Irregular Time Research Articles

Adapting to the stream: an instance-attention GNN method for irregular multivariate time series data

Multivariate time series (MTS) data are vital for various applications, particularly in machine learning tasks. However, challenges such as sensor failures can result in irregular and misaligned data with missing values, thereby complicating their analysis. While recent advancements use graph neural networks (GNNs) to manage these Irregular Multivariate Time Series (IMTS) data, they generally require a reliable graph structure, either pre-existing or inferred from adequate data to properly capture node correlations. This poses a challenge in applications where IMTS data are often streamed and waiting for future data to estimate a suitable graph structure becomes impractical. To overcome this, we introduce a dynamic GNN model suited for streaming characteristics of IMTS data, incorporating an instance-attention mechanism that dynamically learns and updates graph edge weights for realtime analysis. We also tailor strategies for high-frequency and low-frequency data to enhance prediction accuracy. Empirical results on real-world datasets demonstrate the superiority of our proposed model in both classification and imputation tasks.

Association of self-monitoring performance of cognitive performance with personal diurnal preference when sleep-deprived

ABSTRACT In modern society, many workers struggle with sleep deprivation due to their work schedules and excessive workloads. Accurate self-awareness and self-monitoring abilities are crucial for workers to adopt risk-coping strategies and protective behaviors when fatigued. The current study examined the relationship between chronotypes and self-monitoring performance during 24 h of sleep deprivation. The study involved 26 male adults in a two-night experiment, and participants’ diurnal preferences were evaluated using the Morningness-Eveningness Questionnaire (MEQ). Self-monitoring performance was calculated by comparing actual task performance with self-rated predicted or post-estimated performance in the psychomotor vigilance task (PVT) and Digit Symbol Substitution Task (DSST). The study found that task and self-monitoring performances in the PVT and DSST were maintained until around 4:00 h, after which they began to deteriorate. Individuals with a higher MEQ score, indicating a stronger tendency towards a morning type, showed inaccurate self-monitoring, particularly in the final quarter of the sleep deprivation experiment, due to overly optimistic predictions. However, only prediction accuracy and not post-estimation showed this correlation. This study highlights the importance of considering an individual’s chronotype in workplace management, particularly in workplaces with irregular work timings, rotating shifts, and long working hours, to ensure better occupational safety.

Gaussian processes for time series with lead-lag effects with applications to biology data.

Investigating the relationship, particularly the lead-lag effect, between time series is a common question across various disciplines, especially when uncovering biological processes. However, analyzing time series presents several challenges. Firstly, due to technical reasons, the time points at which observations are made are not at uniform intervals. Secondly, some lead-lag effects are transient, necessitating time-lag estimation based on a limited number of time points. Thirdly, external factors also impact these time series, requiring a similarity metric to assess the lead-lag relationship. To counter these issues, we introduce a model grounded in the Gaussian process, affording the flexibility to estimate lead-lag effects for irregular time series. In addition, our method outputs dissimilarity scores, thereby broadening its applications to include tasks such as ranking or clustering multiple pairwise time series when considering their strength of lead-lag effects with external factors. Crucially, we offer a series of theoretical proofs to substantiate the validity of our proposed kernels and the identifiability of kernel parameters. Our model demonstrates advances in various simulations and real-world applications, particularly in the study of dynamic chromatin interactions, compared to other leading methods.

Effects of light, electromagnetic fields and water on biological rhythms.

The circadian rhythm controls a wide range of functions in the human body and is required for optimal health. Disruption of the circadian rhythm can produce inflammation and initiate or aggravate chronic diseases. The modern lifestyle involves long indoor hours under artificial lighting conditions as well as eating, working, and sleeping at irregular times, which can disrupt the circadian rhythm and lead to poor health outcomes. Seasonal solar variations, the sunspot cycle and anthropogenic electromagnetic fields can also influence biological rhythms. The possible mechanisms underlying these effects are discussed, which include resonance, radical-pair formation in retina cryptochromes, ion cyclotron resonance, and interference, ultimately leading to variations in melatonin and cortisol. Intracellular water, which represents a coherent, ordered phase that is sensitive to infrared light and electromagnetic fields, may also respond to solar variations and man-made electromagnetic fields. We describe here various factors and underlying mechanisms that affect the regulation of biological rhythms, with the aim of providing practical measures to improve human health.

