Early Prediction of Memory Disorders Using Primary Healthcare and Social Services Data

Heba Sourkatti; Tunc Asuroglu; Aino Alahäivälä; Anna-Maija Tolppanen; Jouni Ihalainen

Early Prediction of Memory Disorders Using Primary Healthcare and Social Services Data

Heba Sourkatti, Tunc Asuroglu (Corresponding author), Aino Alahäivälä, Anna-Maija Tolppanen, Jouni Ihalainen

BA6308 Health data analytics

University of Eastern Finland

Research output: Chapter in Book/Report/Conference proceeding › Conference abstract in proceedings › Scientific

Abstract

This study focuses on early identification of memory disorders among elderly individuals, utilizing data from social and healthcare services in Kuopio. A cohort of 26,000 citizens aged over 65 as of 2015 was utilized. Through a case-control study, individuals diagnosed with Alzheimer’s disease (AD) and controls were identified. ANOVA and Mutual Information (MI) methods identified significant features including International Classification of Primary Care (ICPC) and International Statistical Classification of Diseases and Related Health Problems (ICD-10) codes, onset age, number of patient visits, and types of services. Logistic regression and SHAP-guided gradient-boosting classifiers demonstrated promising predictive performance, suggesting potential for proactive interventions and targeted monitoring in individuals at risk of memory disorders

Original language	English
Title of host publication	Digital Health and Wireless Solutions
Subtitle of host publication	First Nordic Conference, NCDHWS 2024
Publisher	Springer
Pages	500-502
Number of pages	3
Volume	II
ISBN (Electronic)	978-3-031-59091-7
ISBN (Print)	978-3-031-59090-0
Publication status	Published - 5 May 2024
MoE publication type	Not Eligible
Event	1st Nordic Conference on Digital Health and Wireless Solutions, NCDHWS 2024 - Hotel Lasaretti, Oulu, Finland Duration: 7 May 2024 → 8 May 2024 https://nordic-digihealth.com/welcome/

Publication series

Series	Communications in Computer and Information Science
Volume	2084
ISSN	1865-0929

Conference

Conference	1st Nordic Conference on Digital Health and Wireless Solutions, NCDHWS 2024
Country/Territory	Finland
City	Oulu
Period	7/05/24 → 8/05/24
Internet address	https://nordic-digihealth.com/welcome/

Keywords

Machine learning (ML)
feature selection
Predictive modeling
memory disorder
Electronic Health Records

Access to Document

https://doi.org/10.1007/978-3-031-59091-7

Cite this

@inbook{cb59052a7c114cb1aa24f4c1eb2b4f83,

title = "Early Prediction of Memory Disorders Using Primary Healthcare and Social Services Data",

abstract = "This study focuses on early identification of memory disorders among elderly individuals, utilizing data from social and healthcare services in Kuopio. A cohort of 26,000 citizens aged over 65 as of 2015 was utilized. Through a case-control study, individuals diagnosed with Alzheimer{\textquoteright}s disease (AD) and controls were identified. ANOVA and Mutual Information (MI) methods identified significant features including International Classification of Primary Care (ICPC) and International Statistical Classification of Diseases and Related Health Problems (ICD-10) codes, onset age, number of patient visits, and types of services. Logistic regression and SHAP-guided gradient-boosting classifiers demonstrated promising predictive performance, suggesting potential for proactive interventions and targeted monitoring in individuals at risk of memory disorders",

keywords = "Machine learning (ML), feature selection, Predictive modeling, memory disorder, Electronic Health Records",

author = "Heba Sourkatti and Tunc Asuroglu and Aino Alah{\"a}iv{\"a}l{\"a} and Anna-Maija Tolppanen and Jouni Ihalainen",

year = "2024",

month = may,

day = "5",

language = "English",

isbn = "978-3-031-59090-0",

volume = "II",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "500--502",

booktitle = "Digital Health and Wireless Solutions",

address = "Germany",

note = "1st Nordic Conference on Digital Health and Wireless Solutions, NCDHWS 2024 ; Conference date: 07-05-2024 Through 08-05-2024",

url = "https://nordic-digihealth.com/welcome/",

}

Sourkatti, H , Asuroglu, T , Alahäivälä, A, Tolppanen, A-M & Ihalainen, J 2024, Early Prediction of Memory Disorders Using Primary Healthcare and Social Services Data. in Digital Health and Wireless Solutions: First Nordic Conference, NCDHWS 2024. vol. II, Springer, Communications in Computer and Information Science, vol. 2084, pp. 500-502, 1st Nordic Conference on Digital Health and Wireless Solutions, NCDHWS 2024, Oulu, Finland, 7/05/24. <https://doi.org/10.1007/978-3-031-59091-7>

Early Prediction of Memory Disorders Using Primary Healthcare and Social Services Data. / Sourkatti, Heba ; Asuroglu, Tunc (Corresponding author); Alahäivälä, Aino et al.
Digital Health and Wireless Solutions: First Nordic Conference, NCDHWS 2024. Vol. II Springer, 2024. p. 500-502 (Communications in Computer and Information Science, Vol. 2084).

Research output: Chapter in Book/Report/Conference proceeding › Conference abstract in proceedings › Scientific

TY - CHAP

T1 - Early Prediction of Memory Disorders Using Primary Healthcare and Social Services Data

AU - Sourkatti, Heba

AU - Asuroglu, Tunc

AU - Alahäivälä, Aino

AU - Tolppanen, Anna-Maija

AU - Ihalainen, Jouni

PY - 2024/5/5

Y1 - 2024/5/5

N2 - This study focuses on early identification of memory disorders among elderly individuals, utilizing data from social and healthcare services in Kuopio. A cohort of 26,000 citizens aged over 65 as of 2015 was utilized. Through a case-control study, individuals diagnosed with Alzheimer’s disease (AD) and controls were identified. ANOVA and Mutual Information (MI) methods identified significant features including International Classification of Primary Care (ICPC) and International Statistical Classification of Diseases and Related Health Problems (ICD-10) codes, onset age, number of patient visits, and types of services. Logistic regression and SHAP-guided gradient-boosting classifiers demonstrated promising predictive performance, suggesting potential for proactive interventions and targeted monitoring in individuals at risk of memory disorders

AB - This study focuses on early identification of memory disorders among elderly individuals, utilizing data from social and healthcare services in Kuopio. A cohort of 26,000 citizens aged over 65 as of 2015 was utilized. Through a case-control study, individuals diagnosed with Alzheimer’s disease (AD) and controls were identified. ANOVA and Mutual Information (MI) methods identified significant features including International Classification of Primary Care (ICPC) and International Statistical Classification of Diseases and Related Health Problems (ICD-10) codes, onset age, number of patient visits, and types of services. Logistic regression and SHAP-guided gradient-boosting classifiers demonstrated promising predictive performance, suggesting potential for proactive interventions and targeted monitoring in individuals at risk of memory disorders

KW - Machine learning (ML)

KW - feature selection

KW - Predictive modeling

KW - memory disorder

KW - Electronic Health Records

M3 - Conference abstract in proceedings

SN - 978-3-031-59090-0

VL - II

T3 - Communications in Computer and Information Science

SP - 500

EP - 502

BT - Digital Health and Wireless Solutions

PB - Springer

T2 - 1st Nordic Conference on Digital Health and Wireless Solutions, NCDHWS 2024

Y2 - 7 May 2024 through 8 May 2024

ER -

Early Prediction of Memory Disorders Using Primary Healthcare and Social Services Data

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this