Heterogeneous Series Research Articles

Spatial–Temporal-Correlation-Constrained Dynamic Graph Convolutional Network for Traffic Flow Forecasting

Accurate traffic flow prediction in road networks is essential for intelligent transportation systems (ITS). Since traffic data are collected from the road network with spatial topological and time series sequences, the traffic flow prediction is regarded as a spatial–temporal prediction task. With the powerful ability to model the non-Euclidean data, the graph convolutional network (GCN)-based models have become the mainstream framework for traffic forecasting. However, existing GCN-based models either use the manually predefined graph structure to capture the spatial features, ignoring the heterogeneity of road networks, or simply perform 1-D convolution with fixed kernel to capture the temporal dependencies of traffic data, resulting in insufficient long-term temporal feature extraction. To solve those issues, a spatial–temporal correlation constrained dynamic graph convolutional network (STC-DGCN) is proposed for traffic flow forecasting. In STC-DGCN, a spatial–temporal embedding encoder module (STEM) is first constructed to encode the dynamic spatial relationships for road networks at different time steps. Then, a temporal feature encoder module with heterogeneous time series correlation modeling (TFE-HCM) and a spatial feature encoder module with dynamic multi-graph modeling (SFE-DCM) are designed to generate dynamic graph structures for effectively capturing the dynamic spatial and temporal correlations. Finally, a spatial–temporal feature fusion module based on a gating fusion mechanism (STM-GM) is proposed to effectively learn and leverage the inherent spatial–temporal relationships for traffic flow forecasting. Experimental results from three real-world traffic flow datasets demonstrate the superior performance of the proposed STC-DGCN compared with state-of-the-art traffic flow forecasting models.

Domain adaptation between heterogeneous time series data: A case study on real-time rotary machinery fault diagnosis

Artificial intelligence (AI) in fault diagnosis has emerged as a significant tool for addressing anomalies in rotary machineries (RMs). However, the success of AI models in this context largely depends on situations where the distribution of samples in the training dataset closely aligns with that of the testing dataset. In addition, practically, it is very difficult, costly, or even impossible to collect sufficient time series samples with proper labeling for AI modeling. To address these issues, this paper presents a study developing a structured methodology called frequency-domain supervised domain adaptation (FDSDA) considering heterogeneous data from versatile sensor technologies. In the context of application, two high-dimensional datasets were generated, each following distinct distributions: one being the vibration signals and the other the acoustic signals. The modeling begins with transforming signals from the time domain to the frequency domain through frequency analysis. Subsequently, a domain adaptation model is implemented, adopting an ensemble learning as an estimator to address the covariate shift challenge. Two experiments are conducted with the proposed method for binary and multi-class fault diagnosis of RMs. The results demonstrate that, despite having limited training and target data, the proposed method can effectively detect RMs fault conditions by outperforming the benchmark method.

787. THE CHALLENGE OF GASTRO-ESOPHAGEAL ONCOLOGICAL SURGERY AFTER BARIATRIC PROCEDURES: EXPERIENCE OF A REFERRAL HIGH-VOLUME UPPER GI SURGICAL CENTER

