Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series

Ekaterina Zinkovskaia; Orel Tahary; Izhar Bar-Gad

doi:10.1109/ieeeconf58974.2023.10405072

Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series

Ekaterina Zinkovskaia
, Orel Tahary
, Izhar Bar-Gad

The Gonda Multidisciplinary Brain Research Center

Bar-Ilan University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Behavior consists of a series of repeating yet variable discrete motifs across various timescales. We introduce a framework for temporally aligned segmentation and clustering (TASC) of behavioral time series. TASC is designed to extract such motif recurrences in high temporal resolution. This framework operates iteratively in two steps: (1) embedding of time series segments and calculation of linearly aligned distances within the clustered space, and (2) recalculating of the clustered space after alignment. We evaluated TASC on a semi-synthetic experimental and a clinical dataset, and it demonstrated enhanced segmentation performance. TASC may be applied to other domains where analysis of recurring time series patterns with high temporal precision is needed.Clinical Relevance: This framework enables identifying the temporal structure of discrete pathological behaviors, such as tics properties in Tourette's syndrome.

Original language	English
Title of host publication	2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	173-174
Number of pages	2
ISBN (Electronic)	9798350383386
DOIs	https://doi.org/10.1109/ieeeconf58974.2023.10405072
State	Published - 2023
Event	2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023 - Malta, Malta Duration: 7 Dec 2023 → 9 Dec 2023

Publication series

Name	2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023

Conference

Conference	2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023
Country/Territory	Malta
City	Malta
Period	7/12/23 → 9/12/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

ASJC Scopus subject areas

Artificial Intelligence
Human-Computer Interaction
Information Systems
Decision Sciences (miscellaneous)
Information Systems and Management
Biomedical Engineering
Health Informatics

Access to Document

10.1109/ieeeconf58974.2023.10405072

Cite this

Zinkovskaia, E., Tahary, O., & Bar-Gad, I. (2023). Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series. In 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023 (pp. 173-174). (2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ieeeconf58974.2023.10405072

Zinkovskaia, Ekaterina ; Tahary, Orel ; Bar-Gad, Izhar. / Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series. 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 173-174 (2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023).

@inproceedings{efdfafaf7c164077917e4466c43cd66a,

title = "Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series",

abstract = "Behavior consists of a series of repeating yet variable discrete motifs across various timescales. We introduce a framework for temporally aligned segmentation and clustering (TASC) of behavioral time series. TASC is designed to extract such motif recurrences in high temporal resolution. This framework operates iteratively in two steps: (1) embedding of time series segments and calculation of linearly aligned distances within the clustered space, and (2) recalculating of the clustered space after alignment. We evaluated TASC on a semi-synthetic experimental and a clinical dataset, and it demonstrated enhanced segmentation performance. TASC may be applied to other domains where analysis of recurring time series patterns with high temporal precision is needed.Clinical Relevance: This framework enables identifying the temporal structure of discrete pathological behaviors, such as tics properties in Tourette's syndrome.",

author = "Ekaterina Zinkovskaia and Orel Tahary and Izhar Bar-Gad",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023 ; Conference date: 07-12-2023 Through 09-12-2023",

year = "2023",

doi = "10.1109/ieeeconf58974.2023.10405072",

language = "English",

series = "2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "173--174",

booktitle = "2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023",

address = "United States",

}

Zinkovskaia, E, Tahary, O & Bar-Gad, I 2023, Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series. in 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023. 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023, Institute of Electrical and Electronics Engineers Inc., pp. 173-174, 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023, Malta, Malta, 7/12/23. https://doi.org/10.1109/ieeeconf58974.2023.10405072

Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series. / Zinkovskaia, Ekaterina; Tahary, Orel; Bar-Gad, Izhar.
2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 173-174 (2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series

AU - Zinkovskaia, Ekaterina

AU - Tahary, Orel

AU - Bar-Gad, Izhar

PY - 2023

Y1 - 2023

N2 - Behavior consists of a series of repeating yet variable discrete motifs across various timescales. We introduce a framework for temporally aligned segmentation and clustering (TASC) of behavioral time series. TASC is designed to extract such motif recurrences in high temporal resolution. This framework operates iteratively in two steps: (1) embedding of time series segments and calculation of linearly aligned distances within the clustered space, and (2) recalculating of the clustered space after alignment. We evaluated TASC on a semi-synthetic experimental and a clinical dataset, and it demonstrated enhanced segmentation performance. TASC may be applied to other domains where analysis of recurring time series patterns with high temporal precision is needed.Clinical Relevance: This framework enables identifying the temporal structure of discrete pathological behaviors, such as tics properties in Tourette's syndrome.

AB - Behavior consists of a series of repeating yet variable discrete motifs across various timescales. We introduce a framework for temporally aligned segmentation and clustering (TASC) of behavioral time series. TASC is designed to extract such motif recurrences in high temporal resolution. This framework operates iteratively in two steps: (1) embedding of time series segments and calculation of linearly aligned distances within the clustered space, and (2) recalculating of the clustered space after alignment. We evaluated TASC on a semi-synthetic experimental and a clinical dataset, and it demonstrated enhanced segmentation performance. TASC may be applied to other domains where analysis of recurring time series patterns with high temporal precision is needed.Clinical Relevance: This framework enables identifying the temporal structure of discrete pathological behaviors, such as tics properties in Tourette's syndrome.

UR - https://www.scopus.com/pages/publications/85185559570

U2 - 10.1109/ieeeconf58974.2023.10405072

DO - 10.1109/ieeeconf58974.2023.10405072

M3 - Conference contribution

T3 - 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023

SP - 173

EP - 174

BT - 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023

Y2 - 7 December 2023 through 9 December 2023

ER -

Zinkovskaia E, Tahary O, Bar-Gad I. Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series. In 2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 173-174. (2023 IEEE EMBS Special Topic Conference on Data Science and Engineering in Healthcare, Medicine and Biology, IEEECONF 2023). doi: 10.1109/ieeeconf58974.2023.10405072

Temporally Aligned Segmentation and Clustering (TASC) of Behavioral Time Series

Abstract

Publication series

Conference

UN SDGs

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this