scholarly journals Effect of time-series length and resolution on abundance- and trait-based early warning signals of population declines

2019 ◽  
Author(s):  
A.A. Arkilanian ◽  
C.F. Clements ◽  
A. Ozgul ◽  
G. Baruah
Keyword(s):  
Time Series ◽  
Early Warning ◽  
Generation Time ◽  
Time Series Data ◽  
Series Data ◽  
Series Length ◽  
Warning Signals ◽  
Relative Time ◽  
Early Warning Signals ◽  
Population Collapse

AbstractNatural populations are increasingly threatened with collapse at the hands of anthropogenic effects. Predicting population collapse with the help of generic early warning signals (EWS) may provide a prospective tool for identifying species or populations at highest risk. However, pattern-to-process methods such as EWS have a multitude of challenges to overcome to be useful, including the low signal to noise ratio of ecological systems and the need for high quality time-series data. The inclusion of trait dynamics with EWS has been proposed as a more robust tool to predict population collapse. However, the length and resolution of available time series are highly variable from one system to another, especially when generation time is considered. As yet it remains unknown how this variability with regards to generation time will alter the efficacy of EWS. Here we take both a simulation- and experimental-based approach to assess the impacts of relative time-series length and resolution on the forecasting ability of EWS. We show that EWS’ performance decreases with decreasing length and resolution. Our simulations suggest a relative time-series length between ten and five generations and a resolution of half a generation are the minimum requirements for accurate forecasting by abundance-based EWS. However, when trait information is included alongside abundance-based EWS, we find positive signals at lengths and resolutions half of what was required without them. We suggest that, in systems where specific traits are known to affect demography, trait data should be monitored and included alongside abundance data to improve forecasting reliability.

scholarly journals Eigenstate Transition of Multi-Channel Time Series Data around Earthquakes

2021 ◽  
Vol 11 (23) ◽  
pp. 11407
Author(s):  
Akihisa Okada ◽  
Yoshiyuki Kaneda
Keyword(s):  
Time Series ◽  
Early Warning ◽  
Time Series Data ◽  
Series Data ◽  
Economic Damage ◽  
First Eigenvalue ◽  
Warning Signals ◽  
Early Warning Signals ◽  
The First Eigenvalue ◽  
The Matrix

To decrease human and economic damage owing to earthquakes, it is necessary to discover signals preceding earthquakes. We focus on the concept of “early warning signals” developed in bifurcation analysis, in which an increase in the variances of variables precedes its transition. If we can treat earthquakes as one of the transition phenomena that moves from one state to the other state, this concept is useful for detecting earthquakes before they start. We develop a covariance matrix from multi-channel time series data observed by an observatory on the seafloor and calculate the first eigenvalue and corresponding eigenstate of the matrix. By comparing the time dependence of the eigenstate to some past earthquakes, it is shown that the contribution from specific observational channels to the eigenstate increases before earthquakes, and there is a case in which the eigenvalue increases as predicted in early warning signals. This result suggests the first eigenvalue and eigenstate of multi-channel data are useful to identify signals preceding earthquakes.

scholarly journals Early warning signals predict emergence of COVID-19 waves

2021 ◽  
Author(s):  
Duncan A O'Brien ◽  
Christopher F Clements
Keyword(s):  
Time Series ◽  
Early Warning ◽  
Sequential Analysis ◽  
Time Series Data ◽  
Disease Outbreaks ◽  
Series Data ◽  
Warning Signals ◽  
Early Warning Signals ◽  
Policy Makers ◽  
The Uk

Early warning signals (EWSs) aim to predict changes in complex systems from phenomenological signals in time series data. These signals have recently been shown to precede the initial emergence of disease outbreaks, offering hope that policy makers can make predictive rather than reactive management decisions. Here, using daily COVID-19 case data in combination with a novel, sequential analysis, we show that composite EWSs consisting of variance, autocorrelation, and return rate not only pre-empt the initial emergence of COVID-19 in the UK by 14 to 29 days, but also the following wave six months later. We also predict there is a high likelihood of a third wave as of the data available on 9th June 2021. Our work suggests that in highly monitored disease time series such as COVID-19, EWSs offer the opportunity for policy makers to improve the accuracy of time critical decisions based solely upon surveillance data.

Development of A Pharmaceutical Early Warning Scoring Model for Cardiac Arrest Using Deep Learning: Retrospective Cohort Study (Preprint)

2020 ◽  
Author(s):  
Hsiao-Ko Chang ◽  
Hui-Chih Wang ◽  
Chih-Fen Huang ◽  
Feipei Lai
Keyword(s):  
Time Series ◽  
Cardiac Arrest ◽  
Early Warning ◽  
Time Series Data ◽  
Predictive Accuracy ◽  
Vital Signs ◽  
Sliding Window ◽  
Early Warning Score ◽  
Series Data ◽  
Dynamic Time

BACKGROUND In most of Taiwan’s medical institutions, congestion is a serious problem for emergency departments. Due to a lack of beds, patients spend more time in emergency retention zones, which make it difficult to detect cardiac arrest (CA). OBJECTIVE We seek to develop a pharmaceutical early warning model to predict cardiac arrest in emergency departments via drug classification and medical expert suggestion. METHODS We propose a new early warning score model for detecting cardiac arrest via pharmaceutical classification and by using a sliding window; we apply learning-based algorithms to time-series data for a Pharmaceutical Early Warning Scoring Model (PEWSM). By treating pharmaceutical features as a dynamic time-series factor for cardiopulmonary resuscitation (CPR) patients, we increase sensitivity, reduce false alarm rates and mortality, and increase the model’s accuracy. To evaluate the proposed model we use the area under the receiver operating characteristic curve (AUROC). RESULTS Four important findings are as follows: (1) We identify the most important drug predictors: bits, and replenishers and regulators of water and electrolytes. The best AUROC of bits is 85%; that of replenishers and regulators of water and electrolytes is 86%. These two features are the most influential of the drug features in the task. (2) We verify feature selection, in which accounting for drugs improve the accuracy: In Task 1, the best AUROC of vital signs is 77%, and that of all features is 86%. In Task 2, the best AUROC of all features is 85%, which demonstrates that thus accounting for the drugs significantly affects prediction. (3) We use a better model: For traditional machine learning, this study adds a new AI technology: the long short-term memory (LSTM) model with the best time-series accuracy, comparable to the traditional random forest (RF) model; the two AUROC measures are 85%. (4) We determine whether the event can be predicted beforehand: The best classifier is still an RF model, in which the observational starting time is 4 hours before the CPR event. Although the accuracy is impaired, the predictive accuracy still reaches 70%. Therefore, we believe that CPR events can be predicted four hours before the event. CONCLUSIONS This paper uses a sliding window to account for dynamic time-series data consisting of the patient’s vital signs and drug injections. In a comparison with NEWS, we improve predictive accuracy via feature selection, which includes drugs as features. In addition, LSTM yields better performance with time-series data. The proposed PEWSM, which offers 4-hour predictions, is better than the National Early Warning Score (NEWS) in the literature. This also confirms that the doctor’s heuristic rules are consistent with the results found by machine learning algorithms.

Development of A Drug Early Warning System Model for Cardiac Arrest Using Deep Learning: Retrospective Cohort Study (Preprint)

2020 ◽  
Author(s):  
Hsiao-Ko Chang ◽  
Hui-Chih Wang ◽  
Chih-Fen Huang ◽  
Feipei Lai
Keyword(s):  
Machine Learning ◽  
Time Series ◽  
Cardiac Arrest ◽  
Early Warning ◽  
Time Series Data ◽  
Predictive Accuracy ◽  
Vital Signs ◽  
Warning System ◽  
Series Data ◽  
Dynamic Time

BACKGROUND In most of Taiwan’s medical institutions, congestion is a serious problem for emergency departments. Due to a lack of beds, patients spend more time in emergency retention zones, which make it difficult to detect cardiac arrest (CA). OBJECTIVE We seek to develop a Drug Early Warning System Model (DEWSM), it included drug injections and vital signs as this research important features. We use it to predict cardiac arrest in emergency departments via drug classification and medical expert suggestion. METHODS We propose this new model for detecting cardiac arrest via drug classification and by using a sliding window; we apply learning-based algorithms to time-series data for a DEWSM. By treating drug features as a dynamic time-series factor for cardiopulmonary resuscitation (CPR) patients, we increase sensitivity, reduce false alarm rates and mortality, and increase the model’s accuracy. To evaluate the proposed model, we use the area under the receiver operating characteristic curve (AUROC). RESULTS Four important findings are as follows: (1) We identify the most important drug predictors: bits (intravenous therapy), and replenishers and regulators of water and electrolytes (fluid and electrolyte supplement). The best AUROC of bits is 85%, it means the medical expert suggest the drug features: bits, it will affect the vital signs, and then the evaluate this model correctly classified patients with CPR reach 85%; that of replenishers and regulators of water and electrolytes is 86%. These two features are the most influential of the drug features in the task. (2) We verify feature selection, in which accounting for drugs improve the accuracy: In Task 1, the best AUROC of vital signs is 77%, and that of all features is 86%. In Task 2, the best AUROC of all features is 85%, which demonstrates that thus accounting for the drugs significantly affects prediction. (3) We use a better model: For traditional machine learning, this study adds a new AI technology: the long short-term memory (LSTM) model with the best time-series accuracy, comparable to the traditional random forest (RF) model; the two AUROC measures are 85%. It can be seen that the use of new AI technology will achieve better results, currently comparable to the accuracy of traditional common RF, and the LSTM model can be adjusted in the future to obtain better results. (4) We determine whether the event can be predicted beforehand: The best classifier is still an RF model, in which the observational starting time is 4 hours before the CPR event. Although the accuracy is impaired, the predictive accuracy still reaches 70%. Therefore, we believe that CPR events can be predicted four hours before the event. CONCLUSIONS This paper uses a sliding window to account for dynamic time-series data consisting of the patient’s vital signs and drug injections. The National Early Warning Score (NEWS) only focuses on the score of vital signs, and does not include factors related to drug injections. In this study, the experimental results of adding the drug injections are better than only vital signs. In a comparison with NEWS, we improve predictive accuracy via feature selection, which includes drugs as features. In addition, we use traditional machine learning methods and deep learning (using LSTM method as the main processing time series data) as the basis for comparison of this research. The proposed DEWSM, which offers 4-hour predictions, is better than the NEWS in the literature. This also confirms that the doctor’s heuristic rules are consistent with the results found by machine learning algorithms.

scholarly journals Minimum time required to detect population trends: the need for long-term monitoring programs

2017 ◽  
Author(s):  
Easton R White
Keyword(s):  
Time Series ◽  
Statistical Power ◽  
Time Series Data ◽  
Conservation Status ◽  
Monitoring Program ◽  
Minimum Time ◽  
Series Data ◽  
Series Length ◽  
Time Required

Long-term time series are necessary to better understand population dynamics, assess species' conservation status, and make management decisions. However, population data are often expensive, requiring a lot of time and resources. When is a population time series long enough to address a question of interest? We determine the minimum time series length required to detect significant increases or decreases in population abundance. To address this question, we use simulation methods and examine 878 populations of vertebrate species. Here we show that 15-20 years of continuous monitoring are required in order to achieve a high level of statistical power. For both simulations and the time series data, the minimum time required depends on trend strength, population variability, and temporal autocorrelation. These results point to the importance of sampling populations over long periods of time. We argue that statistical power needs to be considered in monitoring program design and evaluation. Time series less than 15-20 years are likely underpowered and potentially misleading.

Sign in / Sign up

Export Citation Format

Share Document