Inferring temporal phenotypes with topological data analysis and pseudo time-series

Arianna Dagliati; Nophar Geifman; Niels Peek; John H. Holmes; Lucia Sacchi; Seyed Erfan Sajjadi; Allan Tucker

doi:10.1007/978-3-030-21642-9_50

Inferring temporal phenotypes with topological data analysis and pseudo time-series

Arianna Dagliati^*, Nophar Geifman, Niels Peek, John H. Holmes, Lucia Sacchi, Seyed Erfan Sajjadi, Allan Tucker

^*Corresponding author for this work

Division of Informatics, Imaging & Data Sciences (L5)

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

Abstract

Temporal phenotyping enables clinicians to better under-stand observable characteristics of a disease as it progresses. Modelling disease progression that captures interactions between phenotypes is inherently challenging. Temporal models that capture change in disease over time can identify the key features that characterize disease subtypes that underpin these trajectories. These models will enable clinicians to identify early warning signs of progression in specific sub-types and therefore to make informed decisions tailored to individual patients. In this paper, we explore two approaches to building temporal phenotypes based on the topology of data: topological data analysis and pseudo time-series. Using type 2 diabetes data, we show that the topological data analysis approach is able to identify trajectories representing different temporal phenotypes and that pseudo time-series can infer a state space model characterized by transitions between hidden states that represent distinct temporal phenotypes. Both approaches highlight lipid profiles as key factors in distinguishing the phenotypes.

Original language	English
Title of host publication	Artificial Intelligence in Medicine - 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Proceedings
Editors	Szymon Wilk, Annette ten Teije, David Riaño
Publisher	Springer Nature
Pages	399-409
Number of pages	11
ISBN (Print)	9783030216412
DOIs	https://doi.org/10.1007/978-3-030-21642-9_50
Publication status	Published - 2019
Event	17th Conference on Artificial Intelligence in Medicine, AIME 2019 - Poznan, Poland Duration: 26 Jun 2019 → 29 Jun 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11526 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	17th Conference on Artificial Intelligence in Medicine, AIME 2019
Country/Territory	Poland
City	Poznan
Period	26/06/19 → 29/06/19

Keywords

Electronic phenotyping
Longitudinal studies
Type 2 diabetes
Unsupervised machine learning

UN SDGs

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

Access to Document

10.1007/978-3-030-21642-9_50

Cite this

Dagliati, A., Geifman, N., Peek, N., Holmes, J. H., Sacchi, L., Sajjadi, S. E., & Tucker, A. (2019). Inferring temporal phenotypes with topological data analysis and pseudo time-series. In S. Wilk, A. ten Teije, & D. Riaño (Eds.), Artificial Intelligence in Medicine - 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Proceedings (pp. 399-409). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11526 LNAI). Springer Nature. https://doi.org/10.1007/978-3-030-21642-9_50

Dagliati, Arianna ; Geifman, Nophar ; Peek, Niels et al. / Inferring temporal phenotypes with topological data analysis and pseudo time-series. Artificial Intelligence in Medicine - 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Proceedings. editor / Szymon Wilk ; Annette ten Teije ; David Riaño. Springer Nature, 2019. pp. 399-409 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{cd9cc3732f734076a22042bc4f2fa7a8,

title = "Inferring temporal phenotypes with topological data analysis and pseudo time-series",

abstract = "Temporal phenotyping enables clinicians to better under-stand observable characteristics of a disease as it progresses. Modelling disease progression that captures interactions between phenotypes is inherently challenging. Temporal models that capture change in disease over time can identify the key features that characterize disease subtypes that underpin these trajectories. These models will enable clinicians to identify early warning signs of progression in specific sub-types and therefore to make informed decisions tailored to individual patients. In this paper, we explore two approaches to building temporal phenotypes based on the topology of data: topological data analysis and pseudo time-series. Using type 2 diabetes data, we show that the topological data analysis approach is able to identify trajectories representing different temporal phenotypes and that pseudo time-series can infer a state space model characterized by transitions between hidden states that represent distinct temporal phenotypes. Both approaches highlight lipid profiles as key factors in distinguishing the phenotypes.",

keywords = "Electronic phenotyping, Longitudinal studies, Type 2 diabetes, Unsupervised machine learning",

author = "Arianna Dagliati and Nophar Geifman and Niels Peek and Holmes, {John H.} and Lucia Sacchi and Sajjadi, {Seyed Erfan} and Allan Tucker",

year = "2019",

doi = "10.1007/978-3-030-21642-9_50",

language = "English",

isbn = "9783030216412",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Nature",

pages = "399--409",

editor = "Szymon Wilk and {ten Teije}, Annette and David Ria{\~n}o",

booktitle = "Artificial Intelligence in Medicine - 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Proceedings",

address = "United States",

note = "17th Conference on Artificial Intelligence in Medicine, AIME 2019 ; Conference date: 26-06-2019 Through 29-06-2019",

}

Dagliati, A, Geifman, N, Peek, N, Holmes, JH, Sacchi, L, Sajjadi, SE & Tucker, A 2019, Inferring temporal phenotypes with topological data analysis and pseudo time-series. in S Wilk, A ten Teije & D Riaño (eds), Artificial Intelligence in Medicine - 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11526 LNAI, Springer Nature, pp. 399-409, 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Poznan, Poland, 26/06/19. https://doi.org/10.1007/978-3-030-21642-9_50

Inferring temporal phenotypes with topological data analysis and pseudo time-series. / Dagliati, Arianna; Geifman, Nophar; Peek, Niels et al.
Artificial Intelligence in Medicine - 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Proceedings. ed. / Szymon Wilk; Annette ten Teije; David Riaño. Springer Nature, 2019. p. 399-409 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11526 LNAI).

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

TY - GEN

T1 - Inferring temporal phenotypes with topological data analysis and pseudo time-series

AU - Dagliati, Arianna

AU - Geifman, Nophar

AU - Peek, Niels

AU - Holmes, John H.

AU - Sacchi, Lucia

AU - Sajjadi, Seyed Erfan

AU - Tucker, Allan

PY - 2019

Y1 - 2019

N2 - Temporal phenotyping enables clinicians to better under-stand observable characteristics of a disease as it progresses. Modelling disease progression that captures interactions between phenotypes is inherently challenging. Temporal models that capture change in disease over time can identify the key features that characterize disease subtypes that underpin these trajectories. These models will enable clinicians to identify early warning signs of progression in specific sub-types and therefore to make informed decisions tailored to individual patients. In this paper, we explore two approaches to building temporal phenotypes based on the topology of data: topological data analysis and pseudo time-series. Using type 2 diabetes data, we show that the topological data analysis approach is able to identify trajectories representing different temporal phenotypes and that pseudo time-series can infer a state space model characterized by transitions between hidden states that represent distinct temporal phenotypes. Both approaches highlight lipid profiles as key factors in distinguishing the phenotypes.

AB - Temporal phenotyping enables clinicians to better under-stand observable characteristics of a disease as it progresses. Modelling disease progression that captures interactions between phenotypes is inherently challenging. Temporal models that capture change in disease over time can identify the key features that characterize disease subtypes that underpin these trajectories. These models will enable clinicians to identify early warning signs of progression in specific sub-types and therefore to make informed decisions tailored to individual patients. In this paper, we explore two approaches to building temporal phenotypes based on the topology of data: topological data analysis and pseudo time-series. Using type 2 diabetes data, we show that the topological data analysis approach is able to identify trajectories representing different temporal phenotypes and that pseudo time-series can infer a state space model characterized by transitions between hidden states that represent distinct temporal phenotypes. Both approaches highlight lipid profiles as key factors in distinguishing the phenotypes.

KW - Electronic phenotyping

KW - Longitudinal studies

KW - Type 2 diabetes

KW - Unsupervised machine learning

UR - http://www.scopus.com/inward/record.url?scp=85068347062&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-21642-9_50

DO - 10.1007/978-3-030-21642-9_50

M3 - Conference contribution

AN - SCOPUS:85068347062

SN - 9783030216412

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 399

EP - 409

BT - Artificial Intelligence in Medicine - 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Proceedings

A2 - Wilk, Szymon

A2 - ten Teije, Annette

A2 - Riaño, David

PB - Springer Nature

T2 - 17th Conference on Artificial Intelligence in Medicine, AIME 2019

Y2 - 26 June 2019 through 29 June 2019

ER -

Dagliati A, Geifman N, Peek N, Holmes JH, Sacchi L, Sajjadi SE et al. Inferring temporal phenotypes with topological data analysis and pseudo time-series. In Wilk S, ten Teije A, Riaño D, editors, Artificial Intelligence in Medicine - 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Proceedings. Springer Nature. 2019. p. 399-409. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Epub 2019 May 30. doi: 10.1007/978-3-030-21642-9_50

Inferring temporal phenotypes with topological data analysis and pseudo time-series

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this