Temporal patterns in insulin needs for Type 1 diabetes

Isabella Degen; Zahraa S. Abdallah

doi:10.48550/arxiv.2211.07393

Temporal patterns in insulin needs for Type 1 diabetes

Research output: Working paper › Preprint

76 Downloads (Pure)

Abstract

Type 1 Diabetes (T1D) is a chronic condition where the body produces little or no insulin, a hormone required for the cells to use blood glucose (BG) for energy and to regulate BG levels in the body. Finding the right insulin dose and time remains a complex, challenging and as yet unsolved control task. In this study, we use the OpenAPS Data Commons dataset, which is an extensive dataset collected in real-life conditions, to discover temporal patterns in insulin need driven by well-known factors such as carbohydrates as well as potentially novel factors. We utilised various time series techniques to spot such patterns using matrix profile and multi-variate clustering. The better we understand T1D and the factors impacting insulin needs, the more we can contribute to building data-driven technology for T1D treatments.

Original language	English
Publisher	NeurIPS 2022 Workshop on Learning from Time Series for Health
DOIs	https://doi.org/10.48550/arxiv.2211.07393
Publication status	Published - 14 Nov 2022

Bibliographical note

Submitted and accepted for presentation as a poster at the NeurIPS22 Time series for Health workshop, https://timeseriesforhealth.github.io/

UN SDGs

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

SDG 3 Good Health and Well-being

Research Groups and Themes

Interactive Artificial Intelligence CDT

Keywords

cs.LG
q-bio.QM
stat.ML

Access to Document

10.48550/arxiv.2211.07393

Full-text PDF (pre-print)
This is a pre-print version of the article (version of record). It first appeared online via arXiv at https://doi.org/10.48550/arXiv.2211.07393. Please refer to any applicable terms of use of the publisher.
Submitted manuscript, 5.35 MB

Handle.net

Persistent link

Towards Objective Time Series Clustering Validation
Degen, I. E. (Author), Abdallah, Z. S. (Supervisor), Reeve, H. W. J. (Supervisor) & Brown, K. R. (Supervisor), 20 Jan 2026
Student thesis: Doctoral Thesis › Doctor of Philosophy (PhD)

File

Cite this

@techreport{4e02dcfc92cf405c98c2873303740832,

title = "Temporal patterns in insulin needs for Type 1 diabetes",

abstract = "Type 1 Diabetes (T1D) is a chronic condition where the body produces little or no insulin, a hormone required for the cells to use blood glucose (BG) for energy and to regulate BG levels in the body. Finding the right insulin dose and time remains a complex, challenging and as yet unsolved control task. In this study, we use the OpenAPS Data Commons dataset, which is an extensive dataset collected in real-life conditions, to discover temporal patterns in insulin need driven by well-known factors such as carbohydrates as well as potentially novel factors. We utilised various time series techniques to spot such patterns using matrix profile and multi-variate clustering. The better we understand T1D and the factors impacting insulin needs, the more we can contribute to building data-driven technology for T1D treatments.",

keywords = "cs.LG, q-bio.QM, stat.ML",

author = "Isabella Degen and Abdallah, \{Zahraa S.\}",

note = "Submitted and accepted for presentation as a poster at the NeurIPS22 Time series for Health workshop, https://timeseriesforhealth.github.io/",

year = "2022",

month = nov,

day = "14",

doi = "10.48550/arxiv.2211.07393",

language = "English",

publisher = "NeurIPS 2022 Workshop on Learning from Time Series for Health",

type = "WorkingPaper",

institution = "NeurIPS 2022 Workshop on Learning from Time Series for Health",

}

TY - UNPB

T1 - Temporal patterns in insulin needs for Type 1 diabetes

AU - Degen, Isabella

AU - Abdallah, Zahraa S.

N1 - Submitted and accepted for presentation as a poster at the NeurIPS22 Time series for Health workshop, https://timeseriesforhealth.github.io/

PY - 2022/11/14

Y1 - 2022/11/14

N2 - Type 1 Diabetes (T1D) is a chronic condition where the body produces little or no insulin, a hormone required for the cells to use blood glucose (BG) for energy and to regulate BG levels in the body. Finding the right insulin dose and time remains a complex, challenging and as yet unsolved control task. In this study, we use the OpenAPS Data Commons dataset, which is an extensive dataset collected in real-life conditions, to discover temporal patterns in insulin need driven by well-known factors such as carbohydrates as well as potentially novel factors. We utilised various time series techniques to spot such patterns using matrix profile and multi-variate clustering. The better we understand T1D and the factors impacting insulin needs, the more we can contribute to building data-driven technology for T1D treatments.

AB - Type 1 Diabetes (T1D) is a chronic condition where the body produces little or no insulin, a hormone required for the cells to use blood glucose (BG) for energy and to regulate BG levels in the body. Finding the right insulin dose and time remains a complex, challenging and as yet unsolved control task. In this study, we use the OpenAPS Data Commons dataset, which is an extensive dataset collected in real-life conditions, to discover temporal patterns in insulin need driven by well-known factors such as carbohydrates as well as potentially novel factors. We utilised various time series techniques to spot such patterns using matrix profile and multi-variate clustering. The better we understand T1D and the factors impacting insulin needs, the more we can contribute to building data-driven technology for T1D treatments.

KW - cs.LG

KW - q-bio.QM

KW - stat.ML

U2 - 10.48550/arxiv.2211.07393

DO - 10.48550/arxiv.2211.07393

M3 - Preprint

BT - Temporal patterns in insulin needs for Type 1 diabetes

PB - NeurIPS 2022 Workshop on Learning from Time Series for Health

ER -

Temporal patterns in insulin needs for Type 1 diabetes

Abstract

Bibliographical note

UN SDGs

Research Groups and Themes

Keywords

Access to Document

Handle.net

Fingerprint

Student theses

Towards Objective Time Series Clustering Validation

Cite this