A practical computerized decision support system for predicting the severity of Alzheimer's disease of an individual

Research output: Contribution to journalArticle

Abstract

Computerized clinical decision support systems can help to provide objective, standardized, and timely dementia diagnosis. However, current computerized systems are mainly based on group analysis, discrete classification of disease stages, or expensive and not readily accessible biomarkers, while current clinical practice relies relatively heavily on cognitive and functional assessments (CFA). In this study, we developed a computational framework using a suite of machine learning tools for identifying key markers in predicting the severity of Alzheimer's disease (AD) from a large set of biological and clinical measures. Six machine learning approaches, namely Kernel Ridge Regression (KRR), Support Vector Regression, and k-Nearest Neighbor for regression and Support Vector Machine (SVM), Random Forest, and k-Nearest Neighbor for classification, were used for the development of predictive models. We demonstrated high predictive power of CFA. Predictive performance of models incorporating CFA was shown to consistently have higher accuracy than those based solely on biomarker modalities. We found that KRR and SVM were the best performing regression and classification methods respectively. The optimal SVM performance was observed for a set of four CFA test scores (FAQ, ADAS13, MoCA, MMSE) with multi-class classification accuracy of 83.0%, 95%CI = (72.1%, 93.8%) while the best performance of the KRR model was reported with combined CFA and MRI neuroimaging data, i.e., R2 = 0.874, 95%CI = (0.827, 0.922). Given the high predictive power of CFA and their widespread use in clinical practice, we designed a data-driven and self-adaptive computerized clinical decision support system (CDSS) prototype for evaluating the severity of AD of an individual on a continuous spectrum. The system implemented an automated computational approach for data pre-processing, modelling, and validation and used exclusively the scores of selected cognitive measures as data entries. Taken together, we have developed an objective and practical CDSS to aid AD diagnosis.
LanguageEnglish
Pages157-171
Number of pages15
JournalExpert Systems with Applications
Volume130
Early online date10 Apr 2019
DOIs
Publication statusPublished - 15 Sep 2019

Fingerprint

Functional assessment
Clinical Decision Support Systems
Decision support systems
Alzheimer Disease
Support vector machines
Biomarkers
Learning systems
Neuroimaging
Dementia
Magnetic Resonance Imaging
Data acquisition
Support Vector Machine
Processing

Keywords

  • Dementia
  • Alzheimer disease
  • Machine Learning
  • Decision support system
  • Computational modelling
  • Diagnosis prediction
  • Neuroscience
  • Diagnosis support
  • Machine learning
  • Alzheimer's disease
  • Cognitive impairment

Cite this

@article{05fa868ebc7244f4907f00a4dc648270,
title = "A practical computerized decision support system for predicting the severity of Alzheimer's disease of an individual",
abstract = "Computerized clinical decision support systems can help to provide objective, standardized, and timely dementia diagnosis. However, current computerized systems are mainly based on group analysis, discrete classification of disease stages, or expensive and not readily accessible biomarkers, while current clinical practice relies relatively heavily on cognitive and functional assessments (CFA). In this study, we developed a computational framework using a suite of machine learning tools for identifying key markers in predicting the severity of Alzheimer's disease (AD) from a large set of biological and clinical measures. Six machine learning approaches, namely Kernel Ridge Regression (KRR), Support Vector Regression, and k-Nearest Neighbor for regression and Support Vector Machine (SVM), Random Forest, and k-Nearest Neighbor for classification, were used for the development of predictive models. We demonstrated high predictive power of CFA. Predictive performance of models incorporating CFA was shown to consistently have higher accuracy than those based solely on biomarker modalities. We found that KRR and SVM were the best performing regression and classification methods respectively. The optimal SVM performance was observed for a set of four CFA test scores (FAQ, ADAS13, MoCA, MMSE) with multi-class classification accuracy of 83.0{\%}, 95{\%}CI = (72.1{\%}, 93.8{\%}) while the best performance of the KRR model was reported with combined CFA and MRI neuroimaging data, i.e., R2 = 0.874, 95{\%}CI = (0.827, 0.922). Given the high predictive power of CFA and their widespread use in clinical practice, we designed a data-driven and self-adaptive computerized clinical decision support system (CDSS) prototype for evaluating the severity of AD of an individual on a continuous spectrum. The system implemented an automated computational approach for data pre-processing, modelling, and validation and used exclusively the scores of selected cognitive measures as data entries. Taken together, we have developed an objective and practical CDSS to aid AD diagnosis.",
keywords = "Dementia, Alzheimer disease, Machine Learning, Decision support system, Computational modelling, Diagnosis prediction, Neuroscience, Diagnosis support, Machine learning, Alzheimer's disease, Cognitive impairment",
author = "Magda Bucholc and Xuemei Ding and Wang, {Haiying / HY} and Glass, {David H.} and H. Wang and Girijesh Prasad and Liam Maguire and Anthony Bjourson and Paula McClean and Stephen Todd and David Finn and KongFatt Wong-Lin",
year = "2019",
month = "9",
day = "15",
doi = "10.1016/j.eswa.2019.04.022",
language = "English",
volume = "130",
pages = "157--171",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier",

}

TY - JOUR

T1 - A practical computerized decision support system for predicting the severity of Alzheimer's disease of an individual

AU - Bucholc, Magda

AU - Ding, Xuemei

AU - Wang, Haiying / HY

AU - Glass, David H.

AU - Wang, H.

AU - Prasad, Girijesh

AU - Maguire, Liam

AU - Bjourson, Anthony

AU - McClean, Paula

AU - Todd, Stephen

AU - Finn, David

AU - Wong-Lin, KongFatt

PY - 2019/9/15

Y1 - 2019/9/15

N2 - Computerized clinical decision support systems can help to provide objective, standardized, and timely dementia diagnosis. However, current computerized systems are mainly based on group analysis, discrete classification of disease stages, or expensive and not readily accessible biomarkers, while current clinical practice relies relatively heavily on cognitive and functional assessments (CFA). In this study, we developed a computational framework using a suite of machine learning tools for identifying key markers in predicting the severity of Alzheimer's disease (AD) from a large set of biological and clinical measures. Six machine learning approaches, namely Kernel Ridge Regression (KRR), Support Vector Regression, and k-Nearest Neighbor for regression and Support Vector Machine (SVM), Random Forest, and k-Nearest Neighbor for classification, were used for the development of predictive models. We demonstrated high predictive power of CFA. Predictive performance of models incorporating CFA was shown to consistently have higher accuracy than those based solely on biomarker modalities. We found that KRR and SVM were the best performing regression and classification methods respectively. The optimal SVM performance was observed for a set of four CFA test scores (FAQ, ADAS13, MoCA, MMSE) with multi-class classification accuracy of 83.0%, 95%CI = (72.1%, 93.8%) while the best performance of the KRR model was reported with combined CFA and MRI neuroimaging data, i.e., R2 = 0.874, 95%CI = (0.827, 0.922). Given the high predictive power of CFA and their widespread use in clinical practice, we designed a data-driven and self-adaptive computerized clinical decision support system (CDSS) prototype for evaluating the severity of AD of an individual on a continuous spectrum. The system implemented an automated computational approach for data pre-processing, modelling, and validation and used exclusively the scores of selected cognitive measures as data entries. Taken together, we have developed an objective and practical CDSS to aid AD diagnosis.

AB - Computerized clinical decision support systems can help to provide objective, standardized, and timely dementia diagnosis. However, current computerized systems are mainly based on group analysis, discrete classification of disease stages, or expensive and not readily accessible biomarkers, while current clinical practice relies relatively heavily on cognitive and functional assessments (CFA). In this study, we developed a computational framework using a suite of machine learning tools for identifying key markers in predicting the severity of Alzheimer's disease (AD) from a large set of biological and clinical measures. Six machine learning approaches, namely Kernel Ridge Regression (KRR), Support Vector Regression, and k-Nearest Neighbor for regression and Support Vector Machine (SVM), Random Forest, and k-Nearest Neighbor for classification, were used for the development of predictive models. We demonstrated high predictive power of CFA. Predictive performance of models incorporating CFA was shown to consistently have higher accuracy than those based solely on biomarker modalities. We found that KRR and SVM were the best performing regression and classification methods respectively. The optimal SVM performance was observed for a set of four CFA test scores (FAQ, ADAS13, MoCA, MMSE) with multi-class classification accuracy of 83.0%, 95%CI = (72.1%, 93.8%) while the best performance of the KRR model was reported with combined CFA and MRI neuroimaging data, i.e., R2 = 0.874, 95%CI = (0.827, 0.922). Given the high predictive power of CFA and their widespread use in clinical practice, we designed a data-driven and self-adaptive computerized clinical decision support system (CDSS) prototype for evaluating the severity of AD of an individual on a continuous spectrum. The system implemented an automated computational approach for data pre-processing, modelling, and validation and used exclusively the scores of selected cognitive measures as data entries. Taken together, we have developed an objective and practical CDSS to aid AD diagnosis.

KW - Dementia

KW - Alzheimer disease

KW - Machine Learning

KW - Decision support system

KW - Computational modelling

KW - Diagnosis prediction

KW - Neuroscience

KW - Diagnosis support

KW - Machine learning

KW - Alzheimer's disease

KW - Cognitive impairment

UR - https://authors.elsevier.com/a/1YxH63PiGTBUPu

UR - http://www.scopus.com/inward/record.url?scp=85064524072&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2019.04.022

DO - 10.1016/j.eswa.2019.04.022

M3 - Article

VL - 130

SP - 157

EP - 171

JO - Expert Systems with Applications

T2 - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -