Dataset Information

Extended approach to sum of absolute differences method for improved identification of periods in biomedical time series.

ABSTRACT: Time series are a common data type in biomedical applications. Examples include heart rate, power output, and ECG. One of the typical analysis methods is to determine longest period a subject spent over a given heart rate threshold. While it might seem simple to find and measure such periods, biomedical data are often subject to significant noise and physiological artifacts. As a result, simple threshold calculations might not provide correct or expected results. A common way to improve such calculations is to use moving average filter. Length of the window is often determined using sum of absolute differences for various windows sizes. However, for real life biomedical data such approach might lead to extremely long windows that undesirably remove physiological information from the data. In this paper, we:•propose a new approach to finding windows length using zero-points of third gradient (jerk) of Sum of Absolute Differences method;•demonstrate how these points can be used to determine periods and area over a given threshold with and without uncertainty.We demonstrate validity of this approach on the PAMAP2 Physical Activity Monitoring Data Set, an open dataset from the UCI Machine Learning Repository, as well as on the PhysioNet Simultaneous Physiological Measurements dataset. It shows that first zero-point usually falls at around 8 and 5 second window length respectively, while second zero-point usually falls between 16 and 24 and 8-16 s respectively. The value for the first zero-point can remove simple measurement errors when data are recorded once every few seconds. The value for the second zero-point corresponds well with what is known about physiological response of heart to changing load.

SUBMITTER: Wiktorski T

PROVIDER: S-EPMC7575807 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Extended approach to sum of absolute differences method for improved identification of periods in biomedical time series.

Wiktorski Tomasz T Królak Aleksandra A

MethodsX 20201009

Time series are a common data type in biomedical applications. Examples include heart rate, power output, and ECG. One of the typical analysis methods is to determine longest period a subject spent over a given heart rate threshold. While it might seem simple to find and measure such periods, biomedical data are often subject to significant noise and physiological artifacts. As a result, simple threshold calculations might not provide correct or expected results. A common way to improve such cal ...[more]

PMID: 33102157

Similar Datasets

Project description:MotivationMost modern intensive care units record the physiological and vital signs of patients. These data can be used to extract signatures, commonly known as biomarkers, that help physicians understand the biological complexity of many syndromes. However, most biological biomarkers suffer from either poor predictive performance or weak explanatory power. Recent developments in time series classification focus on discovering shapelets, i.e. subsequences that are most predictive in terms of class membership. Shapelets have the advantage of combining a high predictive performance with an interpretable component-their shape. Currently, most shapelet discovery methods do not rely on statistical tests to verify the significance of individual shapelets. Therefore, identifying associations between the shapelets of physiological biomarkers and patients that exhibit certain phenotypes of interest enables the discovery and subsequent ranking of physiological signatures that are interpretable, statistically validated and accurate predictors of clinical endpoints.ResultsWe present a novel and scalable method for scanning time series and identifying discriminative patterns that are statistically significant. The significance of a shapelet is evaluated while considering the problem of multiple hypothesis testing and mitigating it by efficiently pruning untestable shapelet candidates with Tarone's method. We demonstrate the utility of our method by discovering patterns in three of a patient's vital signs: heart rate, respiratory rate and systolic blood pressure that are indicators of the severity of a future sepsis event, i.e. an inflammatory response to an infective agent that can lead to organ failure and death, if not treated in time.Availability and implementationWe make our method and the scripts that are required to reproduce the experiments publicly available at https://github.com/BorgwardtLab/S3M.Supplementary informationSupplementary data are available at Bioinformatics online.

Project description:Background: Digital clinical measures collected via various digital sensing technologies such as smartphones, smartwatches, wearables, and ingestible and implantable sensors are increasingly used by individuals and clinicians to capture the health outcomes or behavioral and physiological characteristics of individuals. Time series classification (TSC) is very commonly used for modeling digital clinical measures. While deep learning models for TSC are very common and powerful, there exist some fundamental challenges. This review presents the non-deep learning models that are commonly used for time series classification in biomedical applications that can achieve high performance. Objective: We performed a systematic review to characterize the techniques that are used in time series classification of digital clinical measures throughout all the stages of data processing and model building. Methods: We conducted a literature search on PubMed, as well as the Institute of Electrical and Electronics Engineers (IEEE), Web of Science, and SCOPUS databases using a range of search terms to retrieve peer-reviewed articles that report on the academic research about digital clinical measures from a five-year period between June 2016 and June 2021. We identified and categorized the research studies based on the types of classification algorithms and sensor input types. Results: We found 452 papers in total from four different databases: PubMed, IEEE, Web of Science Database, and SCOPUS. After removing duplicates and irrelevant papers, 135 articles remained for detailed review and data extraction. Among these, engineered features using time series methods that were subsequently fed into widely used machine learning classifiers were the most commonly used technique, and also most frequently achieved the best performance metrics (77 out of 135 articles). Statistical modeling (24 out of 135 articles) algorithms were the second most common and also the second-best classification technique. Conclusions: In this review paper, summaries of the time series classification models and interpretation methods for biomedical applications are summarized and categorized. While high time series classification performance has been achieved in digital clinical, physiological, or biomedical measures, no standard benchmark datasets, modeling methods, or reporting methodology exist. There is no single widely used method for time series model development or feature interpretation, however many different methods have proven successful.

Project description:BackgroundMore intense tropical cyclones (TCs) are expected in the future under a warming climate scenario, but little is known about their mortality effect pattern across countries and over decades. We aim to evaluate the TC-specific mortality risks, periods of concern (POC) and characterize the spatiotemporal pattern and exposure-response (ER) relationships on a multicountry scale.Methods and findingsDaily all-cause, cardiovascular, and respiratory mortality among the general population were collected from 494 locations in 18 countries or territories during 1980 to 2019. Daily TC exposures were defined when the maximum sustained windspeed associated with a TC was ≥34 knots using a parametric wind field model at a 0.5° × 0.5° resolution. We first estimated the TC-specific mortality risks and POC using an advanced flexible statistical framework of mixed Poisson model, accounting for the population changes, natural variation, seasonal and day of the week effects. Then, a mixed meta-regression model was used to pool the TC-specific mortality risks to estimate the overall and country-specific ER relationships of TC characteristics (windspeed, rainfall, and year) with mortality. Overall, 47.7 million all-cause, 15.5 million cardiovascular, and 4.9 million respiratory deaths and 382 TCs were included in our analyses. An overall average POC of around 20 days was observed for TC-related all-cause and cardiopulmonary mortality, with relatively longer POC for the United States of America, Brazil, and Taiwan (>30 days). The TC-specific relative risks (RR) varied substantially, ranging from 1.04 to 1.42, 1.07 to 1.77, and 1.12 to 1.92 among the top 100 TCs with highest RRs for all-cause, cardiovascular, and respiratory mortality, respectively. At country level, relatively higher TC-related mortality risks were observed in Guatemala, Brazil, and New Zealand for all-cause, cardiovascular, and respiratory mortality, respectively. We found an overall monotonically increasing and approximately linear ER curve of TC-related maximum sustained windspeed and cumulative rainfall with mortality, with heterogeneous patterns across countries and regions. The TC-related mortality risks were generally decreasing from 1980 to 2019, especially for the Philippines, Taiwan, and the USA, whereas potentially increasing trends in TC-related all-cause and cardiovascular mortality risks were observed for Japan.ConclusionsThe TC mortality risks and POC varied greatly across TC events, locations, and countries. To minimize the TC-related health burdens, targeted strategies are particularly needed for different countries and regions, integrating epidemiological evidence on region-specific POC and ER curves that consider across-TC variability.

Dataset Information

Extended approach to sum of absolute differences method for improved identification of periods in biomedical time series.

Publications

Extended approach to sum of absolute differences method for improved identification of periods in biomedical time series.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets