Different Types Of Events Research Articles

Synchrophasor network-based detection and classification of power system events: A singular value decomposition approach

Timely detection and classification of power system events are essential for situation awareness and reliable electricity grid operation. It is also a crucial step with regard to synchrophasor network data management and archiving. In this paper, an event detection and classification method based on the singular value decomposition (SVD) of synchrophasor data is proposed. The detection algorithm exploits the low-dimensionality characteristics of synchrophasor data and identifies the changes in the dimensionality of a sliding data matrix. The SVD-based method assigns several detection flags indicating events and outliers in voltage magnitude, phase angle and frequency data. The proposed classification algorithm comprises a decision tree employing detection flags and singular values to classify events into several categories, e.g., fault, voltage magnitude and phase angle events, and generation-load mismatch events. Moreover, the proposed algorithm identifies whether events are spatially correlated. Field synchrophasor data collected from a smart grid are used to evaluate the performance of the proposed method. The numerical results show that the proposed method can successfully detect and classify different types of events even in the presence of measurement uncertainty.

Model-based prediction of progression-free survival for combination therapies in oncology.

Progression-free survival (PFS) is an important clinical metric for comparing and evaluating similar treatments for the same disease within oncology. After the completion of a clinical trial, a descriptive analysis of the patients' PFS is often performed post hoc using the Kaplan-Meier estimator. However, to perform predictions, more sophisticated quantitative methods are needed. Tumor growth inhibition models are commonly used to describe and predict the dynamics of preclinical and clinical tumor size data. Moreover, frameworks also exist for describing the probability of different types of events, such as tumor metastasis or patient dropout. Combining these two types of models into a so-called joint model enables model-based prediction of PFS. In this paper, we have constructed a joint model from clinical data comparing the efficacy of FOLFOX against FOLFOX + panitumumab in patients with metastatic colorectal cancer. The nonlinear mixed effects framework was used to quantify interindividual variability (IIV). The model describes tumor size and PFS data well, and showed good predictive capabilities using truncated as well as external data. A machine-learning guided analysis was performed to reduce unexplained IIV by incorporating patient covariates. The model-based approach illustrated in this paper could be useful to help design clinical trials or to determine new promising drug candidates for combination therapy trials.

Optimized Ensemble Approach for Multi-model Event Detection in Big data

Event detection acts an important role among modern society and it is a popular computer process that permits to detect the events automatically. Big data is more useful for the event detection due to large size of data. Multimodal event detection is utilized for the detection of events using heterogeneous types of data. This work aims to perform for classification of diverse events using Optimized Ensemble learning approach. The Multi-modal event data including text, image and audio are sent to the user devices from cloud or server where three models are generated for processing audio, text and image. At first, the text, image and audio data is processed separately. The process of creating a text model includes pre-processing using Imputation of missing values and data normalization. Then the textual feature extraction using integrated N-gram approach. The Generation of text model using Convolutional two directional LSTM (2DCon_LSTM). The steps involved in image model generation are pre-processing using Min-Max Gaussian filtering (MMGF). Image feature extraction using VGG-16 network model and generation of image model using Tweaked auto encoder (TAE) model. The steps involved in audio model generation are pre-processing using Discrete wavelet transform (DWT). Then the audio feature extraction using Hilbert Huang transform (HHT) and Generation of audio model using Attention based convolutional capsule network (Attn_CCNet). The features obtained by the generated models of text, image and audio are fused together by feature ensemble approach. From the fused feature vector, the optimal features are trained through improved battle royal optimization (IBRO) algorithm. A deep learning model called Convolutional duo Gated recurrent unit with auto encoder (C-Duo GRU_AE) is used as a classifier. Finally, different types of events are classified where the global model are then sent to the user devices with high security and offers better decision making process. The proposed methodology achieves better performances are Accuracy (99.93%), F1-score (99.91%), precision (99.93%), Recall (99.93%), processing time (17seconds) and training time (0.05seconds). Performance analysis exceeds several comparable methodologies in precision, recall, accuracy, F1 score, training time, and processing time. This designates that the proposed methodology achieves improved performance than the compared schemes. In addition, the proposed scheme detects the multi-modal events accurately.

