Selecting Lung Cancer Patients from UK Primary Care Data: A Longitudinal Study of Feature Trends

Abeer Alzubaidi; Jaspreet Kaur; Mufti Mahmud; David J. Brown; Jun He; Graham Ball; David R. Baldwin; Emma O’Dowd; Richard B. Hubbard

doi:10.1007/978-3-030-82269-9_4

Selecting Lung Cancer Patients from UK Primary Care Data: A Longitudinal Study of Feature Trends

Abeer Alzubaidi
, Jaspreet Kaur
, Mufti Mahmud^*
, David J. Brown
, Jun He
, Graham Ball
, David R. Baldwin
, Emma O’Dowd
, Richard B. Hubbard

^*Corresponding author for this work

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

Abstract

A high proportion of lung cancer cases are detected at a late cancer stage when they present with symptoms to general practitioners (GP). Early diagnosis is a challenge because many symptoms are also common in other diseases. Therefore, this study aims to assess UK primary care data of patients one, two and three years prior to lung cancer diagnosis to capture trends in clinical features of patients with the goal of early diagnosis and thus potentially curative treatment. This longitudinal study utilises data from the Clinical Practice Research Datalink (CPRD) with linked data from the National Cancer Registration and Analysis Service (NCRAS). A comprehensive list of Read codes is created to select features of interest to establish if a patient has experienced a certain medical condition or not. The comparison of the relative frequencies of the identified predictors associated with cases and controls reveals the importance of the following groups of features: ‘Cough Wheeze’ and ‘Bronchitis unspecified’, ‘Dyspnoea’ and ‘Upper Respiratory Infection’, which are frequent events for lung cancer cases, where a high proportion of cases were also identified using ‘Haemoptysis’ and ‘Peripheral vascular disease’.

Original language	English
Title of host publication	Applied Intelligence and Informatics - 1st International Conference, AII 2021, Proceedings
Editors	Mufti Mahmud, M. Shamim Kaiser, Nikola Kasabov, Khan Iftekharuddin, Ning Zhong
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	43-59
Number of pages	17
ISBN (Print)	9783030822682
DOIs	https://doi.org/10.1007/978-3-030-82269-9_4
State	Published - 2021
Externally published	Yes
Event	1st International Conference on Applied Intelligence and Informatics, AII 2021 - Virtual, Online Duration: 30 Jul 2021 → 31 Jul 2021

Publication series

Name	Communications in Computer and Information Science
Volume	1435
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	1st International Conference on Applied Intelligence and Informatics, AII 2021
City	Virtual, Online
Period	30/07/21 → 31/07/21

Bibliographical note

Publisher Copyright:
© 2021, Springer Nature Switzerland AG.

UN SDGs

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

SDG 3 Good Health and Well-being

Keywords

Bronchitis unspecified
Cough
Dyspnoea
Lung cancer
Machine learning
Upper respiratory infection

ASJC Scopus subject areas

General Computer Science
General Mathematics

Access to Document

10.1007/978-3-030-82269-9_4

Cite this

Alzubaidi, A., Kaur, J., Mahmud, M., Brown, D. J., He, J., Ball, G., Baldwin, D. R., O’Dowd, E., & Hubbard, R. B. (2021). Selecting Lung Cancer Patients from UK Primary Care Data: A Longitudinal Study of Feature Trends. In M. Mahmud, M. S. Kaiser, N. Kasabov, K. Iftekharuddin, & N. Zhong (Eds.), Applied Intelligence and Informatics - 1st International Conference, AII 2021, Proceedings (pp. 43-59). (Communications in Computer and Information Science; Vol. 1435). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-82269-9_4

Alzubaidi, Abeer ; Kaur, Jaspreet ; Mahmud, Mufti et al. / Selecting Lung Cancer Patients from UK Primary Care Data : A Longitudinal Study of Feature Trends. Applied Intelligence and Informatics - 1st International Conference, AII 2021, Proceedings. editor / Mufti Mahmud ; M. Shamim Kaiser ; Nikola Kasabov ; Khan Iftekharuddin ; Ning Zhong. Springer Science and Business Media Deutschland GmbH, 2021. pp. 43-59 (Communications in Computer and Information Science).