Delaying bud-break on pecan trees: a Bayesian longitudinal multinomial regression approach

A multivariate Bayesian Probit model is adapted to analyze a longitudinal multiclass-ordinal response, with a linear plateau as the longitudinal model. Measurements on pecan bud growth were collected on irregular time intervals, about a week apart from late March to mid April, using a six-level ordinal scale. The data are from two randomized complete block designs with four blocks each. The experiments were setup and initiated in 2018 in a pecan orchard, at two different locations, to evaluate the effect of two sets of four treatments on delaying growth of recently broken pecan buds to minimize bud loss due to low temperatures. A simulation study was successfully carried out to validate the model implementation. Treatment 3 of Experiment 1 was associated with the greatest reduction in bud growth rate. In Experiment 2, Treatments 2 and 3 had some effect on delaying bud growth. Although treatment effects were not statistically different in either experiment, this paper presents a practical and efficient modeling technique for longitudinal multinomial ordinal data, a common data type in applied agricultural research studies.

Health impacts of a remotely delivered prolonged nightly fasting intervention in stressed adults with memory decline and obesity: A nationwide randomized controlled pilot trial.

Cognitive decline is intricately linked to various factors such as obesity, stress, poor sleep, and circadian rhythm misalignment, which are interrelated in their impact on cognitive health. Irregular food-intake timing further compounds these issues. The practice of prolonged nightly fasting (PNF) may help synchronize food intake with circadian rhythms, potentially mitigating adverse effects of cognitive decline and associated factors. A pilot nationwide, remotely delivered, 2-arm randomized controlled trial was conducted to assess the 8-week outcomes of cognition, stress, and sleep, after a PNF intervention (14-hr nightly fast, 6 nights/week, no calories after 8 pm) compared to a health education (HED) control condition. Participants were living with memory decline, stress, and obesity and had weekly check-in calls to report fasting times (PNF) or content feedback (HED). Participants were enrolled from 37 states in the US; N = 58, 86% women, 71% white, 93% non-Latinx, mean (SD) age 50.1 (5.1) years and BMI 35.6 (3.6) kg/m2. No group differences existed at baseline. Linear mixed-effects models were used to compare outcome change differences between groups. Compared to the HED control, the PNF intervention was associated with improved sleep quality (B = -2.52; SE = 0.90; 95% CI -4.30--0.74; p = 0.006). Perceived stress and everyday cognition significantly changed over time (p < 0.02), without significant difference by group. Changing food intake timing to exclude nighttime eating and promote a fasting period may help individuals living with obesity, memory decline, and stress to improve their sleep. Improved sleep quality may lead to additional health benefits.

A sleep promotion program for insufficient sleep among adolescents: a pilot feasibility randomized controlled trial.

To examine the feasibility, acceptability, and impact of a Sleep Promotion Program (SPP). This pilot trial randomized adolescents (13-15y) with insufficient sleep duration and irregular sleep timing to SPP-continuation (n=24; SPP in month 1, continuation treatment in month 2) or monitoring-SPP (n=20; monitoring in month 1, SPP in month 2). SPP included one clinician session and at-home delivery of web-based reports of each youth's sleep diary data with accompanying intervention questions that prompt youth to engage in sleep behavior change. Attrition rate primarily measured feasibility. Program satisfaction measured acceptability. Total sleep time (TST), sleep timing, and sleep timing regularity were measured via sleep diary at baseline, follow-up 1, and follow-up 2 (each ∼1 month apart). Linear mixed effects models compared treatment arms on changes in sleep from baseline to follow-up 1 (month 1). We also compared changes in sleep during month 1 to changes in sleep during month 2 among SPP-continuation participants. Attrition rate was 8.5%. 96.5% participants rated the quality of care received as good or excellent. In month 1, SPP-continuation youth showed a significantly greater increase in mean TST than monitoring-SPP youth (0.57 vs. -0.38 hours; contrast=0.95; CI=0.14, 1.76, p=0.024). SPP-continuation participants showed an increase in TST during month 1 (0.51h) but a decrease during month 2 (-0.74 h; contrast=-1.24, CI=-2.06, -0.42, p=0.005). No other significant effects were observed. SPP is highly feasible, acceptable, and associated with a significant increase in TST early in treatment. Registry: ClinicalTrials.gov; Name: Targeted Intervention for Insufficient Sleep among Typically-Developing Adolescents; Identifier: NCT04163003; URL: https://clinicaltrials.gov/ct2/show/NCT04163003.