We Need Subject Matter Expertise to Choose and Identify Causal Estimands: Comment on “Estimands for Recurrent Event Endpoints in the Presence of a Terminal Event”

We summarize what we consider to be the two main limitations of the “Estimands for Recurrent Event Endpoints in the Presence of a Terminal Event” (Schmidli et al. 2022). First, the authors did not give detailed guidance on how to choose an appropriate estimand in light of subject-matter considerations. Reasoning about the mechanism by which treatment affects different types of events is central when selecting a causal estimand, and such reasoning can be grounded in the interventionist mediation literature. Second, the article also did not discuss the crucial task of identification when the aim is to study a causal question. Thereby, the authors omit important differences in the uncertainty of the assumptions needed to target each estimand by particular statistical methods. These assumptions have crucial implications for the confidence that can be placed in a given effect estimate, and for the planning and collection of relevant variables in the study design.

How Pre-Event Engagement and Event Type Are Associated with Spectators’ Cognitive Processing at Two Swimming Events

Little is known about event experiences of spectators attending non-mainstream sports, let alone different types of events within the same sport. Past research has demonstrated that pre-event engagement influences event experiences. Therefore, the purpose of this study was to examine the ways in which pre-event engagement and event type were associated with spectators’ event experiences at two different types of swimming events. Surveys were collected from 90 spectators attending the 2016 National Swimming Championships (NSC) in Belgium, and from 246 spectators attending the 2016 World Swimming Championships (WSC) in Canada. Pre-event engagement included behavioral measures, as well as cognitive and affective measures. Event experiences were measured with Madrigal’s (2006) FANDIM scale. A series of regression analyses were used to explore if and how spectators’ pre-event engagement and event type were associated with their event experiences, while controlling for socio-demographic variables. Knowledge of and interest in the sport, as well as awareness of opportunities to participate were associated with aesthetics, evaluation, and flow. Pre-event engagement in terms of prior spectating and active participation were not associated with event experiences. The type of event matters and the WSC generated more engaged event experiences (i.e., aesthetics and fantasy). To improve event experiences at non-mainstream events, managers should seek to implement initiatives to increase interest among potential spectators, including those with little or no behavioral engagement with the sport.

Two Types of Ritual Space at the Poverty Point Site 16WC5

AbstractTwo types of features at the Late Archaic period Poverty Point site in Louisiana—large timber post circles and repetitively used activity surfaces later covered by mounding—are examined as spaces that were the products of historical event sequences involving their construction, their entwined but differing ritual uses, and their final deconstruction or removal. This approach is a particularizing, historicizing way of approaching the features of this monumental hunter-gatherer site, an alternative to a more generalizing view in which its totality as a site or “place” is seen as the interrelated product of many historical acts over time or as a mosaic of historically created and used “spaces.” Analysis of these feature types suggests that spaces of ritual activity in the larger site were used differently, involving different types of events and serving different functions in a developing, perhaps shifting, larger ritual landscape.

CNN disruption predictor at JET: Early versus late data fusion approach

This work focuses on the development of a data driven model, based on Convolutional Neural Networks (CNNs), for the real-time detection of disruptive events at JET. The predictor exploits the ability of CNNs in learning relevant spatiotemporal information straight from 1D plasma profiles, avoiding hand-engineered feature extraction procedures. In this paper, the radiation profiles from both the bolometer horizontal and vertical cameras have been considered amongst the predictor inputs, with the aim of discriminating between the core radiation due to impurity accumulations and the outboard radiation phenomena. Moreover, an innovative predictor architecture is proposed, where two separate CNNs are trained to focus on events with different timescales, that is, the destabilization of radiation, electron density and temperature profiles, and the mode-locking and current profile variations. The outputs of the two CNNs are combined with a logic OR function to provide the disruption alarm trigger. The advantages of this data fusion approach impact on the predictor performance, with a very limited number of false alarms (only 1 in the considered test set), and on the model output interpretability as the two different branches are triggered by different types of events.

Clinical study on the classification of renal artery involvement and comparison of renal function and prognosis of Stanford type B aortic dissection after thoracic aortic endovascular repair

Objective: To investigate the different types of renal artery involvement in Stanford type B aortic dissection (TBAD) and the comparison of clinical effecacy after thoracic endovascular aortic repair (TEVAR). Methods: This is a retrospective cohort study included 330 patients with TBAD and renal artery involvement treated with TEVAR from June 2002 to September 2021 in General Hospital of Northern Theater Command of the PLA. According to aortic CTA image, unilateral renal artery involvement conditions were divided into 5 types: the true lumen type (renal artery opening completely from the true lumen), false lumen type (renal artery opening completely from the false lumen), double lumen type (renal artery opening from the true and false double lumen), compression type (renal artery opening connected with the true lumen, but the renal artery opening was extremely squeezed by the inner membrane), open type (renal artery opening with intimal tear). There were seven types of bilateral renal artery involvement: true-true type (true lumen-true lumen type), true and false type (true lumen-false lumen type), true-double type (true lumen-double lumen type), true-opening type (true lumen-opening type), false-false type (false lumen-false lumen type), false-compression type (false lumen-compression type), double-double type (double lumen-double lumen type). The primary observation index of this study was the comparison of postoperative renal function and the incidence of clinical adverse events of different types of renal artery involvement. One-way ANOVA test, Kruskal-Wallis H test and paired sample rank sum test were used to compare postoperative renal function between different types of bilateral renal artery involvement. The Chi-square test or Fisher's exact probability test were used to compare the near and long term adverse events between different types of bilateral renal artery involvement. Kaplan-Meier method was used to compare the all-cause mortality of patients with severe renal functional injury and non-severe renal functional injury before surgery. Results: The average age of the patients included in this study was (53±11) years, including 276 males (83.6%) and 54 females (16.4%). There were statistical difference in the level of serum creatinine (preoperative:H=18.686, P=0.005, postoperative:H=18.101, P=0.006) and cystatin C (preoperative:H=17.566, P=0.007, postoperative:H=10.433, P=0.016), pre-and post-operative, between the seven groups of TBAD patients with different renal artery involvement types (P<0.05), and the false-false type group shown the worst kidney function. However, no statistically significant differences were shown when comparing their pre- and post-operative change values (P>0.05). The 30-day follow-up result showed that there were statistically significant differences in the incidence of postoperative acute kidney injury (χ2=15.623, P=0.007), aorta-related adverse events (χ2=15.523, P=0.010), and intraoperative endoleak (χ2=17.935, P=0.004) among the seven groups, and the false-false group was the highest (2/9, 5/9 and 5/9, respectively). In terms of long-term follow-up results, there were statistically significant differences in all-cause death (χ2=14.772, P=0.011) and non-aortic death (χ2=15.589,P=0.008) among the seven groups. Kaplan-Meier survival analysis showed that patients with worse pre-operative renal function showed higher long-term all cause death (17.7% vs. 4.8%, P=0.009). Conclusions: For TBAD patients with renal artery involvement, there were differences in renal function among different types, and TEVAR showed no significant effect on renal function in TBAD patients. The long-term all cause death was higher in patients with worse renal function pre-operative.

SmartPatch: Verifying the Authenticity of the Trigger-Event in the IoT Platform

Emerging IoT clouds are playing a more important role in modern lives, enabling users/developers to program applications to make better use of smart devices. However, preliminary research has shown IoT cloud vulnerabilities could expose IoT users to security risks. To better understand the problem, we studied the SmartThings cloud, one of the most popular IoT cloud platforms that support user-defined device automation (SmartApps). Specifically, we found new vulnerabilities in SmartThings that allow attackers to fake events to trigger the SmartApps to operate devices (e.g., open a lock). Exploiting such vulnerabilities, we successfully faked 7 different types of events, which impact 138 (out of 187) SmartThings’ official open-sourced SmartApps. To defeat such attacks, we propose an authenticity-verification-based scheme to deny fake events. Moreover, we designed a tool, <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SmartPatch</monospace> , to help users secure their SmartThings systems. In specific, <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SmartPatch</monospace> automatically patches the vulnerable SmartApps and Device Handlers (input) and outputs the flawless programs, which are ready for users to deploy in their SmartThings systems. We have made <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SmartPatch</monospace> publicly available. With the help of <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SmartPatch</monospace> , we patched all the vulnerable SmartThings’ official open-sourced programs (146 SmartApps and 321 Device Handlers). Experiments have shown the compatibility, effectiveness, and efficiency of our proposed approach.

Predictors of Lethality, Remodeling, and Aorta-Related Events in Different Types of Proximal Aortic Dissection Surgery.

A retrospective observational comparison of the results of surgical treatment of 213 patients with the diagnosis of "DeBakey type I aortic dissection" has been carried out. The participants were divided into three groups: group 1 underwent classic aortic arch reconstruction using hemiarch technique or total reconstruction of the aortic arch with a multiple-branch prosthesis (n=121); group 2 was subjected to the hemiarch technique and implantation of bare-metal (uncoated) stents (n=55); in group 3, the "frozen elephant trunk" correction technique was used (n=37). The diagnosis of all patients included into the study was preoperatively confirmed by ultrasound and tomographic examination. Predictors of negative events have been identified by building the models of logistic regressions. The multivariate model of logistic regression has revealed multiplicative significant predictors of lethality: postoperative neurological complications increased the probability of lethality by 3.39 (1.24-9.18) times and presence of a patent false lumen by 4.17 (1.49-13.68) times.Among the predictors of aorta-related events, the most important were connective tissue diseases (the probability increased by 6.68 (2.98-15.62) times), presence of partial thrombosis of the false lumen (the probability of event development increased by 2.39 (1.07-5.44) times), and aortic valve repair (the probability aorta-event occurrence increased by 2.84 (1.13-7.17) times).Hybrid prosthesis implantation appeared to be the most significant predictor of false lumen thrombosis increasing its probability by 4.19 (1.90-9.44) times among aortic repair methods, while a bare-metal stent implantation in contrast reduced the likelihood of false lumen thrombosis by 0.17 (0.03-0.62) times. Eventually, the type of repair had not any significant impact on the aorta-related events and lethality in the long-term period.

A survey on event detection approaches for sensor based IoT

Over the period, two significant advancements i.e. shrinking of the sensor size and improved communication capability, lead the sensor-based internet-of-things (SBIoT) to mature rapidly. SBIoT has various applications to make human life smarter such as health care IoT, smart cities, agriculture, automated industries, security etc. The sensor collects data by monitoring a natural phenomenon from the environment and the internet is used to communicate that to the end-user. Much research has been done to detect events in such environments. The present work surveys diverse methodologies designed to capture the events of interest from collected data in different application domains of SBIoT. This paper revisits the understanding of detecting events of interest using SBIoT framework and discusses the detailed requirements for event detection. Further, different types of events have been identified, discussed and categorization is made. The authors observed the need to put the different categories of event detection in one paper so that researchers can get idea about state of art of the domain. Previously published surveys focused on a specific domain of event detection such as Health IoT, security in SBIoT, agriculture IoT etc. Further such works defined the occurrences in the deployed field as events of interest in very narrow terms. Thus in most of the existing surveys, the event definition is mostly domain-specific and the focus is on the application area rather than techniques of event detection. Thus, there existed a gap for a holistic survey that conceptualizes different application domains into atomic categories and integrates different techniques available for SBIoT using a common thread of event detection. Hence, the event-based monitoring frameworks in SBIoT are segregated and diversified Event detection models are discussed based on the application domains. The paper describes event detection challenges and methodologies in different domains i.e rule-based, probability and statistics based, multimedia-based, intelligent system based and signal processing based methodologies. Further, the comparative analysis of various recently proposed methods has been carried out in terms of their strengths and weaknesses. The present work also discusses the existing research gaps in the domain of event detection and the authors believe that it will be helpful for researchers in academia and industry to develop new methodologies.

The Official Journal of ATTD Advanced Technologies &amp; Treatments for Diabetes Conference 22‐25 February 2023 I Berlin &amp; Online

The Official Journal of ATTD Advanced Technologies &amp; Treatments for Diabetes Conference 22‐25 February 2023 I Berlin &amp; Online

Future Event Prediction Based on Temporal Knowledge Graph Embedding

Accurate prediction of future events brings great benefits and reduces losses for society in many domains, such as civil unrest, pandemics, and crimes. Knowledge graph is a general language for describing and modeling complex systems. Different types of events continually occur, which are often related to historical and concurrent events. In this paper, we formalize the future event prediction as a temporal knowledge graph reasoning problem. Most existing studies either conduct reasoning on static knowledge graphs or assume knowledges graphs of all timestamps are available during the training process. As a result, they cannot effectively reason over temporal knowledge graphs and predict events happening in the future. To address this problem, some recent works learn to infer future events based on historical event-based temporal knowledge graphs. However, these methods do not comprehensively consider the latent patterns and influences behind historical events and concurrent events simultaneously. This paper proposes a new graph representation learning model, namely Recurrent Event Graph ATtention Network (RE-GAT), based on a novel historical and concurrent events attention-aware mechanism by modeling the event knowledge graph sequence recurrently. More specifically, our RE-GAT uses an attention-based historical events embedding module to encode past events, and employs an attention-based concurrent events embedding module to model the associations of events at the same timestamp. A translation-based decoder module and a learning objective are developed to optimize the embeddings of entities and relations. We evaluate our proposed method on four benchmark datasets. Extensive experimental results demonstrate the superiority of our RE-GAT model comparing to various baselines, which proves that our method can more accurately predict what events are going to happen.

ІСТОРІЯ ФОРМУВАННЯ АРХІТЕКТУРИ КЛУБНИХ БУДІВЕЛЬ ЛЬВОВА

The article shows the history of the formation of the architecture of Lviv clubs from the beginning of their formation to the present day. Today, there is a trend where iconic historical objects, which often have lost their original function, attract the interest of potential investors. In order not to lose valuable objects of the city’s cultural heritage, a policy of integrating the monument into a new functional context is being pursued. Cultural and educational functions prevail among the most popular solutions: media libraries, co-working spaces, social hubs, art and cultural centers. Their main goal is to revitalize cultural communication and public life in the city. Thanks to modern materials and technologies, architects and designers have the opportunity to create flexible, universal spaces that allows quickly respond to changes in user demand and hold a wide range of events. Innovative approaches in the construction process make it possible to preserve iconic objects and bring the architectural environment of the city into European and world contexts. Modern trends in the development of society indicate that there is an acute need for buildings and spaces for the cultural activities of numerous societies and organizations. Recently, as a result of the diversity of urban life, new forms of public communication have appeared, and therefore, new types of buildings in which such communication takes place. The social conditions of each socio-economic formation leave their mark on the type of building, give it a new social meaning and structure. Therefore, in order to predict the development of multifunctional cultural and educational objects today, to determine their optimal functional structure and volume-spatial solution of the auditorium, capable to exist for different types of events, it is necessary to know and use the architectural experience of past times, when cultural and educational buildings and clubs are becoming the most popular, both on our territory and in the world.

Climate Change is an Emerging Threat to Perinatal Mental Health.

In this discussion, we build the case for why climate change is an emerging threat to perinatal mental health. A search of current literature on perinatal and maternal mental health and extreme weather events was conducted in PubMed/MEDLINE and Web of Science databases. Only articles focusing on maternal mental health were included in this narrative review. The perinatal period represents a potentially challenging timeframe for women for several reasons. Necessary role adjustments (reprioritization), changes in one's ability to access pre-birth levels (and types) of social support, fluctuating hormones, changes in body shape, and possible complications during pregnancy, childbirth, or postpartum are just a few of the factors that can impact perinatal mental health. Trauma is also a risk factor for negative mood symptoms and can be experienced as the result of many different types of events, including exposure to extreme weather/natural disasters. While the concepts of "eco-anxiety," "climate despair," and "climate anxiety" have garnered attention in the mainstream media, there is little to no discussion of how the climate crisis impacts maternal mental health. This is an important omission as the mother's mental health impacts the family unit as a whole.

Analyzing Survival Data with Competing Risks Based on R-packages

When it comes to data in survival analysis, it is easy to think of one censored or observed survival time of an individual. In the longitudinal study such as a clinical trial, an individual may be simultaneously exposed to different types of events. It is easy to think of competing risks data as an extension of univariate survival data, but since competing risks events are dependent on each other, if the analysis method of univariate survival data is applied, inferences with bias can be made. In this paper, we briefly introduce the statistical methods developed to analyze competing risks data and apply them to the real examples using the built-in functions in the R package.

Time–Frequency Analysis of Experimental Measurements for the Determination of EMI Noise Generators in Power Converters

In the context of recent decades in which there has been great development in power electronics systems with increasingly better operating characteristics, the study of the factors that affect the behaviour of these systems in terms of electromagnetic interference (EMI) is mandatory. Within this general perspective, it is essential to know the time and frequency characteristics of the switching signals, as they are the main sources of EMI noise. This work analyses the suitability of different spectral analysis techniques, specifically short-time Fourier transform (STFT) and continuous wavelet transform (CWT), to extract the frequency characteristics of the switching signals of a converter. Thus, a test bench based on a half-bridge was designed as a preliminary step in the development of a model of noise generators. After the analysis of the main parameters of these techniques, a comparison of the results obtained was carried out. It was concluded that both techniques are considered valid and complementary; not only that, but they overcome some limitations of the fast Fourier transform (FFT) as they offer the possibility of determining which are the frequency components associated with different types of events occurring at different times.

Statistical Study and Live Catalog of Multispacecraft 3He-rich Time Periods over Solar Cycles 23, 24, and 25

Using ion measurements from Ultra-Low-Energy Isotope Spectrometer observations on board Advanced Composition Explorer and Solar Isotope Spectrometer observations on board the Solar Terrestrial Observatory (STEREO)-A and STEREO-B spacecraft, we have identified 854 3He-rich time periods between 1997 September and 2021 March. We include all event types with observed 3He enhancements such as corotating interaction regions, gradual solar energetic particle (SEP) events, interplanetary shocks, and impulsive SEP events. We employ two different mass separation techniques to obtain 3He, 4He, Fe, and O fluences for each event, and we determine the 3He/4He and Fe/O abundance ratios between 0.32–0.45 and 0.64–1.28 MeV nucleon−1. We find a clear correlation in the 3He/4He and Fe/O abundance ratios between both energy ranges. We find two distinct trends in the 3He/4He versus Fe/O relation. For low 3He/4He values, there is a positive linear correlation between 3He/4He and Fe/O. However, at 3He/4He ∼ 0.3, Fe/O appears to reach a limit and the correlation weakens significantly. We provide a live catalog of 3He-rich time periods that includes the robust determination of the onset and end times of the 3He enhancements in SEP-associated periods for different types of events observed by multiple spacecraft. This catalog is available for public use. New releases will follow after major additions such as adding new periods from new missions (e.g., Parker Solar Probe and Solar Orbiter), identifying event types (impulsive SEP events, etc.), or adding new parameters such as remote observations detailing characteristics of the events’ active regions.

The maraca plot: A novel visualization of hierarchical composite endpoints.

Hierarchical composite endpoints are complex endpoints combining outcomes of different types and different clinical importance into an ordinal outcome that prioritizes the clinically most important (e.g. most severe) event of a patient. Hierarchical composite endpoint can be analysed with the win odds, an adaptation of win ratio to include ties. One of the difficulties in interpreting hierarchical composite endpoint is the lack of proper tools for visualizing the treatment effect captured by hierarchical composite endpoint, given the complex nature of the endpoint which combines events of different types. Hierarchical composite endpoints usually combine time-to-event outcomes and continuous outcomes into a composite; hence, it is important to capture not only the shift from more severe categories to less severe categories in the active group in comparison to the control group (as in any ordinal endpoint), but also changes occurring within each category. We introduce the novel maraca plot which combines violin plots (with nested box plots) to visualize the density of the distribution of the continuous outcome and Kaplan-Meier plots for time-to-event outcomes into a comprehensive visualization. The novel maraca plot is suggested for visualization of hierarchical composite endpoints consisting of several time-to-event outcomes and a continuous outcome. It has a very simple structure and therefore easily communicates both the overall treatment effect and the effect on individual components.

CATNet: Cross-event attention-based time-aware network for medical event prediction

Medical event prediction (MEP) is a fundamental task in the healthcare domain, which needs to predict medical events, including medications, diagnosis codes, laboratory tests, procedures, outcomes, and so on, according to historical medical records of patients. Many researchers have tried to build MEP models to overcome the challenges caused by the heterogeneous and irregular temporal characteristics of EHR data. However, most of them consider the heterogenous and temporal medical events separately and ignore the correlations among different types of medical events, especially relations between heterogeneous historical medical events and target medical events. In this paper, we propose a novel neural network based on attention mechanism called Cross-event Attention-based Time-aware Network (CATNet) for MEP. It is a time-aware, event-aware and task-adaptive method with the following advantages: 1) modeling heterogeneous information and temporal information in a unified way and considering irregular temporal characteristics locally and globally respectively, 2) taking full advantage of correlations among different types of events via cross-event attention. Experiments on two public datasets (MIMIC-III and eICU) show CATNet outperforms other state-of-the-art methods on various MEP tasks. The source code of CATNet is released at https://github.com/sherry6247/CATNet.git.