate the superiority and promising potential of the proposed method in real-time monitoring of epileptic seizures.

Asunto(s)

RESUMEN

Atrial fibrillation, one of the most common persistent cardiac arrhythmias globally, is known for its rapid and irregular atrial rhythms. This study integrates the temporal convolutional network (TCN) and residual network (ResNet) frameworks to effectively classify atrial fibrillation in single-lead ECGs, thereby enhancing the application of neural networks in this field. Our model demonstrated significant success in detecting atrial fibrillation, with experimental results showing an accuracy rate of 97% and an F1 score of 87%. These figures indicate the model's exceptional performance in identifying both majority and minority classes, reflecting its balanced and accurate classification capability. This research offers new perspectives and tools for diagnosis and treatment in cardiology, grounded in advanced neural network technology.

Asunto(s)

RESUMEN

Wastewater treatment plant (WWTP) operation is usually intricate due to large variations in influent characteristics and nonlinear sewage treatment processes. Effective modeling of WWTP effluent water quality can provide valuable decision-making support to facilitate their operations and management. In this study, we developed a novel hybrid deep learning model by combining the temporal convolutional network (TCN) model with the long short-term memory (LSTM) network model to improve the simulation of hourly total nitrogen (TN) concentration in WWTP effluent. The developed model was tested in a WWTP in Jiangsu Province, China, where the prediction results of the hybrid TCN-LSTM model were compared with those of single deep learning models (TCN and LSTM) and traditional machine learning model (feedforward neural network, FFNN). The hybrid TCN-LSTM model could achieve 33.1 % higher accuracy as compared to the single TCN or LSTM model, and its performance could improve by 63.6 % comparing to the traditional FFNN model. The developed hybrid model also exhibited a higher power prediction of WWTP effluent TN for the next multiple time steps within eight hours, as compared to the standalone TCN, LSTM, and FFNN models. Finally, employing model interpretation approach of Shapley additive explanation to identify the key parameters influencing the behavior of WWTP effluent water quality, it was found that removing variables that did not contribute to the model output could further improve modeling efficiency while optimizing monitoring and management strategies.

Asunto(s)

RESUMEN

Brain-computer interface (BCI) enables the control of external devices using signals from the brain, offering immense potential in assisting individuals with neuromuscular disabilities. Among the different paradigms of BCI systems, the motor imagery (MI) based electroencephalogram (EEG) signal is widely recognized as exceptionally promising. Deep learning (DL) has found extensive applications in the processing of MI signals, wherein convolutional neural networks (CNN) have demonstrated superior performance compared to conventional machine learning (ML) approaches. Nevertheless, challenges related to subject independence and subject dependence persist, while the inherent low signal-to-noise ratio of EEG signals remains a critical aspect that demands attention. Accurately deciphering intentions from EEG signals continues to present a formidable challenge. This paper introduces an advanced end-to-end network that effectively combines the efficient channel attention (ECA) and temporal convolutional network (TCN) components for the classification of motor imagination signals. We incorporated an ECA module prior to feature extraction in order to enhance the extraction of channel-specific features. A compact convolutional network model uses for feature extraction in the middle part. Finally, the time characteristic information is obtained by using TCN. The results show that our network is a lightweight network that is characterized by few parameters and fast speed. Our network achieves an average accuracy of 80.71% on the BCI Competition IV-2a dataset.

Asunto(s)