Abstract Background The effectiveness of bariatric surgery for weight loss in morbidly obese patients has been well established and it's becoming more common as many surgeons are trained to perform them safely, even in older patients. Obesity is associated with reflux and hormonal imbalances that increase the risk of Barrett's esophagus, esophageal adenocarcinoma and gastric cancer. Bariatric procedures alter gastric anatomy, vascularization and lymphatic drainage and render subsequent upper gastrointestinal surgery for malignancies a technical challenge, that will present more and more in the future. We present laparoscopic sleeve gastrectomy, gastric bypass and gastric banding conversion to oncological respective surgeries. Methods We retrospectively reviewed the medical records of our referral high-volume center for upper gastrointestinal surgery and identified patients who underwent upper gastrointestinal oncologic surgery from January 1998 to May 2024. We selected all patients who had previously undergone bariatric surgery (laparoscopic sleeve gastrectomy, laparoscopic gastric bypass, gastric banding, or other less common procedures). Three patients met our criteria, and we recorded: demographic characteristics, surgical technique, oncologic surgical outcome, postoperative complications, long-term complications, and oncologic follow-up. Results A hybrid IvorLewis esophagectomy was performed years after a sleeve gastrectomy for distal esophageal adenocarcinoma. The sleeve was used for gastric pull-up, adequately vascularized by the right gastroepiploic artery as demonstrated by intraoperative indocyanine green. A patient developed adenocarcinoma of the cardia after Roux-en-Y gastric bypass, infiltrating the gastric pouch, the gastric remnant and liver S2. After neoadjuvant chemotherapy, an open Ivor-Lewis esophagectomy was performed using the gastric remnant for a pull-up. In a third patient, a locally advanced adenocarcinoma of the stomach developed after gastric banding. After neoadjuvant chemotherapy, an open total gastrectomy with Roux-en-Y esophagojejunostomy was performed. Conclusion We report three cases of successful esophago-gastric oncological surgery in patients with altered anatomy and vascularization due to previous bariatric surgery, without major postoperative complications or mortality. Oncologic benchmarks for both esophageal and gastric surgery were met as a high lymph node yield and oncologic margins were maintained. Our heterogeneous series suggests that sleeve gastrectomy, gastric bypass, and gastric banding surgery can be successfully converted to resective surgeries after a metachronous neoplasia diagnosis in high-volume upper gastrointestinal centers. These and similar modified reconstructive techniques are likely to be used in the future as metabolic surgery and esophago-gastric cancer are increasing.

Electric theft detection based on multi-source heterogeneous time series GRU network

Abstract Electricity theft has long been one of the major problems faced by power supply enterprises. To improve the robustness and accuracy of power theft detection, this article explores the method of multi-source heterogeneous time series feature fusion and designs a gated cyclic unit network model that adapts to its feature fusion. Firstly, through correlation analysis, it is verified that there is a logical correlation between different time series features and classification features, providing a theoretical basis for feature fusion. Then, an encoder decoder model framework is constructed with an attention mechanism to achieve effective fusion and state detection of user multi-source time series features. The experimental results show that compared to a single data source, the fusion of multi-source features can significantly improve detection performance, and the designed model is superior to the control model. This study provides a reference for constructing an efficient power theft detection system and also provides examples of multi-source heterogeneous feature fusion in related fields.

Double outlet right ventricle accompanied by complex congenital anomalies: A rare association

Double outlet right ventricle (DORV) comprises a heterogeneous series of associated cardiac anomalies that involve the RV outflow tract in which both of the great arteries arise entirely or predominantly from the RV. The anatomic dysmorphology is variable, and there may be associated cardiac anomalies. We report here a case of a complex cyanotic heart disease in a 14 year old adolescent comprising of DORV accompanied by multiple congenital cardiac anomalies: single atrium, complete AV canal defect, severe AV valve regurgitation, D-malposition of great arteries, subaortic obstruction and pulmonary valvular stenosis.

ENUMERATION AS A MEANS TO STRUCTURE THE INSTITUTES DOMAIN WITHIN ECONOMIC SYSTEM IN ENGLISH ECONOMIC DISCOURSE

The paper examines the concept of enumeration from the perspective of cognitive linguistics, focusing on economic discourse as its subject matter. Enumeration serves as a tool for organizing and representing professional conceptual information. This cognitive approach to the study of enumeration involves analyzing both homogeneous and heterogeneous enumerative sequences. In economic discourse, homogeneous enumerative series consist of elements within the same semantic field, typically representing economic sectors of human activity. Conversely, heterogeneous enumerative series indicate the intersection of various spheres of human activity within the economic context.The structure of enumerative series in economic discourse has two main components: linear and hierarchical. The linear structure is characterized by elements arranged in direct sequence. In this study, linear enumerations are identified by the presence of two or more elements, with the most effective and accurate information transfer occurring in three-component structures. Additionally, the study notes enumerative block. As a result of the enumerative elements of the main enumerative series is subdivided into a subordinate system were defined series analysis such domains of the economic cases where a sentence contains multiple enumerative series. In case of a hierarchical structure one of the as: the processes domain, the institutes domain, the instruments domain and the indicators domain. The institutes domain is analysed in this paper. The study of homogeneous enumerative series within the institutes domain facilitates differentiation of institutions by type and form of economic activity. The types identified include financial institutions, companies (such as insurance companies and private non-farm establishments), with cooperatives serving as one form of organizational activity. The analysis of heterogeneous enumerative series revealed a more varied set of institutional types, encompassing companies like insurance companies, along with various organizational forms of economic activity, such as partnerships, trade unions, and trade associations. Overall, this research establishes the role of both homogeneous and heterogeneous enumerative series in structuring texts and conceptualizing content within economic discourse.

Improving forecasts for heterogeneous time series by “averaging”, with application to food demand forecasts

A common forecasting setting in real-world applications considers a set of possibly heterogeneous time series of the same domain. Due to the different properties of each time series, such as length, obtaining forecasts for each individual time series in a straightforward way is challenging. This paper proposes a general framework utilizing a similarity measure in dynamic time warping to find similar time series to build neighborhoods in a k-nearest neighbor fashion and improve forecasts of possibly simple models by averaging. Several ways of performing the averaging are suggested, and theoretical arguments underline the usefulness of averaging for forecasting. Additionally, diagnostic tools are proposed for a deep understanding of the procedure.

Mining of heterogeneous time series information for predicting chlorophyll accumulation in oceans

Harmful algal blooms cause environmental harm, financial losses, and disease epidemics. It is also known that the algal blooms cannot be eradicated; hence the best option is to foresee their growth and regulate it. Machine learning algorithms can be used to forecast their presence and further classify the threat that each concentration level presents. In this research work, the dataset collected from Santa Monica, US region is analyzed and processed to predict algae concentration using machine learning algorithms. In this process, the machine learning models such as multiple linear regression, Regression Gradient Boosting Decision Tree (RGBDT), and Hidden Markov Model (HMM) are applied to predict the chlorophyll (Chl-a) content, which serves as a proxy for the presence of algae in the water. The obtained results show that for prediction, the Multilinear regression model outperforms the RGBDT (Regression Gradient Boosting Decision Tree) algorithm. Similarly, for modeling chlorophyll using HMM (Hidden Markov Model), parameter bbp555.00_sd is the best among parameters like aot443.00_sd, kd490.00_sd, poc_sd and pic_sd. The multiple linear regression model gave an adjusted R-squared error of 0.94 with the parameter pic_sd having the least VIF value of 1.78 followed by aot and bbp which have VIF<5 (2.28 and 4.95 respectively). The outcome of the HMM-based model represents the probability of the presence of chlorophyll given the presence of each of the variables individually. From the results, it is observed that bbp has the highest probability of 0.405 implying that there is a 40% chance of chlorophyll in the presence of bbp.

Structured Estimation of Heterogeneous Time Series

Structured Estimation of Heterogeneous Time Series

Prognostic value of preoperative inflammatory ratios in early glottic cancer treated with transoral laser surgery.

There is growing evidence regarding the prognostic utility of ratios such as neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and systemic immune-inflammatory index (SIII) in head and neck squamous cell carcinoma (HNSCC). However, most studies to date include heterogeneous series with different treatments or tumor subsites. We collected data from 201 patients with stage I-II glottic squamous cell carcinoma treated with transoral laser surgery. NLR, PLR, and SIII were calculated from preoperative cell blood count, cut-off points were obtained by ROC curve analysis, and survival rates were calculated. High NLR (p = 0.012) and SIII (p = 0.037), but not PLR (p = 0.48), were associated with worse disease-specific survival (DSS). A similar trend was observed with overall survival (OS), although it did not reach statistical significance. On multivariable analyses, both high NLR (HR = 3.8, 95% CI = 1.5-9.9, p = 0.006) and high SIII (HR = 2.77, 95% CI = 1.1-6.9, p = 0.03) were significantly associated with shortened DSS. Preoperative NLR and SIII emerge as independent prognostic biomarkers for early-stage surgically treated glottic tumors and could guide individualized follow-up.

Heterogeneous Multivariate Functional Time Series Modeling: A State Space Approach

Heterogeneous Multivariate Functional Time Series Modeling: A State Space Approach

Extended Multiresolution Analysis for Filtering Incomplete Heterogeneous Geophysical Time Series

Extended Multiresolution Analysis for Filtering Incomplete Heterogeneous Geophysical Time Series

Leat-associated seizures the possible role of EAAT2, pyruvate carboxylase and glutamine synthetase

BackgroundDrug-resistant epilepsy is a common condition in patients with brain neoplasms. The pathogenesis of tumor-associated seizures is poorly understood. Among the possible pathogenetic mechanisms, the increase in glutamate concentration has been proposed. Glutamate transporters, glutamine synthetase and pyruvate carboxylase are involved in maintaining the physiological concentration of glutamate in the intersynaptic spaces. In our previous research on angiocentric gliomas, we demonstrated that all tumors lacked the expression of the main glutamate transporter EAAT2, while the expression of glutamine synthetase and pyruvate carboxylase was mostly preserved. MethodsIn the present study, we evaluated the immunohistochemical expression of EAAT2, glutamine synthetase and pyruvate carboxylase in a heterogeneous series of 25 long-term epilepsy-associated tumors (10 dysembryoplastic neuroepithelial tumors, 7 gangliogliomas, 3 subependymal giant cell astrocytomas, 3 rosette forming glioneuronal tumors, 1 diffuse astrocytoma MYB- or MYBL1-altered and 1 angiocentric glioma). In order to evaluate the incidence of variants in the SLC1A2 gene, encoding EAAT2, in a large number of central nervous system tumors we also queried the PedcBioPortal. ResultsEAAT2 protein expression was lost in 9 tumors (36 %: 3 dysembryoplastic neuroepithelial tumors, 1 ganglioglioma, 3 subependymal giant cell astrocytomas, 1 diffuse astrocytoma MYB- or MYBL1-altered and 1 angiocentric glioma). Glutamine synthetase protein expression was completely lost in 2 tumors (8 %; 1 ganglioglioma and 1 diffuse astrocytoma MYB- or MYBL1-altered). All tumors of our series but rosette forming glioneuronal tumors (in which neurocytic cells were negative) were diffusely positive for pyruvate carboxylase. Consultation of the PedcBioPortal revealed that of 2307 pediatric brain tumors of different histotype and grade, 20 (< 1%) had variants in the SLC1A2 gene. Among the SLC1A2-mutated tumors, there were no angiocentric gliomas or other LEATs ConclusionsIn conclusion, unlike angiocentric gliomas where the EAAT2 loss is typical and constant, the current study shows the loss of EAAT2 expression only in a fraction of the LEATs. In these cases, we may hypothesize some possible epileptogenic role of the EAAT2 loss. The retained expression of pyruvate carboxylase may contribute to determining a pathological glutamate excess unopposed by glutamine synthetase that resulted expressed to a variable extent in the majority of the tumors. Furthermore, we can assume that the EAAT2 loss in brain tumors in general and in LEATs in particular is more conceivably epigenetic.

Unravelling the Role of Vegetation Dynamics in the Execution of ArcSWAT Hydrological Modeling for Cumulative Streamflow of a Tibetan Watershed

Monitoring vegetation change and hydrological variation is crucial as they are useful means of appraising the ecological environment and managing water resources in water-resource-sensitive regions. The leaf area index (LAI) describes water consumption in hydrologic processes and is an important vegetation variable for water budgeting in catchments. As part of the Soil and Water Assessment Tool (SWAT), LAI is a significant parameter, which links vegetation dynamics with the hydrological cycle. In the current study, we have aimed to describe the Lhasa River (LR) cumulative streamflow based on simulation scenarios obtained with the SWAT model. After dispensing a heterogeneous LAI time series developed by MODIS NDVI as a source of data at the HRU level (SWAT-synthetic LAI scenario), the study has produced a better representation of LR cumulative streamflow in terms of the selected evaluation criteria, encompassing the SWAT-baseline (SWAT-B scenario)-simulated and SWAT-built-in LAI-influenced (SWAT-LAI scenario) LR cumulative streamflow. The study has revealed a close relationship between the observed and the SWAT-SLAI-scenario-generated LR streamflow, with a similar MK trend for the study time span. The LAI has been found to share a close relationship with LR streamflow, as both the LAI and LR streamflow are found to be influenced by the rainfall received in the Lhasa River Basin (LRB). The study is instrumental in understanding the association between LR streamflow, vegetation change, and the climatic conditions of the Lhasa River Basin (LRB).

Partial Ankle Arthroplasty: Talus Resurfacing for Mild-to-Moderate Osteoarthritis and Talus Hemiarthroplasty for Complex Osteochondral Lesions.

We present prospective short-term results of a limited patient series treated with two innovative partial ankle arthroplasties: talar dome resurfacing for mild-to-moderate ankle osteoarthritis and talar shoulder hemiarthroplasty for chronic medial osteochondral lesions of the talus. Eleven patients underwent talus resurfacing and six patients had talar hemiarthroplasty. The outcome was followed by patient-reported measures and by pursuing serious adverse events or implant failures over a 2-year period. Progression of ankle osteoarthritis, peri-implant bone changes, and implant migration were followed radiographically. Active dorsiflexion increased from 3° to 10° in resurfacing and from 15° to 22° in hemiarthroplasty. Patient-reported ankle function, quality of life, and activity level tended to improve only slightly after resurfacing (cumulative Foot and Ankle Outcome Score, from 41 to 42; Foot and Ankle Ability Measures for daily activities [FAAM-ADL], from 43 to 46; EQ-5D-3L time trade-off, from 0.38 to 0.39; Tegner activity scale score, from 1.6 to 2.0) but moderately after hemiarthroplasty (cumulative Foot and Ankle Outcome Score, from 58 to 68; FAAM-ADL, from 37 to 71; EQ-5D-3L time trade-off, from 0.53 to 0.72; Tegner activity scale score, from 3.1 to 3.1). No implant-related radiographic changes, implant failures, or implant-related revision surgeries were recorded. Based on a small and heterogeneous prospective case series, both partial ankle implants investigated were safe and stable over 2-year follow-up, without any radiographic osteoarthritis progression of the remaining joint. However, patient-reported ankle function, quality of life, and activity level showed a tendency toward only minor improvement after resurfacing but a moderate increase after hemiarthroplasty.

INTERSPECIFIC GENETIC DISTANCES IN THE GENUS YPONOMEUTA LATR. (LEPIDOPTERA: YPONOMEUTIDAE) ON COI AND COII FRAGMENTS AND THE USE OF THE RESULTS OF THEIR COMPARATIVE ANALYSIS FOR SOLVING TAXONOMIC TASKS

Comparative genetic analysis of the nucleotide sequences of COI and COII mitochondrial fragments showed that the degree of divergence of ermine moths of the genus Yponomeuta on both fragments is comparable. In the studied taxa, interspecific genetic distances are within 0.2–9.9% for the COI fragment and 0.5–9.6% for the COII fragment. Based on the results of the nucleotide sequences analysis of the COI fragment, the conspecificity of Y. refrigeratus and Y. evonymella, the interval of genetic distances between samples of which is 0.2–0.5%, was proved, and the validity of the synonymization of these species was confirmed. Analysis of nucleotide sequences by COI and COII fragments in samples collected on the Sakhalin Is., as well as the rooting of their clusters in groups with conspecific species, made it possible to identify them and confirm the habitat of Y. polystictus and Y. spodocrossus on Sakhalin Is. In general, the molecular genetic method, based on a comparative analysis of interspecific genetic distances by mitochondrial fragments of COI and COII, is effective for determining the status of a taxon, specimen identification in morphologically weakly divergent and variable species, or in species described by one sex, with an unknown other sex, as well as for delimitation of the species in heterogeneous series.

Extended singular spectrum analysis for processing incomplete heterogeneous geodetic time series

Extended singular spectrum analysis for processing incomplete heterogeneous geodetic time series

Outcomes and Complications After Bone Marrow Aspirate Harvest From the Calcaneus.

Bone marrow aspirate (BMA) is used with the putative goal of enhancing healing of injured tissue. The most common sites to harvest BMA are the iliac crest, the tibia, and the calcaneus. Investigators have found that the tibia and calcaneus have fewer progenitor cells than the iliac crest. This retrospective review evaluates the efficacy and safety profile of harvesting BMA from the calcaneus bone. We hypothesized that harvesting BMA from the calcaneus will have high efficacy and safety profile with low complication rates. A retrospective chart review of patients undergoing bone marrow aspiration from the calcaneus bone from January 2019 to October 2022 was performed. The main data points evaluated were patient satisfaction and pain level, follow-up times, quantity of BMA harvested, fusion rates, and complications including nerve damage, infections, and nonunions. There were 45 (34 female and 11 male) patients who underwent 45 procedures. The average age, BMI, and amount of BMA were 45.1 years (range 23-79), 33.1 (range 19.2-61.3), and 10.3 mL (range 2-40), respectively. There was no pain reported at the final follow-up in 32 patients (71.1%), and there was minimal to no pain in the area of surgery in the remaining 13 patients (28.9%). The 10 cases of fusion successfully healed with combined use of allograft and BMA. The average follow-up time was 12.3 months (range 2.4-33.7). There were no infections, wound complications, or nerve injuries. Harvesting BMA from the calcaneus bone is a safe procedure. In this heterogenous series where BMA augmented other surgical strategies, patients had little to no pain and there were no cases of wound complication or iatrogenic neuropraxia after undergoing bone marrow harvest from the calcaneus. Level IV, case series.

Choose Wisely: An Extensive Evaluation of Model Selection for Anomaly Detection in Time Series

Anomaly detection is a fundamental task for time-series analytics with important implications for the downstream performance of many applications. Despite increasing academic interest and the large number of methods proposed in the literature, recent benchmark and evaluation studies demonstrated that no overall best anomaly detection methods exist when applied to very heterogeneous time series datasets. Therefore, the only scalable and viable solution to solve anomaly detection over very different time series collected from diverse domains is to propose a model selection method that will select, based on time series characteristics, the best anomaly detection method to run. Existing AutoML solutions are, unfortunately, not directly applicable to time series anomaly detection, and no evaluation of time series-based approaches for model selection exists. Towards that direction, this paper studies the performance of time series classification methods used as model selection for anomaly detection. Overall, we compare 17 different classifiers over 1800 time series, and we propose the first extensive experimental evaluation of time series classification as model selection for anomaly detection. Our results demonstrate that model selection methods outperform every single anomaly detection method while being in the same order of magnitude regarding execution time. This evaluation is the first step to demonstrate the accuracy and efficiency of time series classification algorithms for anomaly detection, and represents a strong baseline that can then be used to guide the model selection step in general AutoML pipelines.

ICA and IVA bounded multivariate generalized Gaussian mixture based hidden Markov models

Machine learning (ML), a branch of artificial intelligence (AI), is an area of computational science that is concerned with the analysis and interpretation of patterns and structures in data to enable learning and decision-making without the participation of a human. Hidden Markov models (HMMs), which have been acknowledged for decades but have recently made a significant revival in machine learning, are one of the most impressively powerful probabilistic models. HMMs are frequently employed in machine learning to model heterogeneous time series data. In this paper, we integrate independent component analysis (ICA) and ICA with a bounded multivariate generalized Gaussian mixture model (ICA-BMGGMM) into the HMM approach. One limitation of ICA is that it assumes the sources to be independent from each other. This assumption can be relaxed by combining independent vectors analysis (IVA) and IVA with the BMGGMM (IVA-BMGGMM) into the HMM approach to improve their modeling capability. We validate our proposed models using a variety of applications, such as human action recognition, speech recognition, and energy disaggregation. The results presented in the paper demonstrate the effectiveness of the proposed approaches for modeling different types of data. These data include KTH and Weizmann datasets for human action recognition, TIMIT and SDR for speech recognition, REDD dataset for energy disaggregation and EEG dataset for elliptic seizure classification. For all conducted experiments, our proposed models outperform other comparing models for all performance metrics such as accuracy, sensitivity, and precision. The best detection results were found using the IVABMGGMM-HMM for the reported experiments.