Time Series Domain Research Articles

Quantifying quality of class-conditional generative models in time series domain

Despite recent breakthroughs in the domain of implicit generative models, the task of evaluating these models remains a challenging task. With no single metric to assess overall performance, various existing metrics only offer partial information. This issue is further compounded for unintuitive data types such as time series, where manual inspection is infeasible. This deficiency hinders the confident application of modern implicit generative models on time series data. To alleviate this problem, we propose two new metrics, the InceptionTime Score (ITS) and the Fréchet InceptionTime Distance (FITD), to assess the quality of class-conditional generative models on time series data. We conduct extensive experiments on 80 different datasets to study the discriminative capabilities of proposed metrics alongside two existing evaluation metrics: Train on Synthetic Test on Real (TSTR) and Train on Real Test on Synthetic (TRTS). Our evaluations reveal that the proposed assessment evaluation metrics, i.e., ITS and FITD in combination with TSTR, can accurately assess class-conditional generative model performance and detect common issues in implicit generative models. Our findings suggest that the proposed evaluation framework can be a valuable tool for confidently applying modern implicit generative models in time series analysis.

Smart fault detection of HTS coils using artificial intelligence techniques for large-scale superconducting electric transport applications

Superconducting coil is an essential and critical component in any superconducting apparatus used in large-scale power applications such as in superconducting machines of propulsion systems, in fault current limiters of the distribution system for future cryo-electric aircraft, or in windings of superconducting transformers for power grid applications. The superconducting coils in winding of large-scale power devices operate in kind of harsh environments from both temperature—considering liquid hydrogen or gaseous helium as coolant—(thermal stress) and electro-magneto-mechanical stress, point of views. Reliable operation of the coils in winding is of vital importance for the reliability of the superconducting device and the safety of the application that the device is used for. If the superconducting coil confronts a fault or an abnormal condition in the laboratory-level operation, it is straightforward to test the coil by measuring its critical current, AC loss, etc, to find whether it is damaged or not. However, there would be an urgent need to have faster and more intelligent fault detection and condition monitoring approaches with the possibility to become fully autonomous and real-time, in large-scale power applications, especially in sensitive applications such as in future cryo-electric aircraft, or in the fusion industry. To reach such intelligent fault-finding approaches, artificial intelligence-based techniques have been foreseen to be a promising solution. In this paper, we have developed an intelligent fault detection technique for finding a faulty superconducting coil, named the frequency-temporal classification method. This method has two main steps: first, this paper utilizes the discrete Fourier transform and independent component analysis to convert measurement signals of the healthy and faulty coils from (1) the time-series domain to the frequency domain; and (2) into time-series source signals. Second, this paper trains the support-vector machine using the derived frequency components. This trained model is then used for making fault detection for other superconducting coils. The developed technique can classify a fault with 99.2% accuracy.

The Wavelet Multi-Scale Analysis of Exchange Rate Exposure: An Application to Malaysian Consumer Products and Services Sector

Objective – Past efforts in assessing foreign exchange rate exposure assume homogeneity in the level of exposure across time horizons which seems to be impractical due to the dynamic nature of comovement between firm value and exchange rate. Methodology – Given this, the study aimed to investigate the multi-scale relationships between changes in exchange rates and firm values of 56 multinational corporations in the consumer products and services sector from January 2000 to December 2020. Findings – The novelty of the study lies upon the application of Maximal Overlap Discrete Wavelet Transformation (MODWT) method which allows decomposition of a single time series domain into different time domains. The fragmentation into multiple time domains allows the measurement of scale-dependent foreign exchange exposure. As a result, the study discovered a non-monotonic trend for wavelet scale j of currency exposure across time scales. In particular, there was a gradual increase in the magnitude of beta exchange exposure and the proportion of exposed firms from low to high time scales. This finding suggested that firm value is more sensitive to changes in the exchange rate within a widened time domain. Novelty – The study demonstrated on how the wavelet technique can be used to measure foreign exchange risk and aided firm managers and market participants in managing foreign exchange risk for a specific time interval. Type of Paper: Empirical JEL Classification: E42, F31, F39 Keywords: Currency exposure; multi-horizon exchange rate; Malaysian Sectors; Wavelet Analysis; MODWT. Reference to this paper should be made as follows: Wahab, H.A; Nasir, N.A.M. (2023). The Wavelet Multi-scale Analysis of Exchange Rate Exposure: An Application to Malaysian Consumer Products and Services Sector, J. Fin. Bank. Review, 8(1), 58 – 72. https://doi.org/10.35609/jfbr.2023.8.1(3)

Dynamic Time Warping based Adversarial Framework for Time-Series Domain.