Spatio-temporal analysis of agroforestry systems in hotan using multi-source remote sensing and deep learning

In the Hotan region of Xinjiang, where arable land is scarce, an agroforestry system integrating walnut trees with crops has been implemented to maximize land-use efficiency. While this dense planting enhances land use, it also limits the availability of light to the understory crops, potentially impacting their yield and quality. To address this issue and enhance the system's sustainability and productivity, precise delineation of the planting structure is critical. This study proposes a novel framework that leverages multi-source remote sensing data combined with advanced deep learning techniques to analyze the agroforestry planting structure. The methodological approach consists of three key phases. First, an instance segmentation model was employed to extract farmland parcels from high-resolution imagery, providing a basis for vegetation classification. Next, a time series model using irregular satellite image time series (irSITS) tracked the growth dynamics of the vegetation. Finally, the spatial planting structure of the walnut trees was quantified using the d-LinkNet model, integrated with a template filling algorithm.The results demonstrated a classification accuracy of 97.85 % in extracting parcel-level planting structures, identifying 42,955 farmland parcels, including 21,153 intercropped parcels. The temporal and spatial characteristics of the agroforestry system were then analyzed, leading to a grading of canopy cover and walnut tree density within the intercropped areas. This comprehensive spatiotemporal planting structure offers a valuable foundation for informed local agricultural policy adjustments. In conclusion, this approach advances the understanding of complex agroforestry systems and provides a robust scientific basis for optimizing intercropping practices, contributing to sustainable agricultural development in arid regions.

Aerodynamic interference effect and flow field mechanism of two tandem rectangular columns with a small width–thickness ratio at a high Reynolds number

This paper conducted wind tunnel tests and large eddy simulations to study the aerodynamic interference effect and flow field mechanism of two tandem rectangular columns with a small width–thickness ratio (B/D = 0.25) at a high Reynolds number (Re = 2.1 × 105). The spacing ratio (L/B) varied from 0.2 to 20. Results showed that single-bluff body, reattachment, and co-shedding regimes occur at 0.2 ≤ L/B &amp;lt; 3, 3 ≤ L/B &amp;lt; 10, and 12 &amp;lt; L/B ≤ 20, respectively. In the single-blunt body regime, the mean drag coefficient of the upstream column, the fluctuating lift coefficient of the downstream column, and the Strouhal number of both columns are significantly amplified compared to a single column. These amplification effects are linked to the reattachment of the recirculation flow between columns and a reduced wake recirculation length. In the reattachment regime, the amplification effects in the mean drag coefficient and the fluctuating lift coefficient are diminished, but the Strouhal number still shows a marked amplification due to the short wake recirculation length. In the co-shedding regime, the amplification effects in aerodynamic force coefficients disappear. In addition to the three classic flow regimes, a bistable flow regime was identified at 10 ≤ L/B ≤ 12, where the aerodynamic characteristics observed in the reattachment and the co-shedding regimes alternate randomly at irregular time intervals.

Robust extraction of pneumonia-associated clinical states from electronic health records

Mining of electronic health records (EHR) promises to automate the identification of comprehensive disease phenotypes. However, the realization of this promise is hindered by the unavailability of generalizable ground-truth information, data incompleteness and heterogeneity, and the lack of generalization to multiple cohorts. We present here a data-driven approach to identify clinical states that we implement for 585 critical care patients with suspected pneumonia recruited by the SCRIPT study, which we compare to and integrate with 9,918 pneumonia patients from the MIMIC-IV dataset. We extract and curate from their structured EHRs a primary set of clinical features (53 and 59 features for SCRIPT and MIMIC-IV, respectively), including disease severity scores, vital signs, and so on, at various degrees of completeness. We aggregate irregular time series into daily frequency, resulting in 12,495 and 94,684 patient-day pairs for SCRIPT and MIMIC, respectively. We define a "common-sense" ground truth that we then use in a semisupervised pipeline to optimize choices for data preprocessing, and reduce the feature space to four principal components. We describe and validate an ensemble-based clustering method that enables us to robustly identify five clinical states, and use a Gaussian mixture model to quantify uncertainty in cluster assignment. Demonstrating the clinical relevance of the identified states, we find that three states are strongly associated with disease outcomes (dying vs. recovering), while the other two reflect disease etiology. The outcome associated clinical states provide significantly increased discrimination of mortality rates over standard approaches.

Research on the Operation Status and Development Strategy of Wechat Public Account

With the continuous development and progress of global science and technology, the demand of users is increasing, and Wechat, as an important information exchange platform, plays an indispensable role in daily communication and communication. The Wechat public account, one of its main functions, plays a major advantage in disseminating information and providing facilitation services. However, how to create a fair public account environment so that each content production subject achieves satisfactory communication effects needs to be further studied. This paper studies the operation status and development strategy of Wechat public accounts. By searching and reading relevant materials and literature, it comprehensively analyzes the content production characteristics of different types of Wechat public accounts and finds that the development of science and technology and the needs of users are the main factors that promote the rapid development of Wechat public accounts. It is characterized by diversity of content and subject, fragmentation of content production, low publicity cost and high communication efficiency, accurate positioning of users, and secondary transmission. However, there are still many problems in the operation of Wechat public accounts, such as insufficient attention from content producers, homogeneity and vulgar content, irregular time of content release, fan conversion rate, and insufficient user loyalty. Content producers can win better communication effects by releasing high-quality content, enhancing user loyalty, expanding the dissemination scope by secondary communication, and using Wechat public accounts for product marketing in multiple ways.

Multi-Way adaptive Time Aware LSTM for irregularly collected sequential ICU data

In personalized predictive medicine, accurately modeling a patient’s illness and care processes is essential, given their inherent long-term temporal dependencies. However, electronic medical records contain episodic and irregularly timed data due to patients visiting hospitals based on treatment needs, resulting in unique patterns for each hospital stay. Consequently, when constructing a personalized predictive model, it is crucial to consider these factors in order to accurately capture the patient’s health journey.To address this challenge, we present a novel deep dynamic memory neural network called Multi-Way adaptive Time Aware LSTM (MWTA-LSTM). The primary objective of MWTA-LSTM is to leverage medical records, memorize illness trajectories and care processes, estimate current illness states, and predict future risks, thereby providing a high level of precision and predictive power. To enhance its capabilities, MWTA-LSTM extends the conventional Long Short-Term Memory (LSTM) model in two key ways. Firstly, it incorporates frequency measurement and the most recent observation(last observation) to enhance personalized predictive modeling of patient illnesses, enabling a more accurate understanding of the patient’s condition. Secondly, it parameterizes the cell state to handle irregular timing effectively, utilizing both frequency measurement and elapsed times. Furthermore, the model capitalizes on both to comprehend the impact of interventions on the course of illness on the cell state, facilitating the memorization of illness courses and improving its ability to capture the temporal dynamics of healthcare data, accommodating variations and irregularities in event and observation timing. Lastly, we introduce a novel adaptive pooling strategy that specifically targets and resolves outlier issues that can potentially occur during the analysis of EHR data. By incorporating these features, MWTA-LSTM significantly improves its ability to capture the temporal dynamics of healthcare data, accommodating variations and irregularities in event and observation timing.We validate the effectiveness of our proposed model through empirical experiments conducted on two real-world clinical datasets and three real-world time series datasets. Our results demonstrate the superiority of MWTA-LSTM over current state-of-the-art models and other robust baselines. This showcases the potential of MWTA-LSTM in advancing personalized predictive medicine by offering a more accurate and comprehensive approach to modeling patient health trajectories.

Semi-parametric sensitivity analysis for trials with irregular and informative assessment times.

Many trials are designed to collect outcomes at or around pre-specified times after randomization. If there is variability in the times when participants are actually assessed, this can pose a challenge to learning the effect of treatment, since not all participants have outcome assessments at the times of interest. Furthermore, observed outcome values may not be representative of all participants' outcomes at a given time. Methods have been developed that account for some types of such irregular and informative assessment times; however, since these methods rely on untestable assumptions, sensitivity analyses are needed. We develop a sensitivity analysis methodology that is benchmarked at the explainable assessment (EA) assumption, under which assessment and outcomes at each time are related only through data collected prior to that time. Our method uses an exponential tilting assumption, governed by a sensitivity analysis parameter, that posits deviations from the EA assumption. Our inferential strategy is based on a new influence function-based, augmented inverse intensity-weighted estimator. Our approach allows for flexible semiparametric modeling of the observed data, which is separated from specification of the sensitivity parameter. We apply our method to a randomized trial of low-income individuals with uncontrolled asthma, and we illustrate implementation of our estimation procedure in detail.

How Should Clinicians Partner With Families to Promote Children's Sleep Health?

Short sleep duration, poor sleep quality, and irregular timing of sleep are prevalent sleep troubles for children, but fully assessing children's sleep environments and effectively promoting children's overall sleep health is nearly impossible during brief clinical encounters. This commentary on a case suggests strategies for navigating this problem with a patient- and family-centered approach that prioritizes identifying family sleep-related beliefs, values, and goals and maintaining flexibility when offering evidence-based recommendations to improve children's sleep.

Predicting suicidal ideation from irregular and incomplete time series of questionnaires in a smartphone-based suicide prevention platform: a pilot study

Over 700,000 people die by suicide annually. Collecting longitudinal fine-grained data about at-risk individuals, as they occur in the real world, can enhance our understanding of the temporal dynamics of suicide risk, leading to better identification of those in need of immediate intervention. Self-assessment questionnaires were collected over time from 89 at-risk individuals using the EMMA smartphone application. An artificial intelligence (AI) model was trained to assess current level of suicidal ideation (SI), an early indicator of the suicide risk, and to predict its progression in the following days. A key challenge was the unevenly spaced and incomplete nature of the time series data. To address this, the AI was built on a missing value imputation algorithm. The AI successfully distinguished high SI levels from low SI levels both on the current day (AUC = 0.804, F1 = 0.625, MCC = 0.459) and three days in advance (AUC = 0.769, F1 = 0.576, MCC = 0.386). Besides past SI levels, the most significant questions were related to psychological pain, well-being, agitation, emotional tension, and protective factors such as contacts with relatives and leisure activities. This represents a promising step towards early AI-based suicide risk prediction using a smartphone application.

Novel efficient reservoir computing methodologies for regular and irregular time series classification

Time series is a data structure prevalent in a wide range of fields such as healthcare, finance and meteorology. It goes without saying that analyzing time series data holds the key to gaining insight into our day-to-day observations. Among the vast spectrum of time series analysis, time series classification offers the unique opportunity to classify the sequences into their respective categories for the sake of automated detection. To this end, two types of mainstream approaches, recurrent neural networks and distance-based methods, have been commonly employed to address this specific problem. Despite their enormous success, methods like Long Short-Term Memory networks typically require high computational resources. It is largely as a consequence of the nature of backpropagation, driving the search for some backpropagation-free alternatives. Reservoir computing is an instance of recurrent neural networks that is known for its efficiency in processing time series sequences. Therefore, in this article, we will develop two reservoir computing based methods that can effectively deal with regular and irregular time series with minimal computational cost, both while achieving a desirable level of classification accuracy.

Addressing local sparsity in massive agricultural machinery trajectories: A BiLSTM-GRU approach

Trajectory data acquired from GNSS (Global Navigation Satellite System) terminals on agricultural machinery are crucial for identifying agricultural machinery operation modes, evaluating agricultural machinery operational efficiency and exploring agricultural machinery trans-regional harvesting operation characteristics. However, GNSS terminals often experience signal delays due to factors such as weather conditions and environmental obstructions. These delays result in irregular time intervals between trajectory points, leading to local sparsity within the trajectory data, which subsequently reduces the accuracy of applications and analyses based on agricultural machinery trajectories. To address this issue, we propose a novel approach that leverages Bidirectional Long Short-Term Memory (BiLSTM) and Gated Recurrent Unit (GRU) networks, along with an attention mechanism, to mitigate the problem of local trajectory sparsity, and experiments were conducted using agricultural machinary trajectory data collected during the 2023 wheat harvest period. The results demonstrate the efficiency of our approach by successfully resolving the local sparsity of agricultural machinery trajectories. Moreover, each newly added trajectory point contains all original attributes (e.g., speed and direction). When integrated into state-of-the-art algorithms (e.g., DT, DBSCAN + rules, GCN) for identifying machinery operation modes, our method improves accuracies by 21.83 %, 26.86 %, and 1.17 %, respectively. Our approach effectively addresses the issue of local trajectory sparsity, thus providing assistance for applications and studies based on massive agricultural machinery trajectories.

Fund transfer fraud detection: Analyzing irregular transactions and customer relationships with self-attention and graph neural networks

This paper presents a method for identifying fraudulent fund transfers using real bank data, analyzing customer information, transactional activities, and customer relationships. The preprocessing step transforms high-dimensional, irregular transaction time series into regular time series, then further compresses them in a latent space using a self-attention-based autoencoder. To address the scarcity of fraudulent data samples and mitigate training issues caused by data imbalance, various deep generative models, including the conditional variational autoencoder and Wasserstein generative adversarial network, are applied to generate additional fraudulent raw data and augment fraud samples in latent space. The reparameterization trick is integrated into the encoder–decoder structure to boost the model’s generative capabilities. Additionally, a Graph Neural Network (GNN) is used to model customer relationships. The proposed approach utilizes end-to-end learning, integrating the autoencoder’s reconstruction loss, KL divergence loss (when reparameterization trick is applied), and classification loss for fraud detection. To optimize computational resources, neighborhood sampling for GNN is combined with mini-batch training for the autoencoder, improving both training efficiency and model reliability. Comprehensive experiments demonstrate the effectiveness of the proposed fusion network, highlighting the importance of each component and preprocessing step. For example, the areas under the precision–recall curves for fraud detection show notable improvements in our model. For suspicious transactions identified by the bank’s rules, other models range from 0.66% to 22.15%, while our model reached 27%. For non-suspicious transactions, other models range from 2.53% to 22.00%, with our model achieving 22.90%. This model has potential for wider applications in anomaly detection, particularly in datasets with irregular time series and complex customer relationships.

Intermittent microwave bursts of a semiconductor laser with an ultra-long loop for generating timing-based random bits.

Chaotic dynamics of semiconductor lasers under optical feedback are useful for random bit generation (RBG). By exploring on an ultra-long feedback loop, a single-mode laser in a route-to-chaos unveils an emission intensity with intermittent microwave bursts, for which a fast form of timing-based RBG is demonstrated. Each microwave burst corresponds to a packet of periodic intensity oscillations at the relaxation resonance. Numerous bursts are found intermittently within a round trip. Repetitions of these intermittent microwave bursts are observed across consecutive round trips. Randomness is extracted from the timing of the bursts as far as the feedback is reinitialized. With a 5-km fiber for feedback to the laser, over 104 intermittent bursts of microwave at 7 GHz are obtained per round trip, where the irregular timing in the envelope leads to RBG at 9.6 Gbps. Such experimental results feature the form of timing-based RBG that is very fast by comparison, carried by the microwave, and stored in the feedback nonlinear dynamics.

ChronoVectors: Mapping Moments through Enhanced Temporal Representation

Time-series data are prevalent across various fields and present unique challenges for deep learning models due to irregular time intervals and missing records, which hinder the ability to capture temporal information effectively. This study proposes ChronoVectors, a novel temporal representation method that addresses these challenges by enabling a more specialized encoding of temporal relationships through the use of learnable parameters tailored to the dataset’s dynamics while maintaining consistent time intervals post-scaling. The theoretical demonstration shows that ChronoVectors allow the transformed encoding tensors to map moments in time to continuous spaces, accommodating potentially infinite extensions of the sequence and preserving temporal consistency. Experimental validation using the Parking Birmingham and Metro Interstate Traffic Volume datasets reveals that ChronoVectors enhanced the predictive capabilities of deep learning models by reducing prediction error for regression tasks compared to conventional time representations, such as vanilla timestamp encoding and Time2Vec. These findings underscore the potential of ChronoVectors in handling irregular time-series data and showcase its ability to improve deep learning model performance in understanding temporal dynamics.