@inproceedings{f36c9b640fba42f09e6f647346fccec0,

title = "Selecting Lung Cancer Patients from UK Primary Care Data: A Longitudinal Study of Feature Trends",

abstract = "A high proportion of lung cancer cases are detected at a late cancer stage when they present with symptoms to general practitioners (GP). Early diagnosis is a challenge because many symptoms are also common in other diseases. Therefore, this study aims to assess UK primary care data of patients one, two and three years prior to lung cancer diagnosis to capture trends in clinical features of patients with the goal of early diagnosis and thus potentially curative treatment. This longitudinal study utilises data from the Clinical Practice Research Datalink (CPRD) with linked data from the National Cancer Registration and Analysis Service (NCRAS). A comprehensive list of Read codes is created to select features of interest to establish if a patient has experienced a certain medical condition or not. The comparison of the relative frequencies of the identified predictors associated with cases and controls reveals the importance of the following groups of features: {\textquoteleft}Cough Wheeze{\textquoteright} and {\textquoteleft}Bronchitis unspecified{\textquoteright}, {\textquoteleft}Dyspnoea{\textquoteright} and {\textquoteleft}Upper Respiratory Infection{\textquoteright}, which are frequent events for lung cancer cases, where a high proportion of cases were also identified using {\textquoteleft}Haemoptysis{\textquoteright} and {\textquoteleft}Peripheral vascular disease{\textquoteright}.",

keywords = "Bronchitis unspecified, Cough, Dyspnoea, Lung cancer, Machine learning, Upper respiratory infection",

author = "Abeer Alzubaidi and Jaspreet Kaur and Mufti Mahmud and Brown, \{David J.\} and Jun He and Graham Ball and Baldwin, \{David R.\} and Emma O{\textquoteright}Dowd and Hubbard, \{Richard B.\}",

note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 1st International Conference on Applied Intelligence and Informatics, AII 2021 ; Conference date: 30-07-2021 Through 31-07-2021",

year = "2021",

doi = "10.1007/978-3-030-82269-9\_4",

language = "English",

isbn = "9783030822682",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "43--59",

editor = "Mufti Mahmud and Kaiser, \{M. Shamim\} and Nikola Kasabov and Khan Iftekharuddin and Ning Zhong",

booktitle = "Applied Intelligence and Informatics - 1st International Conference, AII 2021, Proceedings",

address = "Germany",

}

Alzubaidi, A, Kaur, J, Mahmud, M, Brown, DJ, He, J, Ball, G, Baldwin, DR, O’Dowd, E & Hubbard, RB 2021, Selecting Lung Cancer Patients from UK Primary Care Data: A Longitudinal Study of Feature Trends. in M Mahmud, MS Kaiser, N Kasabov, K Iftekharuddin & N Zhong (eds), Applied Intelligence and Informatics - 1st International Conference, AII 2021, Proceedings. Communications in Computer and Information Science, vol. 1435, Springer Science and Business Media Deutschland GmbH, pp. 43-59, 1st International Conference on Applied Intelligence and Informatics, AII 2021, Virtual, Online, 30/07/21. https://doi.org/10.1007/978-3-030-82269-9_4

Selecting Lung Cancer Patients from UK Primary Care Data: A Longitudinal Study of Feature Trends. / Alzubaidi, Abeer; Kaur, Jaspreet; Mahmud, Mufti et al.
Applied Intelligence and Informatics - 1st International Conference, AII 2021, Proceedings. ed. / Mufti Mahmud; M. Shamim Kaiser; Nikola Kasabov; Khan Iftekharuddin; Ning Zhong. Springer Science and Business Media Deutschland GmbH, 2021. p. 43-59 (Communications in Computer and Information Science; Vol. 1435).