Despite the rapid progress on research in adversarial robustness of deep neural networks (DNNs), there is little principled work for the time-series domain. Since time-series data arises in diverse applications including mobile health, finance, and smart grid, it is important to verify and improve the robustness of DNNs for the time-series domain. In this paper, we propose a novel framework for the time-series domain referred as Dynamic Time Warping for Adversarial Robustness (DTW-AR) using the dynamic time warping measure. Theoretical and empirical evidence is provided to demonstrate the effectiveness of DTW over the standard Euclidean distance metric employed in prior methods for the image domain. We develop a principled algorithm justified by theoretical analysis to efficiently create diverse adversarial examples using random alignment paths. Experiments on diverse real-world benchmarks show the effectiveness of DTW-AR to fool DNNs for time-series data and to improve their robustness using adversarial training.

What about the Latent Space? The Need for Latent Feature Saliency Detection in Deep Time Series Classification

Saliency methods are designed to provide explainability for deep image processing models by assigning feature-wise importance scores and thus detecting informative regions in the input images. Recently, these methods have been widely adapted to the time series domain, aiming to identify important temporal regions in a time series. This paper extends our former work on identifying the systematic failure of such methods in the time series domain to produce relevant results when informative patterns are based on underlying latent information rather than temporal regions. First, we both visually and quantitatively assess the quality of explanations provided by multiple state-of-the-art saliency methods, including Integrated Gradients, Deep-Lift, Kernel SHAP, and Lime using univariate simulated time series data with temporal or latent patterns. In addition, to emphasize the severity of the latent feature saliency detection problem, we also run experiments on a real-world predictive maintenance dataset with known latent patterns. We identify Integrated Gradients, Deep-Lift, and the input-cell attention mechanism as potential candidates for refinement to yield latent saliency scores. Finally, we provide recommendations on using saliency methods for time series classification and suggest a guideline for developing latent saliency methods for time series.

ADATIME: A Benchmarking Suite for Domain Adaptation on Time Series Data

Unsupervised domain adaptation methods aim at generalizing well on unlabeled test data that may have a different (shifted) distribution from the training data. Such methods are typically developed on image data, and their application to time series data is less explored. Existing works on time series domain adaptation suffer from inconsistencies in evaluation schemes, datasets, and backbone neural network architectures. Moreover, labeled target data are often used for model selection, which violates the fundamental assumption of unsupervised domain adaptation. To address these issues, we develop a benchmarking evaluation suite ( AdaTime ) to systematically and fairly evaluate different domain adaptation methods on time series data. Specifically, we standardize the backbone neural network architectures and benchmarking datasets, while also exploring more realistic model selection approaches that can work with no labeled data or just a few labeled samples. Our evaluation includes adapting state-of-the-art visual domain adaptation methods to time series data as well as the recent methods specifically developed for time series data. We conduct extensive experiments to evaluate 11 state-of-the-art methods on five representative datasets spanning 50 cross-domain scenarios. Our results suggest that with careful selection of hyper-parameters, visual domain adaptation methods are competitive with methods proposed for time series domain adaptation. In addition, we find that hyper-parameters could be selected based on realistic model selection approaches. Our work unveils practical insights for applying domain adaptation methods on time series data and builds a solid foundation for future works in the field. The code is available at github.com/emadeldeen24/AdaTime .

Spare Parts Demand Forecasting Method Based on Intermittent Feature Adaptation.

The demand for complex equipment aftermarket parts is mostly sporadic, showing typical intermittent characteristics as a whole, resulting in the evolution law of a single demand series having insufficient information, which restricts the prediction effect of existing methods. To solve this problem, this paper proposes a prediction method of intermittent feature adaptation from the perspective of transfer learning. Firstly, to extract the intermittent features of the demand series, an intermittent time series domain partitioning algorithm is proposed by mining the demand occurrence time and demand interval information in the series, then constructing the metrics, and using a hierarchical clustering algorithm to divide all the series into different sub-source domains. Secondly, the intermittent and temporal characteristics of the sequence are combined to construct a weight vector, and the learning of common information between domains is accomplished by weighting the distance of the output features of each cycle between domains. Finally, experiments are conducted on the actual after-sales datasets of two complex equipment manufacturing enterprises. Compared with various prediction methods, the method in this paper can effectively predict future demand trends, and the prediction's stability and accuracy are significantly improved.

Prediction of Water Quantity and Quality Variables Using Deep Learning Algorithms Adopted in Time Series Domain

We developed sequential deep learning models to predict time-series water quantity and quality data. The data set consisting of 10 input variables was prepared from a watershed model Hydrological Simulation Program - FORTRAN (HSPF) fine-tuned to existing conditions in the Nam River Basin on a daily time step for the simulation period 2019-2021. The whole data set was partitioned into training and test sets in the ratio 7 to 3. The predictive accuracy of four deep learning models, created differently in terms of the types of algorithms as well as the number of layers, was tested with respect to mean squared error (MSE). We found that changes in input time steps from 1 to 2 days led to a sharp reduction in prediction errors of all applied models for two target variables during the training and test phases, except for a few cases. In addition, at least 3 important variables were enough to maintain the predictive accuracy of the original deep learning model with 10 variables. The performance of the deep learning model was sensitive to output time steps rather than input time steps. In all test conditions, the MSE values were extremely low, reaching as high as 0.0056. Therefore, sequential deep learning models, regardless of their types and architectures, are most suitable for predictive modeling of time series data compiled on a daily basis or less such as remote sensing data in hydrology and agriculture.

Impact of Design Parameters on the Dynamic Response and Fatigue of Offshore Jacket Foundations

The lifetime of offshore foundations is governed by a combination of harsh environmental conditions and complex service loads. The fatigue limit state (FLS) analysis needs to be performed in the time domain to capture the complex phenomenon. This study aims to investigate different parameters and design modifications that can impact the design life of an offshore jacket foundation. An OC4 jacket foundation is designed in industrial software from DNV and reduced to a super-element model. The super-element model is connected to an NREL 5-MW wind turbine designed in Bladed. The time-series loads are used to compute the fatigue damages faced by the foundation during the service life. The impact of soil non-linearity, marine growth, scour size, the mass of the transition piece, and the grouted connection’s design on the dynamic response and fatigue damages are compared. A 30% increase in life was observed by replacing the concrete transition piece with a lightweight steel configuration. The fatigue damages were considerably greater for the inclined pile in the leg grouted connection than for the leg in the pile concept. The study provides a different perspective by analysing the effect of design parameters and design changes in the complex and computationally expensive time-series domain.

Training Robust Deep Models for Time-Series Domain: Novel Algorithms and Theoretical Analysis

Despite the success of deep neural networks (DNNs) for real-world applications over time-series data such as mobile health, little is known about how to train robust DNNs for time-series domain due to its unique characteristics compared to images and text data. In this paper, we fill this gap by proposing a novel algorithmic framework referred as RObust Training for Time-Series (RO-TS) to create robust deep models for time-series classification tasks. Specifically, we formulate a min-max optimization problem over the model parameters by explicitly reasoning about the robustness criteria in terms of additive perturbations to time-series inputs measured by the global alignment kernel (GAK) based distance. We also show the generality and advantages of our formulation using the summation structure over time-series alignments by relating both GAK and dynamic time warping (DTW). This problem is an instance of a family of compositional min-max optimization problems, which are challenging and open with unclear theoretical guarantee. We propose a principled stochastic compositional alternating gradient descent ascent (SCAGDA) algorithm for this family of optimization problems. Unlike traditional methods for time-series that require approximate computation of distance measures, SCAGDA approximates the GAK based distance on-the-fly using a moving average approach. We theoretically analyze the convergence rate of SCAGDA and provide strong theoretical support for the estimation of GAK based distance. Our experiments on real-world benchmarks demonstrate that RO-TS creates more robust deep models when compared to adversarial training using prior methods that rely on data augmentation or new definitions of loss functions. We also demonstrate the importance of GAK for time-series data over the Euclidean distance.

Adversarial Framework with Certified Robustness for Time-Series Domain via Statistical Features

Time-series data arises in many real-world applications (e.g., mobile health) and deep neural networks (DNNs) have shown great success in solving them. Despite their success, little is known about their robustness to adversarial attacks. In this paper, we propose a novel adversarial framework referred to as Time-Series Attacks via STATistical Features (TSA-STAT). To address the unique challenges of time-series domain, TSA-STAT employs constraints on statistical features of the time-series data to construct adversarial examples. Optimized polynomial transformations are used to create attacks that are more effective (in terms of successfully fooling DNNs) than those based on additive perturbations. We also provide certified bounds on the norm of the statistical features for constructing adversarial examples. Our experiments on diverse real-world benchmark datasets show the effectiveness of TSA-STAT in fooling DNNs for time-series domain and in improving their robustness.