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

TY - GEN

T1 - Selecting Lung Cancer Patients from UK Primary Care Data

T2 - 1st International Conference on Applied Intelligence and Informatics, AII 2021

AU - Alzubaidi, Abeer

AU - Kaur, Jaspreet

AU - Mahmud, Mufti

AU - Brown, David J.

AU - He, Jun

AU - Ball, Graham

AU - Baldwin, David R.

AU - O’Dowd, Emma

AU - Hubbard, Richard B.

PY - 2021

Y1 - 2021

N2 - A high proportion of lung cancer cases are detected at a late cancer stage when they present with symptoms to general practitioners (GP). Early diagnosis is a challenge because many symptoms are also common in other diseases. Therefore, this study aims to assess UK primary care data of patients one, two and three years prior to lung cancer diagnosis to capture trends in clinical features of patients with the goal of early diagnosis and thus potentially curative treatment. This longitudinal study utilises data from the Clinical Practice Research Datalink (CPRD) with linked data from the National Cancer Registration and Analysis Service (NCRAS). A comprehensive list of Read codes is created to select features of interest to establish if a patient has experienced a certain medical condition or not. The comparison of the relative frequencies of the identified predictors associated with cases and controls reveals the importance of the following groups of features: ‘Cough Wheeze’ and ‘Bronchitis unspecified’, ‘Dyspnoea’ and ‘Upper Respiratory Infection’, which are frequent events for lung cancer cases, where a high proportion of cases were also identified using ‘Haemoptysis’ and ‘Peripheral vascular disease’.

AB - A high proportion of lung cancer cases are detected at a late cancer stage when they present with symptoms to general practitioners (GP). Early diagnosis is a challenge because many symptoms are also common in other diseases. Therefore, this study aims to assess UK primary care data of patients one, two and three years prior to lung cancer diagnosis to capture trends in clinical features of patients with the goal of early diagnosis and thus potentially curative treatment. This longitudinal study utilises data from the Clinical Practice Research Datalink (CPRD) with linked data from the National Cancer Registration and Analysis Service (NCRAS). A comprehensive list of Read codes is created to select features of interest to establish if a patient has experienced a certain medical condition or not. The comparison of the relative frequencies of the identified predictors associated with cases and controls reveals the importance of the following groups of features: ‘Cough Wheeze’ and ‘Bronchitis unspecified’, ‘Dyspnoea’ and ‘Upper Respiratory Infection’, which are frequent events for lung cancer cases, where a high proportion of cases were also identified using ‘Haemoptysis’ and ‘Peripheral vascular disease’.

KW - Bronchitis unspecified

KW - Cough

KW - Dyspnoea

KW - Lung cancer

KW - Machine learning

KW - Upper respiratory infection

UR - https://www.scopus.com/pages/publications/85113539458

U2 - 10.1007/978-3-030-82269-9_4

DO - 10.1007/978-3-030-82269-9_4

M3 - Conference contribution

AN - SCOPUS:85113539458

SN - 9783030822682

T3 - Communications in Computer and Information Science

SP - 43

EP - 59

BT - Applied Intelligence and Informatics - 1st International Conference, AII 2021, Proceedings

A2 - Mahmud, Mufti

A2 - Kaiser, M. Shamim

A2 - Kasabov, Nikola

A2 - Iftekharuddin, Khan

A2 - Zhong, Ning

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 30 July 2021 through 31 July 2021

ER -

Alzubaidi A, Kaur J, Mahmud M, Brown DJ, He J, Ball G et al. Selecting Lung Cancer Patients from UK Primary Care Data: A Longitudinal Study of Feature Trends. In Mahmud M, Kaiser MS, Kasabov N, Iftekharuddin K, Zhong N, editors, Applied Intelligence and Informatics - 1st International Conference, AII 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 43-59. (Communications in Computer and Information Science). doi: 10.1007/978-3-030-82269-9_4

Selecting Lung Cancer Patients from UK Primary Care Data: A Longitudinal Study of Feature Trends

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this