Decadal trend analysis of rainfall patterns of past 115 years & its impact on Sikkim, India

Fluctuations or variations in climatic parameters is a recurring phenomenon in India. The effects of climate variability exacerbate existing social and economic encounters across the area. Over the last 115 years, the rainfall patterns in India have changed significantly. The most significant change is the shifting of the monsoon season altogether. This study aims to discuss the significant changes of rainfall pattern and their influence in Sikkim state, India. Daily rainfall data of 115 years from 1901 to 2015 has been processed in the study to find out the monthly variability of rainfall for which Mann-Kendall (MK) Test together with the Sen's Slope Estimator for the determination of trend and slope magnitude. Monthly precipitation trend has been identified here to achieve the objective, which has been shown with 115 years of data. The Mann–Kendall test and Sen's estimation tests indicated significant changing patterns in the frequency and intensity of the rainfall across the Sikkim in an intra-decadal time series domain. It has been noted in this study that the Sikkim region observed drastic rainfall pattern changes after 2001, because of which there was an increase in disasters like landslides and floods., This study also used inverse distance weighted (IDW) interpolation method and reclassify tool to produce the maps for identifying the pattern and region that caused more rainfall. NDVI (Vegetation Index) has also shown the drastic changes in south Sikkim of vegetation due to decreased rainfall, urbanisation. NDVI values gradually risefrom spring to rainy season and fall steadily from autumn to winter.

An efficient human activity recognition model based on deep learning approaches

Human Activity Recognition (HAR) has gained traction in recent years in diverse areas such as observation, entertainment, teaching and healthcare, using wearable and smartphone sensors. Such environments and systems necessitate and subsume activity recognition, aimed at recognizing the actions, characteristics, and goals of one or more individuals from a temporal series of observations streamed from one or more sensors. Different developed models for HAR have been explained in literature. Deep learning systems and algorithms were shown to perform highly in HAR in recent years, but these algorithms need lots of computerization to be deployed efficiently in applications. This paper presents a HAR lightweight, low computing capacity, deep learning model, which is ideal for use in real-time applications. The generic HAR framework for smartphone sensor data is proposed, based on Long Short-Term Memory (LSTM) networks for time-series domains and standard Convolutional Neural Network (CNN) used for classification. The findings demonstrate that many of the deployed deep learning and machine learning techniques are surpassed by the proposed model. TRANSLATE with x English Arabic Hebrew Polish Bulgarian Hindi Portuguese Catalan Hmong Daw Romanian Chinese Simplified Hungarian Russian Chinese Traditional Indonesian Slovak Czech Italian Slovenian Danish Japanese Spanish Dutch Klingon Swedish English Korean Thai Estonian Latvian Turkish Finnish Lithuanian Ukrainian French Malay Urdu German Maltese Vietnamese Greek Norwegian Welsh Haitian Creole Persian // TRANSLATE with COPY THE URL BELOW Back EMBED THE SNIPPET BELOW IN YOUR SITE Enable collaborative features and customize widget: Bing Webmaster Portal Back // ORIGINAL: "; langMenu.appendChild(origLangDiv); LanguageMenu.Init('LanguageMenu', LanguageMenu_keys, LanguageMenu_values, LanguageMenu_callback, LanguageMenu_popupid); window["LanguageMenu"] = LanguageMenu; clearInterval(intervalId); } }, 1); // ]]>

Economic Management Data Envelopes Based on the Clustering of Incomplete Data

In this paper, the economic management data envelope is analyzed by an algorithm for clustering incomplete data, a local search method based on reference vectors is designed in the algorithm to improve the accuracy of the algorithm, and a final solution selection method based on integrated clustering is proposed to obtain the final clustering results from the last generation of the solution set. The proposed algorithm and various aspects of it are tested in comparison using benchmark datasets and other comparison algorithms. A time-series domain partitioning method based on fuzzy mean clustering and information granulation is proposed, and a time series prediction method is proposed based on the domain partitioning results. Firstly, the fuzzy mean clustering method is applied to initially divide the theoretical domain of the time series, and then, the optimization algorithm of the theoretical domain division based on information granulation is proposed. It combines the clustering algorithm and the information granulation method to divide the theoretical domain and improves the accuracy and interpretability of sample data division. This article builds an overview of data warehouse, data integration, and rule engine. It introduces the business data integration of the economic management information system data warehouse and the data warehouse model design, taking tax as an example. The fuzzy prediction method of time series is given for the results of the theoretical domain division after the granulation of time-series information, which transforms the precise time-series data into a time series composed of semantic values conforming to human cognitive forms. It describes the dynamic evolution process of time series by constructing the fuzzy logical relations to these semantic values to obtain their fuzzy change rules and make predictions, which improves the comprehensibility of prediction results. Finally, the prediction experiments are conducted on the weighted stock price index dataset, and the experimental results show that applying the proposed time-series information granulation method for time series prediction can improve the accuracy of the prediction results.

Modeling the Context of the Problem Domain of Time Series with Type-2 Fuzzy Sets

Data analysis in the context of the features of the problem domain and the dynamics of processes are significant in various industries. Uncertainty modeling based on fuzzy logic allows building approximators for solving a large class of problems. In some cases, type-2 fuzzy sets in the model are used. The article describes constructing fuzzy time series models of the analyzed processes within the context of the problem domain. An algorithm for fuzzy modeling of the time series was developed. A new time series forecasting scheme is proposed. An illustrative example of the time series modeling is presented. The benefits of contextual modeling are demonstrated.

Fault diagnosis of rolling bearings using singular spectrum analysis and artificial neural networks

Singular spectrum analysis (SSA) has been employed effectively for analyzing in the time-frequency domain of time series. It can collaborate with data-driven models (DDMs) such as Artificial Neural Networks (ANN) to set up a powerful tool for mechanical fault diagnosis (MFD). However, to take advantage of SSA more effectively for MFD, quantifying the optimal component threshold in SSA should be addressed. Also, to exploit the managed mechanical system adaptively, the variation tendency of its physical parameters needs to be caught online. Here, we present a bearing fault diagnosis method (BFDM) based on ANN and SSA that targets these aspects. First, a multi-feature is built from pure mechanical properties distilled from the vibration signal of the system. Relied on SSA, the measured acceleration signal is analyzed to cancel the high-frequency noise. The remaining components take part in building a multi-feature to establish a database for training the ANN. Optimizing the number of the kept components is then carried out to obtain a dataset called Tr_Da. Based on Tr_Da, we receive the optimal ANN (OANN). In the next period, at each checking time, another database called Test_Da is set up online following the same way of building the Tr_Da. The compared result between the encoded output and the output of the OANN corresponding to the input to be Test_Da provides the bearing(s) health information. An experimental apparatus is built to evaluate the BFDM. The obtained results reflect the positive effects of the method.

Few-Shot One-Class Classification via Meta-Learning

Although few-shot learning and one-class classification (OCC), i.e., learning a binary classifier with data from only one class, have been separately well studied, their intersection remains rather unexplored. Our work addresses the few-shot OCC problem and presents a method to modify the episodic data sampling strategy of the model-agnostic meta-learning (MAML) algorithm to learn a model initialization particularly suited for learning few-shot OCC tasks. This is done by explicitly optimizing for an initialization which only requires few gradient steps with one-class minibatches to yield a performance increase on class-balanced test data. We provide a theoretical analysis that explains why our approach works in the few-shot OCC scenario, while other meta-learning algorithms fail, including the unmodified MAML. Our experiments on eight datasets from the image and time-series domains show that our method leads to better results than classical OCC and few-shot classification approaches, and demonstrate the ability to learn unseen tasks from only few normal class samples. Moreover, we successfully train anomaly detectors for a real-world application on sensor readings recorded during industrial manufacturing of workpieces with a CNC milling machine, by using few normal examples. Finally, we empirically demonstrate that the proposed data sampling technique increases the performance of more recent meta-learning algorithms in few-shot OCC and yields state-of-the-art results in this problem setting.

Causal Mechanism Transfer Network for Time Series Domain Adaptation in Mechanical Systems

Data-driven models are becoming essential parts in modern mechanical systems, commonly used to capture the behavior of various equipment and varying environmental characteristics. Despite the advantages of these data-driven models on excellent adaptivity to high dynamics and aging equipment, they are usually hungry for massive labels, mostly contributed by human engineers at a high cost. Fortunately, domain adaptation enhances the model generalization by utilizing the labeled source data and the unlabeled target data. However, the mainstream domain adaptation methods cannot achieve ideal performance on time series data, since they assume that the conditional distributions are equal. This assumption works well in the static data but is inapplicable for the time series data. Even the first-order Markov dependence assumption requires the dependence between any two consecutive time steps. In this article, we assume that the causal mechanism is invariant and present our Causal Mechanism Transfer Network (CMTN) for time series domain adaptation. By capturing causal mechanisms of time series data, CMTN allows the data-driven models to exploit existing data and labels from similar systems, such that the resulting model on a new system is highly reliable even with limited data. We report our empirical results and lessons learned from two real-world case studies, on chiller plant energy optimization and boiler fault detection, which outperform the existing state-of-the-art method.

LSTM Networks Using Smartphone Data for Sensor-Based Human Activity Recognition in Smart Homes.

Human Activity Recognition (HAR) employing inertial motion data has gained considerable momentum in recent years, both in research and industrial applications. From the abstract perspective, this has been driven by an acceleration in the building of intelligent and smart environments and systems that cover all aspects of human life including healthcare, sports, manufacturing, commerce, etc. Such environments and systems necessitate and subsume activity recognition, aimed at recognizing the actions, characteristics, and goals of one or more individuals from a temporal series of observations streamed from one or more sensors. Due to the reliance of conventional Machine Learning (ML) techniques on handcrafted features in the extraction process, current research suggests that deep-learning approaches are more applicable to automated feature extraction from raw sensor data. In this work, the generic HAR framework for smartphone sensor data is proposed, based on Long Short-Term Memory (LSTM) networks for time-series domains. Four baseline LSTM networks are comparatively studied to analyze the impact of using different kinds of smartphone sensor data. In addition, a hybrid LSTM network called 4-layer CNN-LSTM is proposed to improve recognition performance. The HAR method is evaluated on a public smartphone-based dataset of UCI-HAR through various combinations of sample generation processes (OW and NOW) and validation protocols (10-fold and LOSO cross validation). Moreover, Bayesian optimization techniques are used in this study since they are advantageous for tuning the hyperparameters of each LSTM network. The experimental results indicate that the proposed 4-layer CNN-LSTM network performs well in activity recognition, enhancing the average accuracy by up to 2.24% compared to prior state-of-the-art approaches.

Time-Frequency Representation of Motor Evoked Potentials in Brain Tumor Patients.

Background: The integrity of the motor system can be examined by applying navigated transcranial magnetic stimulation (nTMS) to the cortex. The corresponding motor-evoked potentials (MEPs) in the target muscles are mirroring the status of the human motor system, far beyond corticospinal integrity. Commonly used time domain features of MEPs (e.g., peak-to-peak amplitudes and onset latencies) exert a high inter-subject and intra-subject variability. Frequency domain analysis might help to resolve or quantify disease-related MEP changes, e.g., in brain tumor patients. The aim of the present study was to describe the time-frequency representation of MEPs in brain tumor patients, its relation to clinical and imaging findings, and the differences to healthy subject.Methods: This prospective study compared 12 healthy subjects with 12 consecutive brain tumor patients (with and without a paresis) applying nTMS mapping. Resulting MEPs were evaluated in the time series domain (i.e., amplitudes and latencies). After transformation into the frequency domain using a Morlet wavelet approach, event-related spectral perturbation (ERSP), and inter-trial coherence (ITC) were calculated and compared to diffusion tensor imaging (DTI) results.Results: There were no significant differences in the time series characteristics between groups. MEPs were projecting to a frequency band between 30 and 300 Hz with a local maximum around 100 Hz for both healthy subjects and patients. However, there was ERSP reduction for higher frequencies (>100 Hz) in patients in contrast to healthy subjects. This deceleration was mirrored in an increase of the inter-peak MEP latencies. Patients with a paresis showed an additional disturbance in ITC in these frequencies. There was no correlation between the CST integrity (as measured by DTI) and the MEP parameters.Conclusion: Time-frequency analysis may provide additional information above and beyond classical MEP time domain features and the status of the corticospinal system in brain tumor patients. This first evaluation indicates that brain tumors might affect cortical physiology and the responsiveness of the cortex to TMS resulting in a temporal dispersion of the corticospinal transmission.